<%BANNER%>

Examining Changes in Male and Female Intimate Partner Homicide over Time, 1990-2000

University of Florida Institutional Repository
Permanent Link: http://ufdc.ufl.edu/UFE0022561/00001

Material Information

Title: Examining Changes in Male and Female Intimate Partner Homicide over Time, 1990-2000
Physical Description: 1 online resource (129 p.)
Language: english
Creator: Reckdenwald, Amy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Criminology, Law and Society -- Dissertations, Academic -- UF
Genre: Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Research on intimate partner homicide has increased in the recent years. This is partially due to the dramatic decline witnessed over the last couple of decades in these types of homicides as well as the growth that has occurred in public awareness and policy responses toward domestic violence. Recent intimate partner homicide research has predominately focused around two perspectives to explain the relationship between intimate partner homicide and domestic violence resources ? the exposure reduction hypothesis and the backlash or retaliation hypothesis, with results that support both (Dugan, Nagin, and Rosenfeld, 1999; 2003). The exposure reduction hypothesis proposes that domestic violence resources that reduce the exposure or contact between intimate partners should decrease the probability of intimate partner homicide, while the backlash hypothesis suggests the opposite. That is, domestic violence interventions may have unintended consequences and increase the risk of intimate partner homicide if they threaten male dominance and control over their partners. The contradictory results from these studies have frustrated advocates and have made them question their efforts to make females safer. Societal remedies to lower intimate partner homicide seem to be addressing male-perpetrated intimate partner violence and homicide insufficiently. It is important to gain a better understanding of what factors are truly influencing gender-specific intimate partner homicide, during a time that is marked by significant transformations in domestic violence legislation and response toward domestic violence with the enactment of the Violence Against Women?s Act of 1994. This research helps to address the contradictory findings while controlling for a number of structural factors that have shown to be important in the homicide literature. The current study examines both the arguments of exposure reduction and backlash, in addition to economic deprivation and marginalization to explain the observed patterns in male- and female-perpetrated intimate partner homicide over time, something that has not been done in the literature to date. Measures of key concepts will be collected for 2 decennial points (1990 and 2000). Supplemental homicide files for this time period and census data will be utilized. Poisson regression models will be used to investigate which theoretical perspective is associated with male- and female-perpetrated intimate partner homicide in 1990 and 2000. Pooled cross-sectional fixed effect time series regression will be used to determine whether changes from 1990 to 2000 in key structural indicators influence trends in male- and female-perpetrated intimate partner homicide during this same time period. Overall, results suggest that key theoretical indicators do in fact influence the trends in female-perpetrated intimate partner homicide, but not male-perpetrated intimate partner homicide.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amy Reckdenwald.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Lanza-Kaduce, Lonn M.
Local: Co-adviser: Parker, Karen F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022561:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022561/00001

Material Information

Title: Examining Changes in Male and Female Intimate Partner Homicide over Time, 1990-2000
Physical Description: 1 online resource (129 p.)
Language: english
Creator: Reckdenwald, Amy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Criminology, Law and Society -- Dissertations, Academic -- UF
Genre: Criminology, Law, and Society thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Research on intimate partner homicide has increased in the recent years. This is partially due to the dramatic decline witnessed over the last couple of decades in these types of homicides as well as the growth that has occurred in public awareness and policy responses toward domestic violence. Recent intimate partner homicide research has predominately focused around two perspectives to explain the relationship between intimate partner homicide and domestic violence resources ? the exposure reduction hypothesis and the backlash or retaliation hypothesis, with results that support both (Dugan, Nagin, and Rosenfeld, 1999; 2003). The exposure reduction hypothesis proposes that domestic violence resources that reduce the exposure or contact between intimate partners should decrease the probability of intimate partner homicide, while the backlash hypothesis suggests the opposite. That is, domestic violence interventions may have unintended consequences and increase the risk of intimate partner homicide if they threaten male dominance and control over their partners. The contradictory results from these studies have frustrated advocates and have made them question their efforts to make females safer. Societal remedies to lower intimate partner homicide seem to be addressing male-perpetrated intimate partner violence and homicide insufficiently. It is important to gain a better understanding of what factors are truly influencing gender-specific intimate partner homicide, during a time that is marked by significant transformations in domestic violence legislation and response toward domestic violence with the enactment of the Violence Against Women?s Act of 1994. This research helps to address the contradictory findings while controlling for a number of structural factors that have shown to be important in the homicide literature. The current study examines both the arguments of exposure reduction and backlash, in addition to economic deprivation and marginalization to explain the observed patterns in male- and female-perpetrated intimate partner homicide over time, something that has not been done in the literature to date. Measures of key concepts will be collected for 2 decennial points (1990 and 2000). Supplemental homicide files for this time period and census data will be utilized. Poisson regression models will be used to investigate which theoretical perspective is associated with male- and female-perpetrated intimate partner homicide in 1990 and 2000. Pooled cross-sectional fixed effect time series regression will be used to determine whether changes from 1990 to 2000 in key structural indicators influence trends in male- and female-perpetrated intimate partner homicide during this same time period. Overall, results suggest that key theoretical indicators do in fact influence the trends in female-perpetrated intimate partner homicide, but not male-perpetrated intimate partner homicide.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amy Reckdenwald.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Lanza-Kaduce, Lonn M.
Local: Co-adviser: Parker, Karen F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022561:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101107_AAAADA INGEST_TIME 2010-11-08T02:58:09Z PACKAGE UFE0022561_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 8707 DFID F20101107_AACDJF ORIGIN DEPOSITOR PATH reckdenwald_a_Page_095thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
a5afc4ba47bea73c1d337061c512042a
SHA-1
1e369bc837def4cfcfb447cacd39b03686b9a9da
4651 F20101107_AACDIR reckdenwald_a_Page_087thm.jpg
ddce03b9326db7e64a4552d24c75f72a
1f8ca4a8ab943d1f65cfc48cd85156b28e74168a
116099 F20101107_AACCGD reckdenwald_a_Page_033.jpg
52605898f67496535a82cb0c43e62261
dc3321bc97194e59a80a91b6dc3a39c4aebca491
107075 F20101107_AACCFP reckdenwald_a_Page_018.jpg
9df86b2128f808af2157870256fecc56
3ac4062ebd25369da28cbd4f1cb74dc5ede5a56a
8663 F20101107_AACDJG reckdenwald_a_Page_096thm.jpg
a3e8389b91c50c90d6664128af4465a9
a8d35c6bd960896e945d6d67ae072186ca1bef16
113558 F20101107_AACCGE reckdenwald_a_Page_034.jpg
2e407f258e638fae13cb4e77920167db
31599fe7bfa71d8a8d2cbc8682ec719da121556c
36260 F20101107_AACDJH reckdenwald_a_Page_097.QC.jpg
bcce3c8a08e354aae827b14bab5f7200
dbc67d7c991607b5bdf58fa97f5c094ef36582da
4621 F20101107_AACDIS reckdenwald_a_Page_088thm.jpg
5a3f9b7aca5b12b0bf5688448f9e1cd3
90814b4a842528f3ee75269d9a71c6896921e3e7
109993 F20101107_AACCGF reckdenwald_a_Page_035.jpg
e6f64ee152f6a5ab31954711f3f30b5a
0a4934d5ae2d65c0c69d9010ebefaae95bfd7bef
108839 F20101107_AACCFQ reckdenwald_a_Page_019.jpg
c52efe110471e576094b3d92e5240cec
ab20365025b45f5e906cbfc177d7e730f745a961
8427 F20101107_AACDJI reckdenwald_a_Page_097thm.jpg
a9c94e0ce6746486595cebc9a567ffff
654d4e6471e4b505dd18d2df9c9dcffbd76c020c
6018 F20101107_AACDIT reckdenwald_a_Page_089.QC.jpg
0721d8a6e1b9235eadf58655abe1cecc
a086287cf077771fabfae787f0d0e50e0548e54a
103678 F20101107_AACCGG reckdenwald_a_Page_036.jpg
f27a6daf71daec23cde2402efecb3441
0c08ac957399684a40ea053c4b90fcd7a3cb3bf0
92532 F20101107_AACCFR reckdenwald_a_Page_020.jpg
a25914b1ae34c3a858edda3542629cd2
406ac0436f2cbc544acfdf49317464e98718177f
33688 F20101107_AACDJJ reckdenwald_a_Page_098.QC.jpg
fbc642363027135cc16791eeb652c73a
1c0da0df8ec0912815d997a43500a3cdfdaa81a9
2027 F20101107_AACDIU reckdenwald_a_Page_089thm.jpg
fdae6de682f13dfd6d99f61f9f42b826
43e7387817fcb2256735d75f190cae63a4dfe209
114925 F20101107_AACCGH reckdenwald_a_Page_037.jpg
89a4536b877f4f77f859c07381393280
0ee9b2faed1eb6a39fab91a72016d23a1864df7b
105617 F20101107_AACCFS reckdenwald_a_Page_021.jpg
3d17bd400b9211692fcf71ed28411c54
9c924d4d12815feaad6f0f116d0c67f25940ed1a
8349 F20101107_AACDJK reckdenwald_a_Page_098thm.jpg
8b50e242eb9eb67c30cff8acd5acab66
74955e01ac40fa6fd47c45de8c30111ad7d3d632
33864 F20101107_AACDIV reckdenwald_a_Page_090.QC.jpg
1148324e488926ed81cafaa8d8763215
8b5d66a5d85d72c0da37b10692d74cc273b6b2bb
110081 F20101107_AACCGI reckdenwald_a_Page_038.jpg
2f6861eda52cef36e1f44ad2d46c0b83
6d2a44f967576f885173d4b41b27ac406a63909f
44952 F20101107_AACCFT reckdenwald_a_Page_023.jpg
5211759b01038be5ca6515643ee5024c
44ae7bbb9648d983df1a29885444675107e2419f
35407 F20101107_AACDJL reckdenwald_a_Page_099.QC.jpg
2c93ae758de07b21ee2284ce7d796e55
515497c613aef19a2a7fc432c752e065b95066d6
8498 F20101107_AACDIW reckdenwald_a_Page_090thm.jpg
8e9364be8406279b9b9d7c0e7b613667
c26a576989513a44d58ecbcd4595b9434b498cee
111673 F20101107_AACCGJ reckdenwald_a_Page_039.jpg
58acc3ff63bf7d7d82448dd167f698d7
07fdb1988c24bbf4ccdc9d09dc4df5efc7195804
20751 F20101107_AACCFU reckdenwald_a_Page_024.jpg
916fc11114d73613614ee0a43d25c6dd
3a87b36aff1924fc846d24c190142deb355e35c3
8831 F20101107_AACDKA reckdenwald_a_Page_106thm.jpg
064ac355d0f219ee4a5c45ae13814143
71fb296e277aea64e0b7c23f49e585cffe38ca26
8770 F20101107_AACDJM reckdenwald_a_Page_099thm.jpg
93d9c638aea53ece0c6c8c1bc0cd615b
380035a83d7317e1db924c3e6365f34fb02d82a8
35708 F20101107_AACDIX reckdenwald_a_Page_091.QC.jpg
89df1abdea735076a99be2aecd97fcc3
cfe764421d78176f33711e83825751275fac2074
99735 F20101107_AACCGK reckdenwald_a_Page_040.jpg
ed6a0d1b10e4c830c40badeb310d9df6
94082c6b9bd48f3e74ae92aeb18d47dc9d753a13
31563 F20101107_AACCFV reckdenwald_a_Page_025.jpg
f6f86204468622c5ed6dd9e7d90e5c81
e7cd41d39fe0148f6bcd1fd7a35c25cf36e0787d
34181 F20101107_AACDKB reckdenwald_a_Page_107.QC.jpg
95b8fb6313419f7ffe73f536d09f3070
3beed9133fdc9f731d7b4fb6f9fe8c20a5b52576
36630 F20101107_AACDJN reckdenwald_a_Page_100.QC.jpg
054129343156e1dd48702393c15efff7
2e4c59065be07e31812066de95fbe8de28a230df
8696 F20101107_AACDIY reckdenwald_a_Page_091thm.jpg
f7f30a29e9d951789808e9809843c09b
e8c20e396a8e1e3ad216512096486922f0bdf13b
108089 F20101107_AACCGL reckdenwald_a_Page_041.jpg
491c86aebf8b3488b926655daa5e4da0
40606065c8725ab3b5050313064a2c198426294b
25757 F20101107_AACCFW reckdenwald_a_Page_026.jpg
29af7f18f53709590a3754860ace0b34
25254de4b578218284c9391ed0cab492e2e7bd59
8575 F20101107_AACDKC reckdenwald_a_Page_107thm.jpg
f89df42079aa6ae85ef90351e9c078a7
12f72d779d4b098edb11219ff73644cd19a07fd0
9035 F20101107_AACDJO reckdenwald_a_Page_100thm.jpg
b6c265b218d7261a977021bf0f12b882
ed098801d2c99a6474390db09562ce8c48f0c799
34964 F20101107_AACDIZ reckdenwald_a_Page_092.QC.jpg
4845596e7e7303ddd0f82354af12c1c7
28744e2ada35594770f218cb20911a916ba9e1f6
109023 F20101107_AACCGM reckdenwald_a_Page_042.jpg
6ad6c38b5673569d1464ab0a2af7216f
500d6f0cc48a5da4c9d35262faf74c2dc198d8b0
25065 F20101107_AACCFX reckdenwald_a_Page_027.jpg
d5c906b4580345e56daebf7d27efa6b4
a47f8727e8899b85e014d344db06b038bbab24fb
102279 F20101107_AACCHA reckdenwald_a_Page_058.jpg
f7aa452549a00449f4d4ab29ff3008da
5277b310309b7a7ec6e06e2fcce1c7e8b00aea03
19240 F20101107_AACDKD reckdenwald_a_Page_108.QC.jpg
0eb6679f59e8275066068c3d79021ff3
1385cf16da1269f4e5f754e2c6d4c558c3a19550
34745 F20101107_AACDJP reckdenwald_a_Page_101.QC.jpg
f6442f423d1770c444572c91b21928cf
e911ffd61caa186214cefdb1021fe3fad828bb4b
112831 F20101107_AACCGN reckdenwald_a_Page_043.jpg
13da885f82f960e81f04ad2981ce8237
72a88bc7e5773e34613ba27baa7f2e8fb381b6c6
109678 F20101107_AACCFY reckdenwald_a_Page_028.jpg
e0876c99dc0a36a1c88c04aeef996871
50ac6dfb56141b78d6d655334c5d0c8a419314d2
100703 F20101107_AACCHB reckdenwald_a_Page_059.jpg
9361cd54f9d53ea66eddfd2e9048da61
e306df3d2bd72c6f81d921d4078d0e467eb23b0d
4704 F20101107_AACDKE reckdenwald_a_Page_108thm.jpg
9f2a1b4eb61e2fa88fe3e2f87df269a7
521ad72a2781d223934f2054308bec8e4f6848b8
8710 F20101107_AACDJQ reckdenwald_a_Page_101thm.jpg
5585817138b4c1b773541fe2404d80d8
9516370a4cff850bc25f218856a215fe1abd50df
104589 F20101107_AACCGO reckdenwald_a_Page_045.jpg
ceb83516ae6f184efc1ae88fad5fed07
705e80f838d846353020882394baef67a4fef99a
104337 F20101107_AACCFZ reckdenwald_a_Page_029.jpg
d1681e43d687ccc74a38742345952dca
0bf3177123a9910f46e6eae63d6263acc6115613
107416 F20101107_AACCHC reckdenwald_a_Page_060.jpg
f1d198316ba5d827e03ff14e67a02da7
6c08f9f990905259781ec7dc5e4522ec17dd24fe
3581 F20101107_AACDKF reckdenwald_a_Page_109.QC.jpg
586ce3e34f0e6b0e18566d14a65f94d7
5eaaa48a84caf3a66738aa19994db8bb8d14c8b0
7470 F20101107_AACDJR reckdenwald_a_Page_102.QC.jpg
80f2fa1ecd52d4bf23f76c83c0a19ec6
fe864eb823f8664f014d9eea7793ab3166460bbe
104436 F20101107_AACCGP reckdenwald_a_Page_046.jpg
6250a36f07fa2afc517f4969731ef4e8
88fe7a8424da1d872ca805236362236f3a475410
110314 F20101107_AACCHD reckdenwald_a_Page_061.jpg
f1106d02ac5bad25a2befefd7d8cd167
a2a71331627bfcd4e544deb70f135be8c5d6ec12
936 F20101107_AACDKG reckdenwald_a_Page_109thm.jpg
504cc0d19cb4c13bcd9371b16a82eebc
34adfb2487da0cd5dc6effda45db795a7eae84b3
2110 F20101107_AACDJS reckdenwald_a_Page_102thm.jpg
a306502126bfaf8884f5cdda193613d1
4c8b57c4a83794eccd1edbff3beeaaf390f49ae1
109515 F20101107_AACCHE reckdenwald_a_Page_062.jpg
348aa5aa3d1c2634ccf7a88e521c0aab
a2f5982233527c4ff518d0d519c26f21c4c12c69
77648 F20101107_AACCGQ reckdenwald_a_Page_047.jpg
eb7799e72a013f5dde60f8620aa5b10e
fc66574e3a4af0171abe66153ae430740a07119b
17057 F20101107_AACDKH reckdenwald_a_Page_110.QC.jpg
39a8a033f7a2277cc6ef551eff616b6e
5fbb9aa64fd114dc668ab0cac46502ead4882932
116918 F20101107_AACCHF reckdenwald_a_Page_063.jpg
5f3e3da7a445e336b5c47170baa123cc
b51ff852730f0d10ed624fc4884f2a0d922c1f67
4043 F20101107_AACDKI reckdenwald_a_Page_110thm.jpg
cf7ade7d77a516e5bce74d5376bac762
18208ba7997d2d5059212737dd1ab8ff6b79cc55
28040 F20101107_AACDJT reckdenwald_a_Page_103.QC.jpg
4b63a09caff5bfe101a8458a2ae651e1
f0d754344aabbf38879f32fb3baa9768b9d1f494
103695 F20101107_AACCHG reckdenwald_a_Page_064.jpg
39373dd5871e6800b83eb8e0e2728cb3
492c59c42b433f0410d7f9ce6fca50a3fb59f390
35733 F20101107_AACCGR reckdenwald_a_Page_048.jpg
6a0987458bc87e1f0b0dcafdea28c451
e2f0fda7ca9b555fc4ae7ed2049783a7ab1dac14
3571 F20101107_AACDKJ reckdenwald_a_Page_111.QC.jpg
cb954d5dd04a5aa6fc524aac1ea70ec5
f22fd472c8f575d9af6f6b14faaf45020b3a9d49
7357 F20101107_AACDJU reckdenwald_a_Page_103thm.jpg
ab96bc304b532a81b15cc32a4d961722
cca5ebf9d28b677f29275b3493535dc1945730bb
102174 F20101107_AACCHH reckdenwald_a_Page_065.jpg
8c2861e8c6b6a28685f287cdb0f2a0ac
8c1cf3028f9bd90a697b3c8a92cf4e23e938b101
100809 F20101107_AACCGS reckdenwald_a_Page_049.jpg
c34167856038e350df8401be1039abec
0e52b9ef62d424e6f726ee914b243a5f6116b477
948 F20101107_AACDKK reckdenwald_a_Page_111thm.jpg
7ab1d961e2d546896d75655767e33e5e
e6a5886bbb8291ef34d165548bb96fc3ca4e89a8
34347 F20101107_AACDJV reckdenwald_a_Page_104.QC.jpg
d9545e8e9a6f9fb99fbc4861fca97ec1
52dcb182ac3303d1614a168d705ad94a92f0b94f
102839 F20101107_AACCHI reckdenwald_a_Page_066.jpg
b3b284e857559b143b68da789f324ed6
5a08c0728b7990ed14ceb981d96a0948ad76cb22
101130 F20101107_AACCGT reckdenwald_a_Page_050.jpg
7be22951577ac33667e1c0016318402f
167cbbd746f830b84ee66a5a88f1b6e91d41c5b9
17571 F20101107_AACDKL reckdenwald_a_Page_112.QC.jpg
49c442bd61672c485f983b60780730b4
87fd8f34ae3a958979520074ee72cd6509e2092a
8792 F20101107_AACDJW reckdenwald_a_Page_104thm.jpg
4764c757446205ed5a8a4d6f6ee00719
68d0302a0b5975e437b311eb11a6896a5229647e
106375 F20101107_AACCHJ reckdenwald_a_Page_067.jpg
8a73b06058740e21752d56636d4398dd
30a29e589b1caf80f24643df0299e021aa57d6c1
97447 F20101107_AACCGU reckdenwald_a_Page_051.jpg
d172da2c241107d2401e8ef648637b1a
be45909255d8141e744da124cbb1d5bedd4307a5
37271 F20101107_AACDLA reckdenwald_a_Page_120.QC.jpg
51a770f23e161db521c346794b078b43
26108c31af7e58d19b7f849674824905354b6813
4397 F20101107_AACDKM reckdenwald_a_Page_112thm.jpg
3f3e62826d0437857fa4e6a4cdb34083
bccd608759c14fc48a54a9360e1a74e4cfa18fbf
36094 F20101107_AACDJX reckdenwald_a_Page_105.QC.jpg
2c25aaba00df65b9138f5189b7b663be
84f93f9e4fd57928267d177ae421f72ae7922370
83861 F20101107_AACCHK reckdenwald_a_Page_068.jpg
780beb9d884825504157bba593c2e639
e2f29efe6adf5304621fdffbc79eefd31fb31ca2
102276 F20101107_AACCGV reckdenwald_a_Page_052.jpg
97fab1d4fd161f4c04b5cdf4987c2e4e
3f11d4dffbbacbfb35560c7c641532ade1b9b8cb
9001 F20101107_AACDLB reckdenwald_a_Page_120thm.jpg
dd158b8fc1847e37a581608b6b53d61d
f157d819dc9f14042034563dd8528ae04f18c35a
3704 F20101107_AACDKN reckdenwald_a_Page_113.QC.jpg
4668b4711ef217fcb298e59588d10dba
3ba4f87f55ee36f3f415d4a94c255e8344dc4f34
9087 F20101107_AACDJY reckdenwald_a_Page_105thm.jpg
62bb5b54f5651a8c2dbeeeac5b66bea8
1b5d4af081713e938cb5e30c63ba845ae4fccab8
53201 F20101107_AACCHL reckdenwald_a_Page_069.jpg
2d544026b2fb97b135bd998f7d26e95f
02053794d51b4610bf006fd00356cb70d1668e3c
94138 F20101107_AACCGW reckdenwald_a_Page_053.jpg
1bc609b43aace57298c18636c5436dec
e6e8dc05c24163c6a7ea1a2b4eb47fb8f53ef319
36805 F20101107_AACDLC reckdenwald_a_Page_121.QC.jpg
cffb766f6f6911e1b1b8598f04515a83
8b176b326af6e4528b91c6e326aee5fd088c3e6b
992 F20101107_AACDKO reckdenwald_a_Page_113thm.jpg
3f9ca3211d8b1c985409d1f1e014c2bd
b6c3d0b0f1d98a3a74d21cb0eea02f79c5673fa4
35378 F20101107_AACDJZ reckdenwald_a_Page_106.QC.jpg
3a61f765f9e14e8cb3d86259b5e2a905
d05568760b6560d0f01b53a7333fb25acb4be9f3
53761 F20101107_AACCIA reckdenwald_a_Page_084.jpg
5104faa687dcd0dace73c3d431c4eb4e
edfc531b1dd30bde7e3c80b3d45fa1b33f7e65e6
59221 F20101107_AACCHM reckdenwald_a_Page_070.jpg
0d5ab0529bdcf6669b6b8866ca4a645f
62e9047e9a1d15d99291f32b5c00930a6a3966b5
100764 F20101107_AACCGX reckdenwald_a_Page_054.jpg
eca5fd068d8a6bbb10c667f28cd98da4
9a7b872bf02fb390457716b5ee34733ee8fb36fa
8790 F20101107_AACDLD reckdenwald_a_Page_121thm.jpg
9d730aead33dc024601e4222666e5669
cfd7b064b3fecda9c7d0168165299872927f2056
16054 F20101107_AACDKP reckdenwald_a_Page_114.QC.jpg
95587d700b0de6c96912110ba3963152
f2746bbd003a58ac713ffa7e44a81336e4fa89dc
53559 F20101107_AACCIB reckdenwald_a_Page_085.jpg
061227ab9f5dfe20e470d78849d07e82
6afdb7d1d88710ad6f52dc7c83af50d6d88581fa
56115 F20101107_AACCHN reckdenwald_a_Page_071.jpg
38fd05c23625b6aff0dfbb42e65b7f25
420f244010b19d02b55d8e4fae762327d897ba8d
103194 F20101107_AACCGY reckdenwald_a_Page_056.jpg
ea553c5cf1ed5f83c7290ffeca8b8b75
239720dee0ed829ad983ed669715ec05c56f1ca0
35943 F20101107_AACDLE reckdenwald_a_Page_122.QC.jpg
a6d54dc60cce7ed5e73ac6ce0e8a6cd4
e775b0954502c3e923cba99a7ccfa64b541ffd39
4134 F20101107_AACDKQ reckdenwald_a_Page_114thm.jpg
7fc6ed3bc82295141e32d3fca6cd78d6
e67415ee5e0370918e6371abea53f52dadad21a8
30493 F20101107_AACCIC reckdenwald_a_Page_086.jpg
2536e9bc6680106a9e7073770e386216
91a0b256a1cbd803a46d2e4d2dedc90cef650704
23046 F20101107_AACCHO reckdenwald_a_Page_072.jpg
1f86ae0ed2c417be884cf31d1bc044d9
be1a09db28f88a8dbd3c5164cdee911cf5362340
98062 F20101107_AACCGZ reckdenwald_a_Page_057.jpg
59b347d959d2204a44828ef67f5dabdb
7301730c17f4f1479d39cd50589665912c5d20a1
8964 F20101107_AACDLF reckdenwald_a_Page_122thm.jpg
3bae25baf58dff855ee27a3a87c57a8a
787749ffa8a1123d7644b2bf4855cdc5cb10f842
3689 F20101107_AACDKR reckdenwald_a_Page_115.QC.jpg
f2094099699dc9dc69b03c6fd476075c
92a6d2ec025f207f702b66aab776eec64002a0a7
63553 F20101107_AACCID reckdenwald_a_Page_087.jpg
51c4201708808d607ed061ba709c620e
7d2e28a1b4efcfbaf39138cc0ad2394487fab909
23327 F20101107_AACCHP reckdenwald_a_Page_073.jpg
223ff529a75e06213a21ae12c6b55630
3f332afd46252bcf1ac0d3a29c1fbe8da0e4e1a5
36963 F20101107_AACDLG reckdenwald_a_Page_123.QC.jpg
a49f612bfcfc35cd6ea69023256ff5cd
02c96567d1d8144c6ff9733e5c1f60870fcb13f3
995 F20101107_AACDKS reckdenwald_a_Page_115thm.jpg
b27d1471dcc8dfeafd55852491787214
5a23fa0bab0b2b137a7c74da1915226054336959
62153 F20101107_AACCIE reckdenwald_a_Page_088.jpg
1f159c3dded9163f09bbd4b90174bf74
4d8d06d968e2caf551235fa0994252e491ed4026
98924 F20101107_AACCHQ reckdenwald_a_Page_074.jpg
e64eb1794980ba1046f816ac7d015695
6873ee36fb46f1a61b72937da12a4ac6117e5817
9004 F20101107_AACDLH reckdenwald_a_Page_123thm.jpg
b6dd3f970d4264984001e48d93fcccea
f0e352a08e743ea2a8ff87d62e1deb6bc3a1065f
16057 F20101107_AACDKT reckdenwald_a_Page_116.QC.jpg
4064548778a18af42abedb3d63ce9640
777fde62caef8b95ac526c806bc310456c2e9a41
19025 F20101107_AACCIF reckdenwald_a_Page_089.jpg
cfd0a4636944687ff1199339443076c6
87f275b7d2bebfbe70ed478a0c2afa19d2a522d5
98147 F20101107_AACCHR reckdenwald_a_Page_075.jpg
1a35414a555835208c41b1604ffca69c
84f59efde2b45c2c2fbd72c3e3bddad4d6fb32b6
38422 F20101107_AACDLI reckdenwald_a_Page_124.QC.jpg
7c50fcb326945555e06c16dc921c7b87
1530e925cd459c0ee8168abfe6415a6dbf25a72b
107517 F20101107_AACCIG reckdenwald_a_Page_090.jpg
9a6342c6a84407ca6258b259be442d25
9defc3507e1a94464e614b2214540e9fa4553865
9223 F20101107_AACDLJ reckdenwald_a_Page_124thm.jpg
67f41ad21851bd32452d064d12a67004
57bdb47c47a01eee9d38616d894d7cc0544de4fe
4157 F20101107_AACDKU reckdenwald_a_Page_116thm.jpg
f243a697f71f4e3f1ace32223852de73
19c5827a8747185c19e0b2c13bfaec69b1ef2bc1
108261 F20101107_AACCIH reckdenwald_a_Page_091.jpg
bcf840ad82b89421692949ddacc15e10
c268fbb7996edd1703fe85d9470d5e5f34a461b4
110445 F20101107_AACCHS reckdenwald_a_Page_076.jpg
60c55ca93a8584e0ed818fdf01d49586
33ec24f1f328d5adf4cf0e400fa4bf4e8329dd48
37090 F20101107_AACDLK reckdenwald_a_Page_125.QC.jpg
f65b490f267c23e0a53cfaf8aab13938
0bccd691f7f78e35fbbdfe19422d969c78791950
17453 F20101107_AACDKV reckdenwald_a_Page_117.QC.jpg
7a5f8e7f61d2771a1c04fb2b8de786da
f8193ab11c52f16a82db4735712763dfa909aa61
107600 F20101107_AACCII reckdenwald_a_Page_092.jpg
7fd7fdbbc94c8ecdfa4653e10ca03522
e8e5b3aa246591701dbceac2fe49d0e6e53fa7e4
113196 F20101107_AACCHT reckdenwald_a_Page_077.jpg
90214c42924bd8686d77fe269f22afcc
a3d46847f97d3414bf7b5aad6ce17a75c56191aa
9345 F20101107_AACDLL reckdenwald_a_Page_125thm.jpg
525f2d589e48d85a00101d0d2d654371
fd9e1757d73567379936b950115344cabdedfa7d
4879 F20101107_AACDKW reckdenwald_a_Page_117thm.jpg
bb99f9ecdda0066967106dcd98fca1b5
e7b9b980a6db6dd416f5d3dffb9cdf0014ffa07e
109197 F20101107_AACCIJ reckdenwald_a_Page_093.jpg
ac55abb0069d178fa2706662b4d70bb0
fc92cac823bf6b3914e5b8fe9a2c51bd5fed3845
113352 F20101107_AACCHU reckdenwald_a_Page_078.jpg
1fea38b9d9e97380ab0b518bc6f37914
fd5acc64aeb1a6542dbd2f9254c2f3845cca421e
33761 F20101107_AACDLM reckdenwald_a_Page_126.QC.jpg
6537e38a6bcc9f962218887b576162ce
dbfb78083a30a4de3c8ce185b11138bc8dc5a6f2
3606 F20101107_AACDKX reckdenwald_a_Page_118thm.jpg
7e3a7bad0362909ae2179b8cbd3a4db8
3dd8e2a36ee2a0ea488be4accf505e5cbcc9d0dc
103484 F20101107_AACCIK reckdenwald_a_Page_094.jpg
b06f3989d93fc058fefe0953a9f280f5
434361d42571ffe5af95d1f833103a39bda2584b
106050 F20101107_AACCHV reckdenwald_a_Page_079.jpg
ace28980f37065a34668c638da8db199
b4321e6bd664c47fb2fe793c7756ae5e29728b88
8470 F20101107_AACDLN reckdenwald_a_Page_126thm.jpg
736b72b104d2db8d27cd7daf1d72c2a6
81873c7e2d543f9c6d9b06ca8824436cf36e31f8
33933 F20101107_AACDKY reckdenwald_a_Page_119.QC.jpg
c4d9f8f69ca70aeb0a3135730f590408
44946e9c25beece8d32336aa350ae117bd6b9cbd
108111 F20101107_AACCIL reckdenwald_a_Page_095.jpg
731ed5f8be450ded85932d2f2a817f4c
a3c94e84b09a78d818ebdf7f04d427c3070b7896
109134 F20101107_AACCHW reckdenwald_a_Page_080.jpg
70f162802d4eff844a713369e586d677
ee0f60e0f48636ddfebb8bdc8b7abbf099f8271c
37378 F20101107_AACDLO reckdenwald_a_Page_127.QC.jpg
5634feb30182eb91acaa94bbf4bbc65f
78b8901d07141e8665283b913a5647fb27b4cdd9
8626 F20101107_AACDKZ reckdenwald_a_Page_119thm.jpg
dcbe7f6d11c40fd4cd4d4978ff753a12
99a69859563d7eb91b5341957d2b8c08c3526b36
109665 F20101107_AACCIM reckdenwald_a_Page_097.jpg
62e0e850c7f1b1aa8dda68505498f195
634e5f7f74abe3658808dfc23ce589542fece0c1
110747 F20101107_AACCHX reckdenwald_a_Page_081.jpg
6fd50424a2b0f1d2bc54de7f25f785cc
90f7cb82a2e00c88ef01023a8023166dc6685c01
9434 F20101107_AACCJA reckdenwald_a_Page_111.jpg
9a723561c69c76447633cc343368d103
f5b4c3722d12bb8d9002f1f120767c3212e85ea5
9205 F20101107_AACDLP reckdenwald_a_Page_127thm.jpg
777fe1110823b5b7411c9f4b83179bea
27a7da16dd3a0e35977c5d3b8a28657db8591fa6
102821 F20101107_AACCIN reckdenwald_a_Page_098.jpg
e356d13d192047308bc91c53fbc85c13
6c5f4eebe6e0300b5ee296b4fd643933e95a75bb
110169 F20101107_AACCHY reckdenwald_a_Page_082.jpg
937edadd1c48518ad025029b26293741
67d96226d9034bd912b1bfd520292d5a70a3cc57
10285 F20101107_AACCJB reckdenwald_a_Page_113.jpg
3c208a88eb888a857b2be4bad0eab6ff
c60b174ee45cb4357ec524a4c6727261e126d2cd
25935 F20101107_AACDLQ reckdenwald_a_Page_128.QC.jpg
5566d8f5195343540f9eb4570cc22666
be0fc01e2e66a5ad681a07edf85e69b1f524dd47
108050 F20101107_AACCIO reckdenwald_a_Page_099.jpg
45d0031c3de8877e9dad03167abae8b0
a03c60fe65117c94d74b90d786d91da150b6d628
8353 F20101107_AACCHZ reckdenwald_a_Page_083.jpg
2976bfcd2a7e287f9f2ccc657408234e
4de06231f34cfeb7bfcf3cb687f3db0a09d3f0cf
56200 F20101107_AACCJC reckdenwald_a_Page_114.jpg
512e7df83367f390909dfbf450f49aeb
c64460e18d8e08294a7104844fb6ede66b610832
6374 F20101107_AACDLR reckdenwald_a_Page_128thm.jpg
a04070340c143a9eed2208985f99d08e
28835ade99f11e2382aa522b561a0edfeff324e2
110122 F20101107_AACCIP reckdenwald_a_Page_100.jpg
bd171ddd8fd1dffe76209b0882bc97a0
fa1e422f2071f9ed1fbef7d67a89bb2effe96cc0
10271 F20101107_AACCJD reckdenwald_a_Page_115.jpg
50bb169d884f5f2fa00423247bd2b7e7
93a155983adbd4ac0804e44a52a2269555135ba9
14687 F20101107_AACDLS reckdenwald_a_Page_129.QC.jpg
84041f17b09c4f63585d989c6ec2534a
eecd034972e049627659a52692987a60f4c52eef
108667 F20101107_AACCIQ reckdenwald_a_Page_101.jpg
fce3d9d70b520653fc01fa8ff1f20385
2df09b298ba295a3db767f1d9dd2574b8bfee4e3
57101 F20101107_AACCJE reckdenwald_a_Page_116.jpg
fa9808d46ba8d69de4e2a7e1d2dbb08b
78b65342fd261dafaf8ad45f6af4bde1d0d7a396
3885 F20101107_AACDLT reckdenwald_a_Page_129thm.jpg
28712967e2f7fbb05fa8accfd69b456b
a426207c7fdc4988d3a9869a3767302bf8fb8576
23019 F20101107_AACCIR reckdenwald_a_Page_102.jpg
d99e9e7f879d20eb021dbba67adfb8c4
13355b653782807b3fda4b79428dba9882ee8197
50719 F20101107_AACCJF reckdenwald_a_Page_117.jpg
3179477e83a1d740617a5c9e7eb6c007
99282334a27ed5ad31bb87e8e7c4330c837ceaaa
152832 F20101107_AACDLU UFE0022561_00001.mets FULL
031e979eede52fec7b80928497ebcbbc
057ccc7f847958a3738d380563f2511e0e8de8d0
85490 F20101107_AACCIS reckdenwald_a_Page_103.jpg
618e6779b35c3c312004ecd834a2f29f
9791bbfe7caa7da972e0286bd9fe6d0deeb388ba
40587 F20101107_AACCJG reckdenwald_a_Page_118.jpg
3ceae79cf03a0a89fe8dfb10e407ed27
17f2ad9aa92736efdb84ace063fd98aeca6a361a
115781 F20101107_AACCJH reckdenwald_a_Page_119.jpg
e5e4059dce1158f40931b57c7cd87601
fcd1b7eccac9fde2b69a824d5fcf20234d377c0b
106630 F20101107_AACCIT reckdenwald_a_Page_104.jpg
a349e75061d903838c3c0e94cb383958
c685cf81e14573151fdfd0cd40645f2f969211d8
123802 F20101107_AACCJI reckdenwald_a_Page_121.jpg
ac7879e93411563ecaf6149ccb8e924f
b51525f4a9645c434add6503dce3f6d3e9a0d37a
110270 F20101107_AACCIU reckdenwald_a_Page_105.jpg
f3cf6cb6608a20f28e27b2068146db5e
cbef9259b12a345c7ad847cf11c817e626091f0f
127057 F20101107_AACCJJ reckdenwald_a_Page_122.jpg
f949f41ea659f519424f8ea5ea46b61b
c78587544c68142831e436c222d24f78ca04de0e
107837 F20101107_AACCIV reckdenwald_a_Page_106.jpg
9e893688b20952f2df3511fd6888089d
afdf881f68f3fb136eb2377f4c3837de533c262c
129500 F20101107_AACCJK reckdenwald_a_Page_123.jpg
7e2d5facb94eccb39afc23940127e670
d825ef55af428a5fd5cc9d580340ce33c4dd28cf
106324 F20101107_AACCIW reckdenwald_a_Page_107.jpg
50d7f5f38e46d2f01fb3ed5946a399a0
adce05b11171c43d62a9fb1084fe8d120e39468b
139186 F20101107_AACCJL reckdenwald_a_Page_124.jpg
3cd2b4d120bad7b9f4d0e2361f5c154a
319df05a9ccbbbd35711b40adecf53b8df61e8b4
57964 F20101107_AACCIX reckdenwald_a_Page_108.jpg
93d48d1c50140a315fa09ddcc24ccfbb
3bb5c899e457659b6e5f0e6eb619694e950c6be1
1051959 F20101107_AACCKA reckdenwald_a_Page_011.jp2
7e3cec69d1d17763ace0d0dd7fafe07a
ab9699826b13f01d246555149ac46e306b3069d3
131099 F20101107_AACCJM reckdenwald_a_Page_125.jpg
15afea5ce0a5ead466ae9af8c104d8d1
e6b227adbb19fc10238868af225aa589c25fa96c
9300 F20101107_AACCIY reckdenwald_a_Page_109.jpg
0299d9adac4115156b911a175be54bd6
9bd558f0bbae4da30dcb2d051afab14b3b77d3d2
1051981 F20101107_AACCKB reckdenwald_a_Page_012.jp2
cf14b743b759231982ba06ee24785905
9c4b2e3d0e098ab08867a83fd1c1d89a9c05d951
111521 F20101107_AACCJN reckdenwald_a_Page_126.jpg
570408a845d4aa0040ef285709ff2ba0
0916b8e409eec0b88fdb4fa25dc49a99fafd184f
59192 F20101107_AACCIZ reckdenwald_a_Page_110.jpg
60ce06c8b01da8c603e64a65ef287c57
0be24e1122a877aea7617fc6a4e81e57ed9c19ed
1051941 F20101107_AACCKC reckdenwald_a_Page_013.jp2
5bff079fb1ea23cc49f9851da76a5dbe
57f8f33db8adbf4f5d601285dfaa630887a6d58d
133903 F20101107_AACCJO reckdenwald_a_Page_127.jpg
030d16217d06168905b72a37aa3a49f4
673c1beae0d99a37ecf142c21ada43a5d2c1edc8
1051983 F20101107_AACCKD reckdenwald_a_Page_014.jp2
fddc6eb5fbb7d593f8e82a881eb79bef
4a453094faadc9d6226a46b968043de725f94f74
87403 F20101107_AACCJP reckdenwald_a_Page_128.jpg
55251adb61b2e52928458d5c6fd5587c
6f73cc1d738ee5c6e44daacd511df8299f32cfb8
1051937 F20101107_AACCKE reckdenwald_a_Page_015.jp2
3125a9a7449ed1e2dbce8a9617850740
333bf7616f0cae45ee658da2175e5ae35fb21400
45651 F20101107_AACCJQ reckdenwald_a_Page_129.jpg
c226904191da8253d51e1463fee5a4e1
565e73624f39cab48a22d13ded1768ded5446790
1051970 F20101107_AACCKF reckdenwald_a_Page_016.jp2
58d2f26d587b1a91927cfc527dc8b16d
8265c930259d790c7495950fc9c5ca448573a186
28033 F20101107_AACCJR reckdenwald_a_Page_002.jp2
19d6a861261844d869a6ee03d993f651
66d7be3ca76ce9e9889f52ca286908215e3fac7d
1051969 F20101107_AACCKG reckdenwald_a_Page_017.jp2
d8e4e81826a1c05cf4401ece89aaff77
2c2932260fddf0548763d0b1dcc874aa1143d4ea
584602 F20101107_AACCJS reckdenwald_a_Page_003.jp2
3323c6e8e6722fdde7ec7a121bf6029b
c93c9aa20a4439544ff14ec15f6e33cb8563245a
1051949 F20101107_AACCKH reckdenwald_a_Page_018.jp2
723272be72b7e579ce1595f8f3a1479a
b72240fe21fdd9868bcc21f50102064ec390ada1
1051982 F20101107_AACCJT reckdenwald_a_Page_004.jp2
b66fac6b42034db53352e483e5b2da19
fc85d2c876a9252ce96e2ed8c3c654eb88e1f2d5
1051986 F20101107_AACCKI reckdenwald_a_Page_019.jp2
60476cc655458e99c750dd691494884f
607ad81599a76b519e80b74eaad93a6ab6d0171b
1032946 F20101107_AACCKJ reckdenwald_a_Page_020.jp2
5a42e5a7a7f3c26c43d4cc28569f97a1
8a0461eaa0c6ee122f67969e4938d0eabccdbe84
F20101107_AACCJU reckdenwald_a_Page_005.jp2
9589ed0649581a107b892db478afea07
c1ad4b6ff693251f48ef56c8d7433280b5f3775b
1051965 F20101107_AACCKK reckdenwald_a_Page_021.jp2
ddcd7fa6f50f1c2603656af238b63eb3
dabd3546ffc03a232a79c316cbcbfb0cee421468
265918 F20101107_AACCJV reckdenwald_a_Page_006.jp2
a81b85c12ff63379bbd31f54644be49a
c04e390f88fdd4b58771ae4b168143b3d4904925
1051934 F20101107_AACCKL reckdenwald_a_Page_022.jp2
11b7213856cb52149d62af03a8e178ce
c689cce76884f6ebd645179685c2529e0d2cc973
1051978 F20101107_AACCJW reckdenwald_a_Page_007.jp2
95cd266a3db9461d73c6bcd4ce4f037e
50610702211825607221119bb78dfbba9e53befa
1051974 F20101107_AACCLA reckdenwald_a_Page_042.jp2
0362da1364aef5602436b28fe8c9ece7
9195f9f42168b18c466b3881e10bb906cc87a505
244537 F20101107_AACCKM reckdenwald_a_Page_024.jp2
eb6ab00ca46ffe0bd369abaf430b72e8
db09b9c84593e9c2ce8439c220931897be6f4cab
920619 F20101107_AACCJX reckdenwald_a_Page_008.jp2
8271ddd3d31e0570eb8fa5b84109928b
c5ced8f75c1c988e77ed4c60ffd46f1dbbd498de
1051908 F20101107_AACCLB reckdenwald_a_Page_043.jp2
260fb02ee3ed55a22b5e677d976e4266
a861b1a0bd7236f87489c4adedf6fb985c41d3ee
243465 F20101107_AACCKN reckdenwald_a_Page_027.jp2
5514d1a718181c1088a686cd50e2e96a
b9c841d58458b42eb97022fb8c7881445085a320
F20101107_AACCJY reckdenwald_a_Page_009.jp2
b6e6d9b673a480ef6bd8bb8950f4b02d
6cf18393d8053714f2ce0577426a801677848b41
1051865 F20101107_AACCLC reckdenwald_a_Page_044.jp2
fbe5fa6b338eb56a295c1777a7fccca3
14896786348917f136a64fb7173165e8c39f839b
F20101107_AACCKO reckdenwald_a_Page_028.jp2
37a334f87b32ecffcee8b6c65c964015
16039ebfdb76e19a7fe45260b21dc32a9d90b5db
917479 F20101107_AACCJZ reckdenwald_a_Page_010.jp2
e6f92c54c4b507475c9b73e8850200c9
7d5a93501c28d8e49fb3db92c5bfdf93c1a4e37d
1051975 F20101107_AACCLD reckdenwald_a_Page_045.jp2
cf88ead380249216919cdf3d7e827297
c1ea4fb2d0612578b742728cddaf062d7becd4a9
F20101107_AACCKP reckdenwald_a_Page_030.jp2
f7b753dc956867faddb6da91b95f6c77
3998ced1543f63d83ee6a8a99b8b35d779b03f57
1051954 F20101107_AACCLE reckdenwald_a_Page_046.jp2
4e1ba5d28993eee4e651d7f3783bf7c9
3702029859edb2fbd23ecf97ca183a3d5cd9fe2b
1051985 F20101107_AACCKQ reckdenwald_a_Page_031.jp2
457df4f20460be90cb7d5f9490782752
5b748a8bd5ac53ceb3fe9b6b8f515630082caea9
821325 F20101107_AACCLF reckdenwald_a_Page_047.jp2
e6b20f6fca83fcdecbd24bdb992d3c29
5c520c4d2b60193c141eb762ed2f0dbcf98a93ac
1051932 F20101107_AACCKR reckdenwald_a_Page_032.jp2
94a82bb1cf64d69b4b047fd4594a61f8
a1124cdc154f5d5e691778d9874f4d41ab2c7d97
364264 F20101107_AACCLG reckdenwald_a_Page_048.jp2
a35a9647e57d5a93a6900874a081d4b8
3dd6839067b970a996360dae5a1712ef0d601669
F20101107_AACCKS reckdenwald_a_Page_033.jp2
2cce413d38d26e0eb9b27dad59ae602a
d823be8630b350407c141526ff0c850480e3cf53
F20101107_AACCLH reckdenwald_a_Page_049.jp2
a41339bc6dde956d3c2ef9fa44bec4f9
fe005559dbc815eb1f50e045532200f4a93b1446
1051904 F20101107_AACCKT reckdenwald_a_Page_034.jp2
719db045951fb09e8edc3a0a8fb385cd
e6684144e6891e60990fe9514e1cf1f606b2589d
1051958 F20101107_AACCLI reckdenwald_a_Page_050.jp2
cfe34c5df9380d9b2d756342685dbd43
efe3a4a6b09c3b96d1c9b14716f8764183db75e4
F20101107_AACCKU reckdenwald_a_Page_035.jp2
6484f4590a78b3ff33c386c199a5c381
6cf052875bd3a4a11a57c5fed5378015bbfd3427
1051973 F20101107_AACCLJ reckdenwald_a_Page_051.jp2
c699eaf4ba39146edec099ad9fe67528
49ebb17db687960b549b4e53545adbfd6d26594a
1051962 F20101107_AACCLK reckdenwald_a_Page_052.jp2
733a671f61230192ced491498ea089d4
52af92954ea11112b41e6b99f012ebcd5b77ce98
1051933 F20101107_AACCKV reckdenwald_a_Page_036.jp2
7a431aef178c4bfaaa4a65c1a9057915
6a31be9c56a439a99fa021cba6ec00d51faebee8
1024881 F20101107_AACCLL reckdenwald_a_Page_053.jp2
7c18eb9b083803f9e223b047f28dd1e9
cdc09f598aef9d0b807a461ec3e5d9cc78093feb
F20101107_AACCKW reckdenwald_a_Page_037.jp2
724e2cd9d2ebd770334162b5703fd498
35feeff6492c4a3bcb3d58dc722c3fa7e6f65387
1051964 F20101107_AACCLM reckdenwald_a_Page_054.jp2
4110b047e9fa8c9415ca7ff2c57a1388
5bb6e6083cbbf24e54534bd309dbdf7c8f53f3f5
1051957 F20101107_AACCKX reckdenwald_a_Page_038.jp2
f31645d397fb3afe20733050efc142f6
96c4d7bb486046e1a2412a33acdd5790fe79960c
925717 F20101107_AACCMA reckdenwald_a_Page_068.jp2
61a4c772c4a496a6c3d6ba554a012a1d
2c866b5098f1789c7d2eebd4d1363b881dbe89a5
1051903 F20101107_AACCKY reckdenwald_a_Page_039.jp2
468e2770eea0c47d0c760e04dad5416e
ed6e42a2ffbd619f0f91b6843e3142c2c04d5d81
594196 F20101107_AACCMB reckdenwald_a_Page_069.jp2
5cd98d58b8d5cd416bb4851eff5d65e8
9d216d2030c01b158f9ba6381bfea4f1eef206d3
1025816 F20101107_AACCLN reckdenwald_a_Page_055.jp2
79d0a9c7d4ed68f847c210640fb0f9e7
ac2438be397f436b99d2cc1f18d28eef6c0b6dbb
F20101107_AACCKZ reckdenwald_a_Page_041.jp2
d7e4fad0f5fae8b23e749c8f1e64ae6a
378c6b155120b4785b3fb3e140c66c8d0ebce22b
630491 F20101107_AACCMC reckdenwald_a_Page_070.jp2
4b7311dfc7c25c250b31da2ba4c5cf94
6ede3844f5dd3cc3f682998c56a32b33bcb9821d
1051953 F20101107_AACCLO reckdenwald_a_Page_056.jp2
288e0bf0e933af623b2b9539994a5627
81ea75900190ed5fa825fdb05ffe64ff6dcf57a7
224946 F20101107_AACCMD reckdenwald_a_Page_072.jp2
70abfe13a0fa7e17ba6ea1dc2af0f3eb
5e846544f77295c164fe14e8d44207645a3bd77a
F20101107_AACCLP reckdenwald_a_Page_057.jp2
27b2dd7cb6f41913541431d80e051f2e
72b33a7a3563a0f8482c5a7fa39986a8df5eea5b
225943 F20101107_AACCME reckdenwald_a_Page_073.jp2
e08a828a11f68813461c18b21c5d9c92
d00aeca0b0d797117b385e8010ca520a1680cafa
F20101107_AACCLQ reckdenwald_a_Page_058.jp2
3a9d5b3c0594d15f8ec4d4510b5a27ca
2fa009a89a927c4e860484e4c3599bf1bc9c69c4
F20101107_AACCMF reckdenwald_a_Page_074.jp2
fe60070a3dc30bd93483ea3b2ab8bc94
7c99e83de038194e2015d247a302b6a372a13aa5
F20101107_AACCLR reckdenwald_a_Page_059.jp2
e221053f5831eb1c08cd2a749692ddc7
6d503d021c7e7693a2c21e7af497fd298f400b55
F20101107_AACCMG reckdenwald_a_Page_075.jp2
bf2548939ef6f96e93d7ec55a8541a37
5b4c81ef2eb361f52a25f62ab1b138177fb1456e
F20101107_AACCLS reckdenwald_a_Page_060.jp2
386d4aa968c2893dd36bc6db4f7d0a1f
18b5efdbdb4c117bf34cd345deac33c44a01ab0a
1051928 F20101107_AACCMH reckdenwald_a_Page_076.jp2
e8d01e487390ee0b6a68c3ae70b9b76b
fd1b92f1650ef514c72754b1d4318bd1c4041663
1051945 F20101107_AACCLT reckdenwald_a_Page_061.jp2
3b1461c8e0a927fe67f9a0490924fb0d
600b18f0f60eccea06c798d1cdf557ecc8adcc6b
1051885 F20101107_AACCMI reckdenwald_a_Page_077.jp2
77a85ad03618ec4a04fd7f140d58e839
b871c2e8d0596b5f82f19ce24749d05e4ff24c1e
F20101107_AACCLU reckdenwald_a_Page_062.jp2
1e9e388d5f0808513260b2634709417b
ae8af824c5b7a3bacffd9e20a7ceab49473a5836
1051971 F20101107_AACCMJ reckdenwald_a_Page_078.jp2
34b2afde44519def8293eb9331da7244
d76fa65bed0067041fad2aee8ea5fa37889d0f88
1051961 F20101107_AACCLV reckdenwald_a_Page_063.jp2
d30a41cfb2966933edd6ee3d1c1effdf
188ddf6ebc46b46d559e448f7f267277e6f3a34e
1051980 F20101107_AACCMK reckdenwald_a_Page_079.jp2
80cb9e363d71012c4b857e878de20c93
a96e8e39af0fcb533158bd93fc15a1900dfdcfce
F20101107_AACCML reckdenwald_a_Page_080.jp2
3c8cda3e2b904e62ab0d69a9fa33e7dd
6db51cb7324bf3c93486d10eb7e3d6d483437d24
1051940 F20101107_AACCLW reckdenwald_a_Page_064.jp2
bd359ed17519d2acfffea8a9dbc1cad3
502cf4316ccae33b1761faa550130f8fab447382
1051880 F20101107_AACCNA reckdenwald_a_Page_096.jp2
899f99d25ee3008fb4ccb93f6211708d
5078ec352e6d5791897c99fea370a31d39deba67
1051922 F20101107_AACCMM reckdenwald_a_Page_081.jp2
11b91e85630de5b1e40cc5ff6d75084e
3b2ffc7b355e99f89ae844de70ec3148f0d211c4
1051960 F20101107_AACCLX reckdenwald_a_Page_065.jp2
1cbce6d1455fee78d9ccb5da8aa9ba01
63a0c2337cc02e42c8c85858b10889a976136b62
1051939 F20101107_AACCNB reckdenwald_a_Page_097.jp2
29ddf58a0e92314f91528dc20a3f5d86
5b7ee7c6bc8d8bb7937d0d7b5f5319947d47e03e
1051956 F20101107_AACCMN reckdenwald_a_Page_082.jp2
e99893105f93a179ae8d738cf0452947
3fc6e6cc9c449d44c42acc0dd43667881653b231
F20101107_AACCLY reckdenwald_a_Page_066.jp2
1a38b0ef3dfe78e25c9bb456fcc883d8
dfdb3f69ae88dab39b30f0968082ae22622c4a2c
1051919 F20101107_AACCNC reckdenwald_a_Page_098.jp2
9503e02b0a10c9e23bf2c61940e54403
64792f4bbd33d08fa01b479559b05f03890f5215
72847 F20101107_AACCMO reckdenwald_a_Page_083.jp2
468c4f206632eafe4ab6084de9b4ea26
fe99c11d241af078a305b2c6ac1272f418fe2b02
F20101107_AACCLZ reckdenwald_a_Page_067.jp2
2268e5f718b855222b4c93ccf9cfc0fa
1a08a1d96d5fba7802fb800519ac7d634fedf3d8
F20101107_AACCND reckdenwald_a_Page_099.jp2
1ecedc477767e89be589e2d7a2756b08
fffc56d556a2532a267fd9075b8617c90591ef60
586835 F20101107_AACCMP reckdenwald_a_Page_084.jp2
f337b299be66fd903e4c93ca5c163fc4
dbc0c53a867d9988371d818e82f59161ab2ca6d6
F20101107_AACCNE reckdenwald_a_Page_101.jp2
af633012269f1c7486a3ff8c4ddf9395
7e2b46348ffc504aa0885e5aa2e32ce6e73418e0
587853 F20101107_AACCMQ reckdenwald_a_Page_085.jp2
567c10bd3239f720789962ffd4f1dc1b
4fc2d468cbf817af82be6195bb39cb09b6837e4f
226630 F20101107_AACCNF reckdenwald_a_Page_102.jp2
552711b1a0fd30064a2574c481ce6e4e
7b78efd498f8d3ac56c5cec88fab48cd9ca81b00
703265 F20101107_AACCMR reckdenwald_a_Page_087.jp2
8bf4e66a3e279f1253044b7165682d7e
60b87f3af2eaa6ceda3518d79e3c85ab01a2fb4c
934399 F20101107_AACCNG reckdenwald_a_Page_103.jp2
714b882fa8ebd1fb5ce1770418a8173b
fd91a3412852b8810b7fabac19e83f246cdfe65a
685728 F20101107_AACCMS reckdenwald_a_Page_088.jp2
8fbbd5bc7c81729776abd7f35932d1e1
11a0c0a2f9d9781fa2c6979a4a585c409dbed083
F20101107_AACCNH reckdenwald_a_Page_104.jp2
915965ee019c4cca8d92321b41daaa76
29d5e5abbda5dc872aa6240e49b6b4a2d7e18ccf
158251 F20101107_AACCMT reckdenwald_a_Page_089.jp2
e35c3e542060a517e7248a6073cd54e2
8cd4c85a8b6ea2fd32d167df0badb9ccf032fb9c
F20101107_AACCNI reckdenwald_a_Page_105.jp2
77b90f4ab5644d137628081f739d8138
7a540716d945c08e41f4616dfcd45a646c2f0154
1051968 F20101107_AACCMU reckdenwald_a_Page_090.jp2
40f50b64233fed5b38ef38c7929f1dbc
e2b3b429fd2212334a88496a2e607b9e4be7f2f5
F20101107_AACCNJ reckdenwald_a_Page_106.jp2
2b1c16e002c4beafee2a8e14c0ea31f5
f7e52021664a0ebff5558eaab6099417c69c52a2
1051948 F20101107_AACCMV reckdenwald_a_Page_091.jp2
99a378644d2993ace1be98cca4a63b1a
4ddf731ca06702d93214319ca52bbf97c5dc8d84
F20101107_AACCNK reckdenwald_a_Page_107.jp2
80ba3156e757ba864be1a7e41b9b8d9e
211ee2c97d5034dc45d20a1faa67393487ee22b0
F20101107_AACCMW reckdenwald_a_Page_092.jp2
c09edfdb049357d98ea6f8ea650f48cd
9487067b1545d60379c0f4728f71f70253419718
77022 F20101107_AACCNL reckdenwald_a_Page_109.jp2
a702f60b4bc43568d75ec8de2200000c
4d5ae7f1da2112ec4fa3522c21b80e7c7c001b10
975289 F20101107_AACCNM reckdenwald_a_Page_110.jp2
05008f2993d68d16137a3c474138dafa
5ebe05da16fd2275ed3f4efa95908101fd119da5
F20101107_AACCMX reckdenwald_a_Page_093.jp2
9b5fe089eaca10df028f042b7ea4da4c
3a79fd89168ddcc1c6fd1ed61bf74ab9521a038c
F20101107_AACCOA reckdenwald_a_Page_125.jp2
f73d5c13327dabc9c19e356b98d23de3
40b78ab0002fd7510a54ed1213355409af7ba26d
77410 F20101107_AACCNN reckdenwald_a_Page_111.jp2
dac2699c6680c2f297a22be850f8afab
2425a958194f802c2afc47213949c9ac9bab0830
F20101107_AACCMY reckdenwald_a_Page_094.jp2
18b1dd290cf99c04b2f365e716327e86
170360a560ad19cb679288c53a37b632ac81aa19
1051867 F20101107_AACCOB reckdenwald_a_Page_126.jp2
bcc6c7e69258c735fc091a1ca06dfac5
e8e2e0976f3f90d7e6527ac7941fbe387caecf29
980608 F20101107_AACCNO reckdenwald_a_Page_112.jp2
b3f18f805bee76f954ec56e4a4398e05
1561e013bf03b737d78058d200f304b99c0b2338
1051906 F20101107_AACCMZ reckdenwald_a_Page_095.jp2
816ec5cc78e9917f1dcf693818a62db4
cd91b6372eacde2866315afaba18e36ad445d72c
F20101107_AACCOC reckdenwald_a_Page_127.jp2
7469070fea40a391b033cd43bc2272dd
7c45c2ddece37e4ad34a3bded0f0c17f2f9e48c1
84895 F20101107_AACCNP reckdenwald_a_Page_113.jp2
cd7623e21f559764f6b194ba98c980b2
0b0919c407646430094d8776a5b987036457ac59
972080 F20101107_AACCOD reckdenwald_a_Page_128.jp2
fcdf65b5b53e4529ee102b97fed08620
b339e6faa6999ea001452973f6dde9cd4beed99b
905139 F20101107_AACCNQ reckdenwald_a_Page_114.jp2
21122302a08ab82a690ae5ee6fe61bba
3326300b940b2c518870562c21b8889c5f368401
472175 F20101107_AACCOE reckdenwald_a_Page_129.jp2
f3bcc82baf85b7d552ceadf715d147ad
7f7078fbe620e357d0c6aad6a2bd4a1cd05aea0b
85018 F20101107_AACCNR reckdenwald_a_Page_115.jp2
7703f76209c30b4aa308ec64e79d5f36
8148f7e53b32ff5e6e9986e71a9b87cc043a63e7
25271604 F20101107_AACCOF reckdenwald_a_Page_001.tif
d96ffe36de2213ce6d7a54170d6f0fe1
86933bbd40274089f44425bce32cd1d38322d579
912838 F20101107_AACCNS reckdenwald_a_Page_116.jp2
7378b3fed25dd3fd88aa59f2a143d897
9d0f401b027836012be7c6c8a1bdfd3ad503c1f4
F20101107_AACCOG reckdenwald_a_Page_003.tif
f5623ece08c952c2cd99cf3ea59660aa
27d95d35fd53e42a6750660ca14b24c4859b69cf
387338 F20101107_AACCNT reckdenwald_a_Page_118.jp2
a03f8499227cc3399baae5a7f84f15b0
0af00d0bf81af89ecc78e3dc022fa20a2c98a1d6
F20101107_AACCOH reckdenwald_a_Page_004.tif
9987aad51bbaf8286a03763f82f8fccf
da62986daadbc10617daef1ac77d8f3966c64269
F20101107_AACCNU reckdenwald_a_Page_119.jp2
091d3bef97590b9d4a83e354cfa1891b
5bc1537ead439300a2ad9b2126fd26d8840d719d
F20101107_AACCOI reckdenwald_a_Page_005.tif
df160ad34b17db0f9896922fff28a592
d3f383b6833c0b1c66883d64ec926da8cb8aad71
1051947 F20101107_AACCNV reckdenwald_a_Page_120.jp2
33a3ba5f9080773ea268e4c5a94d6bfa
8ac4e4ba21fea7fc850907d5295e721e84e64d02
F20101107_AACCOJ reckdenwald_a_Page_006.tif
c755db79b41dc5deb809d9a0d77827bf
9ac5a3a978b5dd836cd7f98fc802797b23fbe5b2
F20101107_AACCNW reckdenwald_a_Page_121.jp2
11d9c84510708d4c7fc747543c9a9e5a
3fc4cbef6770ce490c237eb84460856fd03c6488
F20101107_AACCOK reckdenwald_a_Page_007.tif
b63307dd2a59befc6e1df722d7aba55e
4c84a58bd59d38ecf6f7aaa040b1d589bfc6e415
F20101107_AACCNX reckdenwald_a_Page_122.jp2
8638dbfc3f37e60892fffb5a0e33b745
3e1aa81d8e031d9198545552188e1eb19ad09dd8
F20101107_AACCOL reckdenwald_a_Page_008.tif
64f1a0c8877683c1953bc0f6570978c6
9d6d9008620986b35043994e847cd6de1e3defd4
F20101107_AACCPA reckdenwald_a_Page_023.tif
7e04b1d82faf9e01d9f1a9e40a295000
514be2a916750e5dcf028be6f35a3fd088138ebd
F20101107_AACCOM reckdenwald_a_Page_009.tif
ae3b6c638dfcb181bedfb9bae723ed06
da9b736bbb8c72c7ba30c858c2faf28a39ce0f75
F20101107_AACCNY reckdenwald_a_Page_123.jp2
8b81de9ad6ae79ba883f475da6a97027
ad4bae9eace8accb0de44516480ded96649fa0b2
25265604 F20101107_AACCPB reckdenwald_a_Page_024.tif
ac4b051e2c07cd3c85f346323a8b1453
f43f3d7ed20d3824c913053d32269295f470ddd2
F20101107_AACCON reckdenwald_a_Page_010.tif
7327816895f3defdb402a2e87879faff
11417db29c84d864859bd5e8b2bca2706218f5cd
F20101107_AACCNZ reckdenwald_a_Page_124.jp2
bb0881fe291719ac460fe00fae8e992a
d7f838433a6de73df9e8afca72ee5ade5370dc1d
F20101107_AACCPC reckdenwald_a_Page_025.tif
100baeaae8b8c4c619241c4986d528f7
a52123b26f315fdc7de371d31f1a2b3ca28605ac
F20101107_AACCOO reckdenwald_a_Page_011.tif
a8abbc95cbef2d0996571498179d20ad
ca9b848009b3fcb94b77496be1d0e354ea4c4ca2
F20101107_AACCPD reckdenwald_a_Page_026.tif
e3e123401b54d9c500a737a6fec40683
26f0fe9eb9b8180748e935176452fb17ec9ad510
F20101107_AACCOP reckdenwald_a_Page_012.tif
3285d1535e881b5df06bac1e4b3df950
b74f4475dea5f0550b1f8c30ef2f9a740f088b3a
F20101107_AACCPE reckdenwald_a_Page_027.tif
c6d5b64c7ee6e05aa1aa14b29637e2dd
59234e1ed2266c2709d6e4092a41855ddc3f1157
F20101107_AACCOQ reckdenwald_a_Page_013.tif
ed9ab7ee671998fcefc71b34fb3a0a08
bad7d472ddef8ec2e476c2c4127e716cc6078bf1
F20101107_AACCPF reckdenwald_a_Page_028.tif
08990af4c4bc90bc029e9e89bb2ecc7b
f704daebd01fc44797c71fdd276f77892f5bdbb1
F20101107_AACCOR reckdenwald_a_Page_014.tif
8ca8028e18c11e1a0c39f7070a2e8fc9
f412b7eeffc73a9122435ca4a4c908d05b129c6f
F20101107_AACCPG reckdenwald_a_Page_029.tif
bce7fa0d1f26181cf48fd8135ff1e03d
f66a9d65e7559b92998d8b36517465346320bada
F20101107_AACCOS reckdenwald_a_Page_015.tif
7fc6e8ccaf92b424352762a8c49b72ec
7cc51b7dabd820fb29e5975b34772df0459c2f98
F20101107_AACCPH reckdenwald_a_Page_030.tif
e1a07512d73ac5c8690186e965a8f1fa
7c526170a8f3cb185b8feef447c2068b7ccb6d0c
F20101107_AACCOT reckdenwald_a_Page_016.tif
4e4ec6c3c666b04f45daf2657b4433a0
6803cd30a9a1cc85e01657dd2cf62e6f2bdbd24f
F20101107_AACCPI reckdenwald_a_Page_031.tif
a3dc7c05ff46911fd940b8343188cde2
500561954ca5663136c396eba411ffa3336c056b
F20101107_AACCOU reckdenwald_a_Page_017.tif
4f5bf4beeb85b2df555576e96db999cc
4b9cba2bdbcc46c104162a19bc6e940e25a31e67
F20101107_AACCPJ reckdenwald_a_Page_032.tif
93fd9ba6c9d4b03b6a0b41fbbd8d3262
a62ca51b53b5d6537cfc1558fea77d946bed48ad
F20101107_AACCOV reckdenwald_a_Page_018.tif
7e2b6731f8e2acbdfce027b990fea467
ed5b16bb1aa3e5fa073c979fe238afb319f5c28c
F20101107_AACCPK reckdenwald_a_Page_033.tif
15cc02c72cbc28dccb5ab1cdb89e9960
6043c6708ed212e00275b2a23fa716551baae172
F20101107_AACCOW reckdenwald_a_Page_019.tif
80e12cf910e658a7bb2e9aba5cf4861f
e357a378f36b3a37b0fc7ad1a0ec3755ef12f706
F20101107_AACCPL reckdenwald_a_Page_034.tif
a5c767272ee0f4429a3e51d8d783279d
fc4fc6102a01d09d675e5923d5cb0bbfdf2e3f0e
F20101107_AACCOX reckdenwald_a_Page_020.tif
6eb03b174f94a870ac6f16c00a437213
cd83658a7f0934a12cb7128f55461dac9dbdce7b
F20101107_AACCQA reckdenwald_a_Page_050.tif
e3df3d9e521520b1b88026b8f5192cad
ca1cc9ae03bb9efec7162ab2ba8f2d767eedfcb1
F20101107_AACCPM reckdenwald_a_Page_035.tif
21dc16dab261b3310621b0cb8fd41442
b2fde8c1a9d134a4ccbadbd6d97101c238ee1497
F20101107_AACCOY reckdenwald_a_Page_021.tif
5883ffc7835e813f5c1107b49edead8f
b1964656687391e2d3653599ec415baa604a7567
F20101107_AACCQB reckdenwald_a_Page_051.tif
5097ea7a626e2e6d7dcd19ec65b48a7a
bb39eb8ac0eed7ddd20bf508e2408c4649a3a310
F20101107_AACCPN reckdenwald_a_Page_036.tif
c6388e42e64b0218fc8c7b32acbb4436
74055ac732683450e5563274ce3aeac70df11384
F20101107_AACCQC reckdenwald_a_Page_052.tif
678eb3dfe453987e736245561533f7fd
0f0054922a6aea3957c1c4a3ac6021a99a2fa156
F20101107_AACCPO reckdenwald_a_Page_037.tif
12f84644f3faf1f38f820b0482711f21
cc4a060e351b616a0ef14d15629eee511bcdf989
F20101107_AACCOZ reckdenwald_a_Page_022.tif
6574cdb61d009c57f8a39f7f47f7e02e
c236f14b9553c283a33969904f9a2bbfde84737e
F20101107_AACCQD reckdenwald_a_Page_053.tif
0a354303fe3e4df695922c821b0dc45e
776f75a3f004f62dc828a9a8f77c87a0c43aa6d5
F20101107_AACCPP reckdenwald_a_Page_038.tif
98437536096406079b01c8ee05e2669f
12a0d1fce3c6c12c564a2bdaf254755bd2e59935
F20101107_AACCQE reckdenwald_a_Page_054.tif
98c14450558bcfba66658c6389070260
24b4ca32098ce5b58293a4828d78f97a9ea6d77a
F20101107_AACCPQ reckdenwald_a_Page_039.tif
2a961868f6165d40064d06160526e770
1d562bd10dbf0a58b7323f70ea2765aff26c1f3d
F20101107_AACCQF reckdenwald_a_Page_055.tif
533abf19d059b0e3fcae97de19c98d23
407bcad08c5d1559d4fde1b0fe4d50fa6da9ade3
F20101107_AACCPR reckdenwald_a_Page_040.tif
5b7a8854566e0ae859ffcbcca44da622
224e6057d9ef8c755bae86352ab8a2a2a4cbf0c6
F20101107_AACCQG reckdenwald_a_Page_056.tif
b1e1f7641600c334494d2bf984fe974a
a9656be13e78e0f9163e22f431e53da5eb7cd3ca
F20101107_AACCPS reckdenwald_a_Page_041.tif
11c11de7f3c12543cf0fcc88e63f929b
6b601bd35bf69cef3c17adad7ec616e1c40dcb17
F20101107_AACCQH reckdenwald_a_Page_058.tif
3cc53474642d63a78ffaee66dd2a4801
8ee6898f019b4abbb8f231c7e7fdbf24979a45b2
F20101107_AACCPT reckdenwald_a_Page_042.tif
334eb8a1c70b1a7d6638f0d9cb412af8
33a2d8bafbe3a61a68adb98910c1e1479fcfe729
F20101107_AACCQI reckdenwald_a_Page_059.tif
26d9692558480159b2bbe519d187f1ad
db230dd3ccb80eb5d0123d96dc2c262efc498b10
F20101107_AACCPU reckdenwald_a_Page_043.tif
91db9484b7b91b650c4546bbbee4d6a6
2c1e073a83c27602821b91750b92b69d1c8978b8
F20101107_AACCQJ reckdenwald_a_Page_060.tif
b6edefe8e8bd09fa3c7ce441eed74f0f
bbba40e6600aeeac7b64985eb088ac031d4f0670
F20101107_AACCPV reckdenwald_a_Page_044.tif
b0fdc815b9b6de15f6fcc7999868d4a3
3dccf38947ed561e6d05eefa071d873cc62ac4f8
F20101107_AACCPW reckdenwald_a_Page_045.tif
be30c4498c58bd26a2eb8b6de6a4e078
70cdd575f9fd7ac91015413d88605cab64039aac
F20101107_AACCQK reckdenwald_a_Page_061.tif
ef844ef161280aca7ada6c2832a46b9f
eadf6bffa56c340d507130b464f72f671b01301d
F20101107_AACCPX reckdenwald_a_Page_046.tif
9c6902f315aad68559bed1d8c89d6759
447567c3462fbcc986a236b988ae21f3eef73642
F20101107_AACCQL reckdenwald_a_Page_062.tif
c9cd6151e116b3d1612b4eb2a90fc2a4
3cea3135aaa994735a5c209b224208389a7ff122
F20101107_AACCPY reckdenwald_a_Page_047.tif
c387cb7305bf08f7fd2100f03ffeb802
c3ff0d4e17f675597c6452d118cdb02cca5755b2
F20101107_AACCRA reckdenwald_a_Page_078.tif
687e871f23e2b10c828ff0edd35b12e5
d3160343c996e32d7f059867faecd9988809d7b0
F20101107_AACCQM reckdenwald_a_Page_063.tif
f9b56413977a51c9f6bf6a31c51c1d3c
a65e398b7a2e619a79b89a910499f3d0e498b815
F20101107_AACCPZ reckdenwald_a_Page_048.tif
34900438371e6a9e30db6161ade9db63
13c38ca7fa18ca850f61476404a78569bc16f45f
F20101107_AACCRB reckdenwald_a_Page_079.tif
2f7701866248f170001b7c2455903547
3799504ff8ae60ec1d8f119d62708abc2e98772d
F20101107_AACCQN reckdenwald_a_Page_064.tif
6fddd170bd5183fb9a673c4566ee41aa
85d8b94ac5277aaf3b96ef1a74b3ea2e1886de21
F20101107_AACCRC reckdenwald_a_Page_080.tif
0ca6c3beb51965cd220b79a178148fd7
e78f4d3da54b9b7551da7cdd7185b2af3a2a2fc3
F20101107_AACCQO reckdenwald_a_Page_065.tif
2f74902f2f9597d07bd53842720cbaa8
55f1b98449886eee71d6b17055fb737fdf4b0ec7
F20101107_AACCRD reckdenwald_a_Page_081.tif
3b1d970b0d6d10eea5aed62571e56e18
445f26b12a78df5fd6dca5848564aa46bc316906
F20101107_AACCQP reckdenwald_a_Page_066.tif
de3c766cb67c8d9117e7b0165e30ac15
22d97cadd4de3ad19adb6d8ffce55cd4a62e57df
F20101107_AACCRE reckdenwald_a_Page_082.tif
1e8be64b407d554260bd15a2952cba8c
b604b9ea1c1741971d0978dbc6a21e6c235b81df
F20101107_AACCQQ reckdenwald_a_Page_067.tif
275fdc97600d3993bd9570597acfe181
0fb9a597e876860aa35b4fd4936887b40384291b
F20101107_AACCRF reckdenwald_a_Page_083.tif
306a45357b953a2c1f2049c6b5725008
c392c0f221ffd73c28dace9743a18efa0344ce9a
F20101107_AACCQR reckdenwald_a_Page_068.tif
df0fe98ddd5db3c5fdde6c2a91a05ae1
9b1d9f00ac71bed09f5a4b54bfeb27a43a710be0
F20101107_AACCRG reckdenwald_a_Page_084.tif
5e2b1dee5986eeb2b74a1b59ee85b2fd
ee5376ebcd85063c7adcdfc3728cd3b9646798fd
F20101107_AACCQS reckdenwald_a_Page_069.tif
1a634e43e3b94f2aab4d480df3157c21
f8c530b117ce4d30ef8e3746fa28a3c0f354cefc
F20101107_AACCRH reckdenwald_a_Page_085.tif
2e79fc3ea479e95282bed2a634ff596d
927e5d5124ca60bb96f2a84280cf6cd58b393b41
F20101107_AACCQT reckdenwald_a_Page_070.tif
20cba8716f25f5412825682dc01c22c1
935b5eb7c73a57a25849b83b3fe8ac3d37fae5c2
F20101107_AACCRI reckdenwald_a_Page_086.tif
7304f2d2f4890723fdf75647dfd8a870
0de84adfcec30fafe5245cfb5be4006cb6b420e6
F20101107_AACCQU reckdenwald_a_Page_071.tif
c78fec869f1a06efc54f760742a179da
e47eebe9915a55c5d4a49b133d5e940b441fdb84
F20101107_AACCRJ reckdenwald_a_Page_087.tif
79e72019d24da1ef12d81fdf31d22926
8de5ecdd3895ddb099ec9d89b9f6ddeb1a34ab24
F20101107_AACCQV reckdenwald_a_Page_072.tif
b3b53298b3a4d9d222014770bcbd8a3f
434d55dc2700cd0703a27d6375896b6326c7adf0
F20101107_AACCRK reckdenwald_a_Page_088.tif
1f9c0c3d127d71e186ef49c91f5c376c
bca01a844f97423742a6b7eea08a6a8609bed59f
F20101107_AACCQW reckdenwald_a_Page_073.tif
59ee032a0c358ebcc38629241902333a
9b87b246416a5421f5c3996afdaad978401f9162
F20101107_AACCRL reckdenwald_a_Page_089.tif
829bb6d9ef157e1e7d6c9659948faa75
d7a6601557d8815042c1c16fa50af4792333217f
F20101107_AACCQX reckdenwald_a_Page_075.tif
d0b32a8fadb0108146b668e3617442be
a7d3c91cfccc6271eadcc34e2937cd26b73f2fce
F20101107_AACCSA reckdenwald_a_Page_104.tif
bcf0e78f782a15052fa151f7d9667b3e
c4b2cc417d03699b917ed1b515a5a311f8256db1
F20101107_AACCRM reckdenwald_a_Page_090.tif
8f7e87f6a361e371fc45fd8a1845d83a
4b4d0b6ff2b10f8e425fa6c00f57d789bf19c76d
F20101107_AACCQY reckdenwald_a_Page_076.tif
bc4f8438ccc117ca462bc45bd076c4a9
b8e9fded74f12c81d80ab9f229995d1632c06a58
F20101107_AACCSB reckdenwald_a_Page_105.tif
a4f789ee1c9f900aca22faf74858756d
f328e1982caf071695d8a5c5b9e5a66f481ad487
F20101107_AACCRN reckdenwald_a_Page_091.tif
69d03dccee0236a3d105faca90acc355
bb78d91364a72c9f6e9f4c2902feb23780301f77
F20101107_AACCQZ reckdenwald_a_Page_077.tif
e1bc5125a8532d79e9fea3b25f92ceea
0dcc1b3f67fdc3ea1528a86ac3c6edaecc7ce12e
F20101107_AACCSC reckdenwald_a_Page_106.tif
39c8c7e158f39e910454cdabffb70cb7
b80a3c41e34ad259e6abeec9f7436aba5d586a5b
F20101107_AACCRO reckdenwald_a_Page_092.tif
de76d264fb1c70f92b239989bb2a12a8
e4ee5203e099f7aed51674c66e10eaa76c5c70ec
F20101107_AACCSD reckdenwald_a_Page_107.tif
001f877f8385d0f2a032402567d34be7
a26625c4f075e7b9eb653f94ad81ce057bf6d717
F20101107_AACCRP reckdenwald_a_Page_093.tif
2ea6cbfa589e4eef961986ef484dd2bf
acf1b4a67a494fb16524a7229054847ff0d441e5
F20101107_AACCSE reckdenwald_a_Page_108.tif
290b5960aea4365bc6eebac9a2ae6df3
8bc66e8162db92660307b0208f67e17314d0558e
F20101107_AACCRQ reckdenwald_a_Page_094.tif
5baa5cc6e3cf687434a9a837c16c498c
70725b5e2bcc78b2d18259e95ebf78ceab288d0a
F20101107_AACCSF reckdenwald_a_Page_110.tif
ee80c7328b31e9d947e29f5addcf225d
caa33f5b5c4c3b877d6e4dddb779ca714dfff704
F20101107_AACCRR reckdenwald_a_Page_095.tif
ffd293cfa781b9ee58f9efb26efafe52
331c3c91462fc17815895f0cdf119799d59b3413
F20101107_AACCSG reckdenwald_a_Page_111.tif
54c507b215d79f941ecd96ae621ea3a8
f5c43ab5fefc2e9df7dfa52f37e68745097704b8
F20101107_AACCRS reckdenwald_a_Page_096.tif
ec94123f8c30b17c7a1f647b67275689
cb3b91bbc402e038385681816c7a9e007d106013
F20101107_AACCSH reckdenwald_a_Page_112.tif
a956fc085e27466d85391cfac7754097
8f3370683780fbd5b045b8d1e19f8d53692ebe26
F20101107_AACCRT reckdenwald_a_Page_097.tif
1989316e57c743fb76d3d6cd45f9d97d
0deedcba1e29ff08ccd4b19e97457f7ab86e5bdc
F20101107_AACCSI reckdenwald_a_Page_113.tif
f5faa8f89577fea79fba32e9a9d9fcdf
badaebb9cbde32e097d4015674087226f334b5f2
F20101107_AACCRU reckdenwald_a_Page_098.tif
46c76d6b9a6367fe394bdced3c974340
a90c857fe75a255e859da341d8d88094f3b34c18
F20101107_AACCSJ reckdenwald_a_Page_114.tif
3a331356b9a9da305ca30db3835616bf
b234fb8e2ff9e13b5735285eaff5927e71be0ecf
F20101107_AACCRV reckdenwald_a_Page_099.tif
44a71b599d63d813b8a70027a63e0c80
0a5e15a9b00227a0516d1d7ce341e93c5c7a544d
F20101107_AACCSK reckdenwald_a_Page_115.tif
8c744009787180ac2a7a91c4a6a5c5c6
ab3f2ef074f45ec4ad4a3448e39042127bb0b7af
F20101107_AACCRW reckdenwald_a_Page_100.tif
d86acd28343213d6b36a8df16d7ecac1
de3f37e9aaf70a893941a3b8d33d21fcc38b5338
F20101107_AACCSL reckdenwald_a_Page_116.tif
7b843d268ea95f42da26610f3cf1fe93
310628b9e1779267c2a354d9b5da3341853ce9a2
F20101107_AACCRX reckdenwald_a_Page_101.tif
bebefc3d5e0ad6bb859ff9e6ef294cd9
6f00501c498c57ef85d52d1942e88e6b977e5df0
F20101107_AACCSM reckdenwald_a_Page_117.tif
1e7e120d41b7c9ffc83e31b9ea3597a3
7e719f02e0a3d14f54a7cd84231ae9e737b53b21
F20101107_AACCRY reckdenwald_a_Page_102.tif
6c54ee861384f4d3d4a77cfa15c1679a
66f80b676a0b92f8515fa1e586351b64b998a4b4
866 F20101107_AACCTA reckdenwald_a_Page_002.pro
6aa0f64896aa211f3cd07c2cf32625f9
562248b0c79bb7168ab04d8f346d720c2a71f4ab
F20101107_AACCSN reckdenwald_a_Page_118.tif
f7c68beefb2263df4fb75b19de95741c
59077655dd7ff2cc54783a07e651f1178efac957
F20101107_AACCRZ reckdenwald_a_Page_103.tif
cc45334374bde34050f03a1fcb2fd1aa
014d22a36f1ef05b5a59b49f66c85d3a1fd0faf2
25990 F20101107_AACCTB reckdenwald_a_Page_003.pro
072f5eb2d656ae2209031787184d3938
28986953324ac9f44115d5399c2030ad3a0291f1
F20101107_AACCSO reckdenwald_a_Page_119.tif
1b7fe9f7a7e960941d75fd7ca22a249e
a6a139dc037875eca34949cc24c5837fdf64ac1f
97597 F20101107_AACCTC reckdenwald_a_Page_004.pro
42a41afb3b10e769262a11c76f835f73
6d5a957e3b4b5722f334c9e4616b41b30575c104
F20101107_AACCSP reckdenwald_a_Page_120.tif
93ab6499aa2cd4815eb2ac797723a2b4
d4d9652a2a684c3cf5b074a652334946333b1c33
98990 F20101107_AACCTD reckdenwald_a_Page_005.pro
f71b9f0d3a3e1546af29cca7bb0c1d9e
9a73cf3b4072a474ba13d5a7657d901b0f008526
F20101107_AACCSQ reckdenwald_a_Page_121.tif
158199e6959fce487f0808b085d9bafd
1728635fd9a0151669b36d22aa6221494e65ca16
9500 F20101107_AACCTE reckdenwald_a_Page_006.pro
bdd57cecdf6e3bf948c84a20fcf2b67e
6de04c328fcbfb1b6abebc03da7bc04a7c109f6d
F20101107_AACCSR reckdenwald_a_Page_122.tif
df546ea476c9be4976864f6f4d9259a5
0263df63008915dda808c3a3616d76d0fbbc1c59
43495 F20101107_AACCTF reckdenwald_a_Page_007.pro
aa655bd5f9dfd054e457dd3854cce7db
9947e8327528f337bc1be242cf39fcf217ecfffe
F20101107_AACCSS reckdenwald_a_Page_123.tif
1516fc6057839bc64fab732ae1f5dddf
6b7716171125fa34982e8fc2b0d8a945f4c10664
24338 F20101107_AACCTG reckdenwald_a_Page_008.pro
795e9ec1cb8bd1c9980b2f77d8d7fa75
ba8cec9f80702d03ea365a0affd079a5fe87fcc4
F20101107_AACCST reckdenwald_a_Page_124.tif
0958df4b5ee8e5cec10a4a8005229aaa
43065ca1b2b507100767edda48a8c0ae3c08307b
47207 F20101107_AACCTH reckdenwald_a_Page_009.pro
93bc03f16cd66c9a09e25f4ed45f8265
3fa4287c6c9799d4e0c4fc5e320244a9685bdbfe
F20101107_AACCSU reckdenwald_a_Page_125.tif
9319e804f6ed774d7aa4427d5b4cc2d1
80e819876924b14c17c61c80db88f8537799e8c4
65816 F20101107_AACCTI reckdenwald_a_Page_011.pro
c317641dcd98a231a28fe3d57a519193
d5f868179a61d5fa9ad32cab1a3aa49c80642b21
F20101107_AACCSV reckdenwald_a_Page_126.tif
a45cc6d2569d47942ddc903704336112
2abe6496222deaff84a22a890ccd9d457f824720
57637 F20101107_AACCTJ reckdenwald_a_Page_012.pro
4a7fd4435481717258f5e610ebc97ee4
547070fbf252fc6166fd966e0357119128639311
F20101107_AACCSW reckdenwald_a_Page_127.tif
1115ab8938717dc74faca6927386f24f
84b55b1c8d955834027e18bbe0e020e16b840c23
54721 F20101107_AACCTK reckdenwald_a_Page_013.pro
573cdbd64a3ae3da916efd1acf81064d
2fbabf99d1b651e9aeec5f0d4e8abdcfcd5c543e
F20101107_AACCSX reckdenwald_a_Page_128.tif
f310b41a51bdcc86ad986f197b1e74d3
e0a5a895c45b9f309f6beb00bcd1b4c75fbb5b22
53561 F20101107_AACCTL reckdenwald_a_Page_014.pro
bc605d37d1894b312c1937b651d5d1c1
3a303b967a3d2ca6a9ca4f1a7ed2cfa608c02521
F20101107_AACCSY reckdenwald_a_Page_129.tif
1ad896aa5ef437314cd57e5add6ccfec
cc4a029b3ce37eaebc67eda92ab8648b81237ca5
53202 F20101107_AACCUA reckdenwald_a_Page_030.pro
c0a8b134d2229fd4cd6a4dda7e4b3a9c
8eea84481749a56b73f92bf24004097ec257bb7c
56963 F20101107_AACCTM reckdenwald_a_Page_015.pro
7556225879f2131262861ee2a6a8392e
5d49a2e98df72e529f028749358997d555714d89
8352 F20101107_AACCSZ reckdenwald_a_Page_001.pro
fd83c8fcd0dd7ae7769a62ef9450cd0c
1de6f28b1407985301d90a991435195bc93fa31f
52501 F20101107_AACCUB reckdenwald_a_Page_031.pro
07f77d6ad23d40915e41dd9cd869761b
54b7e30b55e471fd1a0a780c5dd54992505da991
52335 F20101107_AACCTN reckdenwald_a_Page_016.pro
24ea0b6d98b3223c0c3cf321d797ad93
4e137fb29016810e9468c414ceafa0b9d319fe49
54308 F20101107_AACCUC reckdenwald_a_Page_032.pro
88edad53931a3d33ac4297faa64e1c93
b92df44f80d83d300cb95cf9507fc3d81d33ead0
52992 F20101107_AACCTO reckdenwald_a_Page_018.pro
cf57fb5c34f74cf3f8c892b4fa555d89
d8decd77f5a6214e03f94982a5a911e56a7c5bb3
56856 F20101107_AACCUD reckdenwald_a_Page_033.pro
c7fbdb366dcb5349dd6bf5ee70fb2c0e
c7dadccc0782cfdc26aa19481c469651688d7303
54309 F20101107_AACCTP reckdenwald_a_Page_019.pro
6114811bde74da218c0192fd3a62ed20
a61580c448c0027041c3d0670fc92d84efabb523
54640 F20101107_AACCUE reckdenwald_a_Page_035.pro
026e609617929397c7d3a9016fbe5c86
d9e26e565f84520655b730f7a60f3d7b13d37950
46433 F20101107_AACCTQ reckdenwald_a_Page_020.pro
341b94e55417589e717e3c2e8fd45aa0
621455024ee4c87b35a23c5f8cf528e46672dc78
57194 F20101107_AACCUF reckdenwald_a_Page_037.pro
6c1e6afdd97af05765ae60a2f0f8581e
355a1b56d87cc5d756d71f1bdbd0c3b24f4efad1
51448 F20101107_AACCTR reckdenwald_a_Page_021.pro
8f394d9c080e4a32deb6b58aeb19c887
8448479bbbf83d93323a04fce652153e93d5d4b7
1607 F20101107_AACDAA reckdenwald_a_Page_070.txt
6edab7f6d255823c1e4a040c3c6b24ba
5bf1db094d2a24bb5420f54a960e11804e634dae
55184 F20101107_AACCUG reckdenwald_a_Page_038.pro
825cd99e718234dfa9127f77b366b523
c72511cfbfc9a095d36a361402ec37d9e0e10ccd
55612 F20101107_AACCTS reckdenwald_a_Page_022.pro
a5221afe2ae0fd0eed4e84ccbc2d5b9c
cceadbc47954ddecebcc098a424955b923ffd26c
2159 F20101107_AACDAB reckdenwald_a_Page_071.txt
87f8f3b3e9e4a34fa3ad09cb0b2921c9
4d8813660540a7a1d2b861c310eaec5bc4c09778
54820 F20101107_AACCUH reckdenwald_a_Page_039.pro
a670ef71062dad8d933e83cc53f17bdd
ffb85aa0ef4605f22fa965357c97fb31bab8d25f
20760 F20101107_AACCTT reckdenwald_a_Page_023.pro
087ef9b3df37d52d4b460a2838fecabc
dcc486ee68643cb17e4184c7aeb13cdce561abd2
644 F20101107_AACDAC reckdenwald_a_Page_072.txt
b363b41451bc06769e79326d1e10abe1
2a910aa5cf0058fe2cab3e5242e741705af69170
56394 F20101107_AACCUI reckdenwald_a_Page_041.pro
b4b788b0df7aba364bdba85f8b8d1600
bb4fcae8c855c99a51cac6e0bfbf380a61d16c34
9740 F20101107_AACCTU reckdenwald_a_Page_024.pro
613d8e718befa9667fd91c8d6c31bd67
cc380cd72df8f37dfea8069cad37f6d8c58c3f6a
648 F20101107_AACDAD reckdenwald_a_Page_073.txt
0ce365d97aca23caa1fed758deabd5ab
8d5e90809970131a9b5cb3428c72b417fb35b5e9
52943 F20101107_AACCUJ reckdenwald_a_Page_042.pro
d452d83bf9f91d44eea38543693c84d8
9699d19bffe466d26b40c51bc48c8053c8ca18d3
10065 F20101107_AACCTV reckdenwald_a_Page_025.pro
4615166cd417dc0f963986b21bd4f2ee
5d35ebc48b303f081d9b4dc0736d288db74d0871
2030 F20101107_AACDAE reckdenwald_a_Page_074.txt
f060cb06f8048d0c1b7f5a8defd2538f
341633865685153b2704ffd1e1c14fd5c100163e
54819 F20101107_AACCUK reckdenwald_a_Page_043.pro
f826566abaa92b1feea52d2e261bac8c
4fe8dbf9f064e2d43aef4e8c9929d6d5aef47c44
8483 F20101107_AACCTW reckdenwald_a_Page_026.pro
f6e4689464c8cae9dce4139864b6df84
c2c54d38886dc85ea4a10e730d8286cd9652d4f6
1948 F20101107_AACDAF reckdenwald_a_Page_075.txt
6c7c8264ff1bc042ec9ab3b3c0659510
015f2eff2eae3ecb068a036c79d3351cc522d4fd
51511 F20101107_AACCUL reckdenwald_a_Page_044.pro
49fd9c8981cec311a2adf51eb1a1739b
0bb8bf302b358c3f91f12bc60a7666de6066df29
10718 F20101107_AACCTX reckdenwald_a_Page_027.pro
50bbc65c4ab27a384ae7053b40772e77
47d9fc9209174ad1fcb7534cfe9aa7612935a9e0
2166 F20101107_AACDAG reckdenwald_a_Page_076.txt
ee77fc32d0707c01c4c6379a63fb1944
42f2d20e2e5cb7cd630c906dca12095bc3589823
50205 F20101107_AACCVA reckdenwald_a_Page_059.pro
1bdb8b6d3a1ed9466ab2db205c10cd2d
d895dba9ee1b8b332e962272cf9273fff01b5128
50599 F20101107_AACCUM reckdenwald_a_Page_045.pro
3658331fc67e1f0ba0aad231fd561779
c268bd51e260f8e04dbe4c4621693743dd93d4a5
53171 F20101107_AACCTY reckdenwald_a_Page_028.pro
568206edbc0579a4510282246d3e897a
057e5b249a904154ec2b20e9a68e5e1444d6c433
2230 F20101107_AACDAH reckdenwald_a_Page_077.txt
144001d70ef5660402fc5388e1c43557
f069f3704f6bf3437053542afcd941da79d61539
52186 F20101107_AACCVB reckdenwald_a_Page_060.pro
6c77bd1788d06f5c7a3607580c1d9de5
277e85cdb53f4e2c1b0e4bf2a6a4cea5f0229474
51717 F20101107_AACCUN reckdenwald_a_Page_046.pro
d686c41de4d63e716f35bcbe199aa937
c44f4872b43438611a674aae7b87b936617db35f
51410 F20101107_AACCTZ reckdenwald_a_Page_029.pro
3bb06ae97837cc8fbbb0329dbe7f3d33
0a1e3fdfe9cfe99059d69ce98ba3530bf1875f75
2241 F20101107_AACDAI reckdenwald_a_Page_078.txt
16b0731faad082d9eff42eeea95ee3bf
bd31e876706142ff4ea855e37d1e4ef52d39e1b4
55440 F20101107_AACCVC reckdenwald_a_Page_061.pro
8c5c473343c92de7bd07c4556ee1a5dd
8245064cec9b0d6f3d5dcb007d2adc789583a2cf
36012 F20101107_AACCUO reckdenwald_a_Page_047.pro
03c8507be6c4966cc21250ab39a00a30
24e02408aaad884da677986942d8e0962c05239c
2124 F20101107_AACDAJ reckdenwald_a_Page_079.txt
fc7ad16f91c4a2f954e8527d6c5bf441
fa308f795d92c8d292b395f71fb486ca63167f48
54890 F20101107_AACCVD reckdenwald_a_Page_062.pro
83b61675494f52ef483be8de6f645bca
0cc88f34d4f40fc4f96dec263e2700cde5783696
15863 F20101107_AACCUP reckdenwald_a_Page_048.pro
a65d77cdad1859dd6be2a449bfd706b8
983f3cdeafb6752805febeec044820bc4c860412
66965 F20101107_AACCVE reckdenwald_a_Page_063.pro
c7f67dda081585b0f68f680a7f5cde34
8eaa4ef4e1eebf15cbf73695a29215b5e0592983
50130 F20101107_AACCUQ reckdenwald_a_Page_049.pro
4b0d256af62532cec404bd16a5e4824f
8cf3ee96edfef4daf62113abc0169c14f05cb233
2147 F20101107_AACDAK reckdenwald_a_Page_080.txt
c641f8dacf22078e2d9d1eaef35cd2f4
b3efe5b74d48e47a67f373ea481a545cfa40ab46
50144 F20101107_AACCVF reckdenwald_a_Page_064.pro
1af4ec2e05f311cb9c5556680de77dcc
cc08e8cd7cbab38e1c74d1468f518a0674f46985
51352 F20101107_AACCUR reckdenwald_a_Page_050.pro
169919195c90565ff1f4a26c178aec23
33beafe1d7f5d25b655ff883a937d9e790e3aabe
2189 F20101107_AACDBA reckdenwald_a_Page_096.txt
e83ffc51ebe2038644ba3666c7ae47b2
2906126cdaf80d538e7500c5dd98ffa6aa4fd203
2171 F20101107_AACDAL reckdenwald_a_Page_081.txt
93eff2eae09a482921b89220b9539cf3
1ff4c0fcc2af532616397565e02dd27a05a9103d
50785 F20101107_AACCVG reckdenwald_a_Page_065.pro
109dcc5c6bb5048f16bd02156e50a1dc
82be4e735041e1b65b9c9ecce6802c2886d3c193
46977 F20101107_AACCUS reckdenwald_a_Page_051.pro
632fef01727b663129e650c2da209c6e
ed7353330dfc81e53748ad41328ab01b1a0805c9
2179 F20101107_AACDBB reckdenwald_a_Page_097.txt
606e4ea91c3b2345473226069ea95688
9d4aadd8f9d3754a3cfd2abca5eaad60212e2ff8
2162 F20101107_AACDAM reckdenwald_a_Page_082.txt
70d1b3713df2f25f70612abf99277e84
febe94dc8ca3f7c71ed46591e222a88c70abd830
50773 F20101107_AACCUT reckdenwald_a_Page_052.pro
9e06dae3d586513a722ae43c8f8b5e3e
3bfdb7e258fab435eaec996460462e48ac386227
2087 F20101107_AACDBC reckdenwald_a_Page_098.txt
02d71fbcf74523e7cf955241bcf110c2
99a664abbb6be9941b559206b641110226708647
123 F20101107_AACDAN reckdenwald_a_Page_083.txt
67677221669ac053087173be0b745ed8
2c7f34197b8f43dfbd78472520d6a14d491c687c
55005 F20101107_AACCVH reckdenwald_a_Page_066.pro
a73ed9b8f0ebe9984f7042b6b3ee2b35
6d1a9dee26ec7b86fc3d08b1f6d660abe96f63ff
45188 F20101107_AACCUU reckdenwald_a_Page_053.pro
4bbadebc32aecaeebd13ff22c907689a
296892e498bad01b8a7aace689cd9793dd73901e
2255 F20101107_AACDBD reckdenwald_a_Page_099.txt
a668fa234f668e95f3ea1df4329c90a7
1cbd7028e4ea14ed9bdbeff6255338874de2c51b
1370 F20101107_AACDAO reckdenwald_a_Page_084.txt
8908ad43bd2dfa34932adb94b2ea7ada
2d3899bcfd56d1505a343c5c619f89f249acce95
55130 F20101107_AACCVI reckdenwald_a_Page_067.pro
1afdbed84e2a880ea71a830a572ad8f7
2f5d168ad8d7616e5cb8334a2299f8c69e352f89
49729 F20101107_AACCUV reckdenwald_a_Page_054.pro
84bb96646bef99c4bdee71f601d781dd
cccee86066a77bc7e6e11a909c87712ec8266c9e
2237 F20101107_AACDBE reckdenwald_a_Page_100.txt
d62e9f1567794b362da96b844385b022
6ada04344aa97d8004ca293fe176b473ff9adc42
1348 F20101107_AACDAP reckdenwald_a_Page_085.txt
be414b87d59889503153481e3b557099
618be11cb9aad2999a2cf996e758b73cff4b8a05
41223 F20101107_AACCVJ reckdenwald_a_Page_068.pro
1959511086df2a29d4aa5b0be1f6d9b5
049e986c76ed1aee942fcf8d7cac7d1768c3fecf
45972 F20101107_AACCUW reckdenwald_a_Page_055.pro
9410cfe697339670854fb7c02d68c867
6dc718196c7f2f2a659bd0aa6a0d67f4cd9cceea
2175 F20101107_AACDBF reckdenwald_a_Page_101.txt
b3b74c67fd715c6cbe5265bcd3d6d449
13d23ee18eec72daef07d30cb291464a8978f73e
754 F20101107_AACDAQ reckdenwald_a_Page_086.txt
a7d2d87c2dfa0d89d2fa36d0f4e3f2cf
e945d03fc9734ba560130e92a308ae39702d608a
31087 F20101107_AACCVK reckdenwald_a_Page_069.pro
4f55e5b0c343ba66341aafbc0cd1afdb
63bba707d9214289577c472a45be47701c079dd0
49567 F20101107_AACCUX reckdenwald_a_Page_056.pro
cd6aab0665ddb7912cf205ee2edf11da
3d93477ff7e8feed21d8c013f93db11ce13a3ea9
396 F20101107_AACDBG reckdenwald_a_Page_102.txt
149dc409c276bfc428dd41071f43a77e
8da1417ccec58a468f9a158e9797a315568847fd
48732 F20101107_AACCWA reckdenwald_a_Page_087.pro
4d4e37d2bc48cbba594be56c0acb8754
e927fe8dab4b85b4cb020d9b7c02489dd651c01a
2423 F20101107_AACDAR reckdenwald_a_Page_087.txt
91581d1fd1734999b9885897205c1777
affb8d53a3ccece0d5a96e3f31382b3c188907ce
34189 F20101107_AACCVL reckdenwald_a_Page_070.pro
53b3f9e6b3bfd257078067bcf56ce9de
4b164bebce0097da879f11f2423f52282353c733
49881 F20101107_AACCUY reckdenwald_a_Page_057.pro
cb05fd5fa6bf8ab8bfbe0ec26268ae8f
58adbaddfc9682c6791effe5369f22fef0ee09b0
1692 F20101107_AACDBH reckdenwald_a_Page_103.txt
494df4c96fc20dceb27d2c25b234390d
07897206991521e3b77346ca2a4e01b93c7eef29
2405 F20101107_AACDAS reckdenwald_a_Page_088.txt
23fd744acc9d547f9023194a0effbfeb
9a6878f4f278b604819ca9568b1d52c1c6afd697
40759 F20101107_AACCVM reckdenwald_a_Page_071.pro
f2c5170310d010e39fb352b288bc6baf
ed7a192c1469e95f7fe1a5316579cab4c960e8e0
49294 F20101107_AACCUZ reckdenwald_a_Page_058.pro
91e264391981a4a632e78eb87c7b374d
098da736a7f3b655265c0109b842965060f08239
2041 F20101107_AACDBI reckdenwald_a_Page_104.txt
285b1cc76dd00e38b7f9bf51a8c89bab
2d372eb85635f603249de2b5f42676a0508dc322
47958 F20101107_AACCWB reckdenwald_a_Page_088.pro
3cd9ce8ee508f7fede1b1a6e2c6e7360
66f713cf0c6cfce0a686411884b8a1e5a6e149d2
382 F20101107_AACDAT reckdenwald_a_Page_089.txt
09e87ea44349b51b4de64606c2399993
e8e9bc9f9f2f5b59bde2072c586e05e56b9464be
12505 F20101107_AACCVN reckdenwald_a_Page_072.pro
32e41a8db776b7f9050ba9dad8e5553f
e26f6e99478cdd6fb8459088fdd6ab5e36147bc8
2114 F20101107_AACDBJ reckdenwald_a_Page_105.txt
1699e19190c17f56f17aa19c706fb852
3bd64761d47a0f72bc12cd1d1a65ae32c616f3e8
7241 F20101107_AACCWC reckdenwald_a_Page_089.pro
77bdf6b2a7e001ccb761bc6a6621af51
1df15146483dcde62f21d9fc5f65577659a0d82d
2198 F20101107_AACDAU reckdenwald_a_Page_090.txt
2c6a3725bd9d312680b12711b712e1be
2c76b18b53ff4f4a6157374c252558d11c29d2b0
12365 F20101107_AACCVO reckdenwald_a_Page_073.pro
907a49d9f3f9611f5326b75195b3a7df
8a4652d4b2c8762a7374acfc7ad783872435a44d
2090 F20101107_AACDBK reckdenwald_a_Page_106.txt
ea563285ccc564b12a3dfc503ac56abe
42f9e5899285e64101111774b41a3fe5b078aac2
53161 F20101107_AACCWD reckdenwald_a_Page_090.pro
efc32a7f1e2a56ece928ef88f14b5fc8
d845907712cbe296467f8afc8e8684d4f8a7c62c
F20101107_AACDAV reckdenwald_a_Page_091.txt
d7fbb15e60b5d308525f8ba3b95b527f
7d6502da89c25150b0f6d151e3abc10ae27d13bd
48563 F20101107_AACCVP reckdenwald_a_Page_074.pro
ffbaf9e64e55b71841e08fd704452feb
cf7fa4516aa7e44634df346b41cdf9b108faf100
55012 F20101107_AACCWE reckdenwald_a_Page_091.pro
839a2c51b11dab1bcaf7a73d2fa6a39e
230a9c94cd79cd7a58b9ff995dc20e9cd2a30fe0
F20101107_AACDAW reckdenwald_a_Page_092.txt
cafa721b21f101221d6dc71fbf939907
909fadd9559032e4bb41222a2f70c640bca16e2d
54715 F20101107_AACCVQ reckdenwald_a_Page_076.pro
507d86e03bd1950fc2bb85dce8db65c4
837f6a418b68ee7432cf9f5db32ebef826246273
2644 F20101107_AACDCA reckdenwald_a_Page_123.txt
eb447646a63343dc22fe5d48a6359bf5
32a69155901cd55909755423c646a0414755196c
2088 F20101107_AACDBL reckdenwald_a_Page_107.txt
840d99d094f7b07565b56a73b1cb07fc
afdc025be91e9a153e091808853a9b3b70f6a7d0
53937 F20101107_AACCWF reckdenwald_a_Page_092.pro
8fb28f2b3ea686afc4380d40c53eae65
8565cb94eb212789e34ea9fa2706f9172507b083
2188 F20101107_AACDAX reckdenwald_a_Page_093.txt
f456ee355804e2dceea0ab320671cae6
caa5addb7946e2fc047f3320766b564070dadb02
56454 F20101107_AACCVR reckdenwald_a_Page_077.pro
1bd574b307cf9e7e970fbc70a4b41a64
51e9ceddff99099cf3ea43783ad5b7fff0db902b
2716 F20101107_AACDCB reckdenwald_a_Page_124.txt
94bb56fb01927a3e2569393a409d7f46
246845aa9af530b9469fbf2b243dfd98203177cc
1063 F20101107_AACDBM reckdenwald_a_Page_108.txt
7f69b567e431c3b9313530289021991a
c36bdb6d31347d7ec5adbc6898eb57d2981def91
55648 F20101107_AACCWG reckdenwald_a_Page_093.pro
10e700c07bea7f7ad2968b7377f436e1
5afae79ff64f0b96e653fd5ac0ec499c3671dd43
2024 F20101107_AACDAY reckdenwald_a_Page_094.txt
0bd0e526c6336c3a9c33e9c3e8c96dc6
27d1d9f7707e47a538b312c995a36707bff2fce6
57113 F20101107_AACCVS reckdenwald_a_Page_078.pro
fef2605ac36b7f1bcec666417e78b79b
fa642fa30125a3716bf87847ca162a61f06b7fd0
2574 F20101107_AACDCC reckdenwald_a_Page_125.txt
fa040866500c81e33c83e0b0f67ad18c
96a4c20ecc7651e92de0bd907a177ad3c55b735d
114 F20101107_AACDBN reckdenwald_a_Page_109.txt
3f7d2aa49e2e82f59e5dd50af7283160
0291c112c4d7f37567c73e1b358e27dd7f466ac3
51360 F20101107_AACCWH reckdenwald_a_Page_094.pro
c9609be96f485a48843d675c5c9b4859
ee68f95aa24b5bb9b2da9ed498f8d10eab102814
2145 F20101107_AACDAZ reckdenwald_a_Page_095.txt
15d44ffe88544b6db2ad732b44fbcece
cc924d21fa563619ee01921e78b026a5d8052953
53654 F20101107_AACCVT reckdenwald_a_Page_079.pro
ece86abce2f0c411ec549bb333ef70f2
00b5a27f3283f8d4cfe01107da84b303d2d0e55f
2252 F20101107_AACDCD reckdenwald_a_Page_126.txt
8bff3b61b32c013ed7be6b502bd463e5
f2a369d6ab9af5440214ee083e0ae54e2d17d0ff
2934 F20101107_AACDBO reckdenwald_a_Page_110.txt
49a173977e3c1038dc74e3bd305f360e
568c36e2e1ffc9b3ccd3af31a774c25c843bd7f3
54407 F20101107_AACCWI reckdenwald_a_Page_095.pro
c0fff25ab2d91dcc15a67cf43b12d53e
f3996d3e82af56e2866d70ddc2cdb9fb4023e048
54227 F20101107_AACCVU reckdenwald_a_Page_080.pro
afc8cd54ca0e3773e18d2705e3f95db3
3484ebfc50767b01eefce5d199f67370535002e0
2727 F20101107_AACDCE reckdenwald_a_Page_127.txt
4b0f42b4f9ef6fd2a0cb5216eb836ad6
6eb10011bbf93e5d31017c3c1ede9d6ff4e14d7f
115 F20101107_AACDBP reckdenwald_a_Page_111.txt
c7cf58c599dd8c1d870d5bff61a881e5
617816569763bf90ed7a9e6cf71ed13ec03a05a5
55801 F20101107_AACCWJ reckdenwald_a_Page_096.pro
be32aac330bf5da661b4877f17d7f423
3d855bae15c7a410ad63736f3d5f1fdb286e3fd3
55256 F20101107_AACCVV reckdenwald_a_Page_081.pro
e713cbc54194a1f5eb4407442d1a17bf
f9c18ca9fc732423a3277995abf3366c804279f2
3031 F20101107_AACDBQ reckdenwald_a_Page_112.txt
40b0c85a791c74daff5de74c75107dfd
fb1fd6469c68f7389822f576f1cb2e875da1e4eb
55529 F20101107_AACCWK reckdenwald_a_Page_097.pro
1b1fc0692d94a630853debf2925ec373
4bead371b591707c4cf940bb1e68315d97cfab18
2985 F20101107_AACCVW reckdenwald_a_Page_083.pro
8f8eb27c2e9ac8be9a46789dca989c68
3ade72e118b95d309f965a5807eb01510ae06da4
1758 F20101107_AACDCF reckdenwald_a_Page_128.txt
126df0136daf83778144f2f1659b39d0
a7d443c9e9929ea13438811d470adc3ceaa77710
F20101107_AACCXA reckdenwald_a_Page_114.pro
4595f2df0c86d988ebfa09863fba6d4c
be0b279a26476f11d27c71810e9c97dca0f95ca7
147 F20101107_AACDBR reckdenwald_a_Page_113.txt
ee1c287e09c6b5a3e0a208b58beae9bb
e236bed5d4699faee5852d0e1147c5d958ee0857
51877 F20101107_AACCWL reckdenwald_a_Page_098.pro
77d3ec58504c504937ac2904deb1c38e
b8fc17e039795169afab39818698e357b3ea1943
30275 F20101107_AACCVX reckdenwald_a_Page_084.pro
c0181c04f5b52359be7391cca2067e3b
a7acdb2cd8e1543f81ca30bc9d77e82c4ce27ff9
840 F20101107_AACDCG reckdenwald_a_Page_129.txt
0bccb16c2da3663d96a856387971f690
fc137cc070c1362778f888c7dc7eea5b51f8e584
2506 F20101107_AACCXB reckdenwald_a_Page_115.pro
cf85cb227ac5bdb0e203c4d23cc10cae
ab7b320f83b73161940a987d6403af14bfad50c7
2346 F20101107_AACDBS reckdenwald_a_Page_114.txt
7c1117405eabd4a72a440a13e3f4b7d9
0d38d7b26ed5d2a5c403304b4921609dbbab91f4
56440 F20101107_AACCWM reckdenwald_a_Page_099.pro
41f5516a5948bd45a61d882ef631e5a8
43562496d71c34500fa866e346498426090105ce
30222 F20101107_AACCVY reckdenwald_a_Page_085.pro
2b90196662677d08d34553a458bb79c7
77eeb986fa08ff4144e82a6403b9b11c76aa6fac
2119 F20101107_AACDCH reckdenwald_a_Page_001thm.jpg
d906dfd3150308ce5f7c7a6279de790e
02df3fe6cbbbfaf27e1fe5d3bf8d68ced999bee0
F20101107_AACDBT reckdenwald_a_Page_115.txt
fd187bfdf8a4c19b12c7ebc3402b3a93
e9cb8bfef1eda29599d2ff813b0ff05f916d5e78
55639 F20101107_AACCWN reckdenwald_a_Page_100.pro
9d3c96d230cab2dcabede49586f1cdaf
ff961e9c103104573d6cef76b2c50cf9f2d72bcb
16219 F20101107_AACCVZ reckdenwald_a_Page_086.pro
cb006e0bb1d1934c0728bdf416ba11c9
2487a22491db52de3f4ef66e3bbeab7b546e7a3c
461720 F20101107_AACDCI reckdenwald_a.pdf
63bf76b57dc129f0c67c80ed572ebe11
a03ae9084e8a8daad0bb1d8015688f9c55c42e4c
25437 F20101107_AACCXC reckdenwald_a_Page_117.pro
c25cde4ff6a27ea64b24d3cb9388faed
0b9184fab2ce2bcb2dc45ad152b092ec2a1ee186
2622 F20101107_AACDBU reckdenwald_a_Page_116.txt
f47b014974cd0b84532e9256908eaa33
51febe4c927e0bb9096029d72a0482fb83941bc5
55370 F20101107_AACCWO reckdenwald_a_Page_101.pro
87d2c0e7bfddb1f650c1f4a9594a0341
665fa69ec5339fe2aac481bbd920dc88b138c9fe
8554 F20101107_AACDCJ reckdenwald_a_Page_001.QC.jpg
7aa0dbc3dfd88b29705b5c4bdfd0bc76
a16ff4d5939babeaf4e9b3046ff000a792b39b39
22112 F20101107_AACCXD reckdenwald_a_Page_118.pro
556f04cafc70c21ab9e5aac5bdcf91b3
1272f8185e2afbee2736a1a012cc2f6a573830c5
1259 F20101107_AACDBV reckdenwald_a_Page_117.txt
94eba1a8612cd8054c596eb21fdc1eac
53c8023b8ed6c7b6dac919d1958366dccb60c517
9957 F20101107_AACCWP reckdenwald_a_Page_102.pro
ab738d754d0ad09df51756489909e2e8
b73dd6564535116352d6c9955f3533ac430de9e2
1186 F20101107_AACDCK reckdenwald_a_Page_002.QC.jpg
ed5c84c0b7952c034343d1a488fcda32
bce0f616e2ee947532063c1aa5c8b53f498ca8ec
58778 F20101107_AACCXE reckdenwald_a_Page_119.pro
590996c4fae4b2451e62a42ceadf250d
6b25e5c3b3f685702cbc0fdb6cc7f0622a8d994f
1086 F20101107_AACDBW reckdenwald_a_Page_118.txt
579e70f8e7906491b2a92084d130e22b
dcdd53e4dbe71ea92d664c874be4a622ba3e66a4
41219 F20101107_AACCWQ reckdenwald_a_Page_103.pro
4dfc61f48d8e81182fe8e939825b4a14
10f747c89323f4e178d440c33e3b7cf4faec6e8a
548 F20101107_AACDCL reckdenwald_a_Page_002thm.jpg
f14d0590988443e512b113d0499bf280
a46ddc100d46e29421f07f038340c2c82f6b9aa9
62005 F20101107_AACCXF reckdenwald_a_Page_120.pro
703f6ab4e61b1de3c08eaef6b68af8fd
c21e745a2c0eb01eba45583f01706e095f26419b
2581 F20101107_AACDBX reckdenwald_a_Page_120.txt
d140df8d4519951562f87ee8df511340
7cc888ddf6aab22980601d1f6a46d9b404bd8fc0
54125 F20101107_AACCWR reckdenwald_a_Page_105.pro
2e02b0fba0e8a20decbcc93a60136427
d6c9ed3d03c225b412121624659997604108dffb
6737 F20101107_AACDDA reckdenwald_a_Page_010thm.jpg
eb63893970acc7246fb723c6eae32144
797ce1b4a3b4ca2e3ef9b7c13423483633c42787
61455 F20101107_AACCXG reckdenwald_a_Page_121.pro
1223b981328add005a45d7dad2f0a41c
98b9a9a1eef16c8aa005cf016aa0e62bbb15b03e
2557 F20101107_AACDBY reckdenwald_a_Page_121.txt
3e2961bb3aab5b41a5bc1abd4d83a58b
4b3caf84179a96a036caf3c4ebb280b4c72b09e3
53578 F20101107_AACCWS reckdenwald_a_Page_106.pro
e8cff555975e0fec7f288c0c84f2fcf3
85116d5bfd001b246f45449e64ff24d929b5c27f
34821 F20101107_AACDDB reckdenwald_a_Page_011.QC.jpg
3122359fe0b40d2c18b648afbb81070f
a47367482e42bef7f9a393036c2d920998f6684b
17719 F20101107_AACDCM reckdenwald_a_Page_003.QC.jpg
ac596995db5dbb67c88064e395cf67e7
6a47cf03f42ff023bc621709d98248a8548f9bcc
63128 F20101107_AACCXH reckdenwald_a_Page_122.pro
c243f216c783dd8762c765685784cef9
185a016f6a814d49a5c59f51ee7b6ae90e967946
2628 F20101107_AACDBZ reckdenwald_a_Page_122.txt
c850c84c56f96af6ce4082baf3960b88
9334e6169890990370897e3892974fdb375eb976
53481 F20101107_AACCWT reckdenwald_a_Page_107.pro
2571e944fb6f6bd367fa98386ab85d92
76c941c5a013a7e4ab7cd7dd0730a05406b75305
8439 F20101107_AACDDC reckdenwald_a_Page_011thm.jpg
ac09debe7905d27676e946b76749aa43
58d1b659d20b3474a713f2143c35a1a30fa5080f
25248 F20101107_AACDCN reckdenwald_a_Page_004.QC.jpg
a61721e49293041b334ec043efd4fd94
e5cb9a978392cfec613f95b178c6435e1ba7e7bc
63452 F20101107_AACCXI reckdenwald_a_Page_123.pro
97329e1f8a0226329be4f0f96ddb7661
3b5703935258d35700a51d6a63e40d28177c3925
26796 F20101107_AACCWU reckdenwald_a_Page_108.pro
16f3df62bab233780a592dc42e0aa9cf
719e7edfa906b3e8bdf8591f6070fb4e647c7f18
37303 F20101107_AACDDD reckdenwald_a_Page_012.QC.jpg
d303f393e0ee9dbb2ed9c449b7f9a7a1
64fc6d8c33159257002046b134858d144950e1cd
6341 F20101107_AACDCO reckdenwald_a_Page_004thm.jpg
5e99e46630fa6e044a2e74ec2a8a6952
61b05401f9fe140ef9d2709d22375f5ce6cc55c6
61732 F20101107_AACCXJ reckdenwald_a_Page_125.pro
a98cea87fff0f869cc1ceca2c2f91c95
46dbce71a53b58b5b416a8817c147fa849c7ec34
F20101107_AACCWV reckdenwald_a_Page_109.pro
b7419618433c79d28bef2974605a7add
aad94f9b6718ac172b7e0858ce4947730b39f000
9172 F20101107_AACDDE reckdenwald_a_Page_012thm.jpg
e7cac46e9f7722c4c7f2cba666ace47f
0125086fc9129d6ef34f59b98de9d8bb5aedc04b
31176 F20101107_AACDCP reckdenwald_a_Page_005.QC.jpg
e5b3a0f2def883c2fc5e928baf3183d0
1cb64a84497ccd074670d78c99d870d26243002f
54251 F20101107_AACCXK reckdenwald_a_Page_126.pro
3f976454a5115a54534af73d51c15346
15e8d4921ab4be68b46038312fda48665b1537d8
65008 F20101107_AACCWW reckdenwald_a_Page_110.pro
78f868148123bbd38909a3b31acc27d8
0e3597935bd6350f73eaa239c618774268cbdc03
35075 F20101107_AACDDF reckdenwald_a_Page_013.QC.jpg
ba7c4a317028a918955aaf51113eee95
92a08064721e8ca19c072336773906da75ba611f
7312 F20101107_AACDCQ reckdenwald_a_Page_005thm.jpg
5cbab219a0e8f3c95028db90ebc90c74
993572da6181e7f7785e45fae41be89d383a6d94
65179 F20101107_AACCXL reckdenwald_a_Page_127.pro
9ac7b4f974233a50be6f6d520584f783
9cb8aab84ace91fd0bbd941909cc228fc8f7f045
2205 F20101107_AACCWX reckdenwald_a_Page_111.pro
7269e79f599bd7bee27e8621dd5d64ae
7c0fc938f851225858e6f22d9d9fd03d499d02c8
8796 F20101107_AACDDG reckdenwald_a_Page_013thm.jpg
8be45bfc56db0948d09a4588e7a85b55
9001676571b35d6f4c442e42bd857b1c20a3ae30
2056 F20101107_AACCYA reckdenwald_a_Page_016.txt
cff470e536d2b9b9c70f0d5ec0ecc6a2
08b697bf636cbf129ba88d1ad6d857b4373314ae
5251 F20101107_AACDCR reckdenwald_a_Page_006.QC.jpg
e39e082d4dd2a82e1e0b43485ba9f54c
c671b463b0cb42a28814ff081110d8c0394dd548
20136 F20101107_AACCXM reckdenwald_a_Page_129.pro
0a1274728a13a1b5da2e0f5455bdede5
055267775a58bd0b37cc08ef7958c200f0040b9f
66897 F20101107_AACCWY reckdenwald_a_Page_112.pro
18990e21cc01edbdef7e2d53ed543fb6
628145528a7586eef8ba650689d425e46904b140
36047 F20101107_AACDDH reckdenwald_a_Page_014.QC.jpg
b7003873c7bf4ef0ca901f14ea71de4a
3dc4cecbb0660cbcfb0ee75249bea0a806424f22
F20101107_AACCYB reckdenwald_a_Page_017.txt
21065c2568de209ad12305e1950f802a
3305514db94e89cd4987e8b8e146f4ded2d54e7e
1601 F20101107_AACDCS reckdenwald_a_Page_006thm.jpg
95ebdf23797c34820f044af21a4ef8f3
52331cd15a425e1751736f311dd97475addcc12e
481 F20101107_AACCXN reckdenwald_a_Page_001.txt
caa8e63fb25f0c2c2f877cba3b0cb15f
093079ddd96a7fb094471d8c7b33754ff1709e54
F20101107_AACCWZ reckdenwald_a_Page_113.pro
9e17639c9e0cfef02c86cf06d156105b
4dca4d54575f070e50846012d942fa1426bf84fc
8506 F20101107_AACDDI reckdenwald_a_Page_014thm.jpg
f40c1995bf6ebeb5b9eba45c5dcef21d
1a2eb16d5d085e4d0f143b55ae5b9578c3e14669
2083 F20101107_AACCYC reckdenwald_a_Page_018.txt
e05912c6118e53ff0e1fa4a8c483866a
6ba12b7b0d2fea5e0e0701efb66d9169fc027dea
22946 F20101107_AACDCT reckdenwald_a_Page_007.QC.jpg
783983c738292a651472a2479696f9f1
98bd17f81b47b854e1ab6bd0076e9c41ef163902
91 F20101107_AACCXO reckdenwald_a_Page_002.txt
622b73271f01e0322978fb546a3c10ef
141c98293465a6b1d8be6f0c9c69ca1bfb342567
36835 F20101107_AACDDJ reckdenwald_a_Page_015.QC.jpg
69efc3368c2202dba95db7c79b1efc1f
179e61746975f42dd13da22d215c6043bf3d8b2b
6006 F20101107_AACDCU reckdenwald_a_Page_007thm.jpg
dcfd5c87acb21debf0c78a96ffbff900
3ae7f5e962a9f0e4b08ce7837155d1452d4feeed
4065 F20101107_AACCXP reckdenwald_a_Page_004.txt
73fd89f96486c70208463e50ef990f3a
3728a4e3cbdb5bf0ef7813dab2f3e5ed8e05e7d0
9094 F20101107_AACDDK reckdenwald_a_Page_015thm.jpg
4b65bce8f82e996416c1474a0edb820b
ad8bcd4ea44ab5218ab7db4f108c6ffc969cdb6b
2134 F20101107_AACCYD reckdenwald_a_Page_019.txt
19f599f6833cda8a68aaa3f74b247ac1
6b7e5808f35b06f47eb44df35974a5a23e62be33
14023 F20101107_AACDCV reckdenwald_a_Page_008.QC.jpg
a28d5bbcfd46d7a1d97d1f452c7bb8e2
b850a952f9c214e41c6a63cf545fa59793ba7d07
4023 F20101107_AACCXQ reckdenwald_a_Page_005.txt
610b5aa95d6bccef39fe69115f40741a
6be2d4d8d75de92c338f84474337b1d50a2fb651
34468 F20101107_AACDDL reckdenwald_a_Page_016.QC.jpg
05e8a1194a43cdd6a8fba973c06bc446
81053e05aa5c39138c73046975d2803df5e76a61
1982 F20101107_AACCYE reckdenwald_a_Page_020.txt
c7780724bd1d2a04b53889b34dae345f
c630da5ecf0ba333a67a1defca622b5c12effe63
3622 F20101107_AACDCW reckdenwald_a_Page_008thm.jpg
c6b57c0b625fa5e5b36f3c9f27558e55
5fe11dda5aad0ce0c0a65609c2caeabc3c962155
523 F20101107_AACCXR reckdenwald_a_Page_006.txt
33b0b6b985284ff4f169705ff20446bd
a51fb2d5cbd47a3f9ad5bc99738c9387f760534c
12074 F20101107_AACDEA reckdenwald_a_Page_025.QC.jpg
08208b1e513392ad1f8409dcd9642e70
17bed067c1fce9d1d5e0ca6b6a8ab467adbe55b5
36520 F20101107_AACDDM reckdenwald_a_Page_017.QC.jpg
c1681bad86eb3c12e055750366945235
25ce5513f7c66eab9cc6a8c4ee1b9face95165ca
2029 F20101107_AACCYF reckdenwald_a_Page_021.txt
e783a024a9ac89ec525b3c9364502c63
c4fbd776fc287df5db322eeda37a7042fd132d2f
30745 F20101107_AACDCX reckdenwald_a_Page_009.QC.jpg
1a929d6cb122c252d810d50d13510841
2cff443dbd081c829681d3f9499416cd0f67e9cd
1808 F20101107_AACCXS reckdenwald_a_Page_007.txt
b5629bd922e676fec24878c9635a409e
496e5ad0fb74029741bfd8e9a92178232a9aad97
3646 F20101107_AACDEB reckdenwald_a_Page_025thm.jpg
fe0205d4b2369b6b5c7cecb42f2ebc6a
7e4b29b48f05f91f6145e3bcd73ddefe28dd2fed
2221 F20101107_AACCYG reckdenwald_a_Page_022.txt
df3a23359bcedd63e5b495a4c77c83b0
345eed5bc0698c0fd20277e53ee3ad3be6c9623c
7567 F20101107_AACDCY reckdenwald_a_Page_009thm.jpg
5ef741ed49d3221cec6b014a48bcffb2
9a83bc853b6a6ad37f05b34ac82d5010d5bb08a1
2057 F20101107_AACCXT reckdenwald_a_Page_009.txt
f89f85e23241645988a34ff4d4f50c32
6d7c0bc27cdfc4e98363cf632317e3da9a7e9e45
9744 F20101107_AACDEC reckdenwald_a_Page_026.QC.jpg
29487aa590e93675cafce5447764d663
e934cfe774dedfcfe20c1b48c365a52c3ebb8300
8717 F20101107_AACDDN reckdenwald_a_Page_017thm.jpg
4aa7db74dfc4b801adeb196dd3ddf567
3c82ccb3f35999f72df68c00bf94b26125b5650b
833 F20101107_AACCYH reckdenwald_a_Page_023.txt
139db3a5b60f944a82bc22d8e546996d
feceb71f381fda1b47328f46ac34c222d8df5f9a
27685 F20101107_AACDCZ reckdenwald_a_Page_010.QC.jpg
58c8a3570be1da0f4b38ed631e082fa0
2509868d22dfdd6e59626497d2dd7a0f8eab7490
1618 F20101107_AACCXU reckdenwald_a_Page_010.txt
8908d860a30b0650934e47a7e5d83ea1
070b46183ead54cd842ff1fb711ad97991352d00
3205 F20101107_AACDED reckdenwald_a_Page_026thm.jpg
bc4fb67af9e7410c15c112a0cd5a1bf0
10ec531e95939f9d0e706d8b24954e885487a154
8816 F20101107_AACDDO reckdenwald_a_Page_018thm.jpg
d3fba85b33feba8a7a244ddef52c666e
0d34e05fe5d3efa856b77bad3399503ae24ed0fd
712 F20101107_AACCYI reckdenwald_a_Page_024.txt
feed3f703f3917e736587a898aa17a49
996dce1e03def11c8a2c835055acea5303dede96
2662 F20101107_AACCXV reckdenwald_a_Page_011.txt
179aa4977a58bc3596eaf77eef696338
3f30f63f6c084e4258abf0ae4c41fdb63e714556
F20101107_AACDEE reckdenwald_a_Page_027.QC.jpg
56cd0242c18cc6f10b8c162354484e72
ff59d307f8bbf38227d936e0fe56190060f0dc51
35017 F20101107_AACDDP reckdenwald_a_Page_019.QC.jpg
8c52c946ec8dd508e51f80c25284f34c
be12f71c8214770c3302ec2a2eaed5487fdbfd17
811 F20101107_AACCYJ reckdenwald_a_Page_025.txt
e46e6b1906f71747841f3e148c93b5fa
768e15362324743fa881e3c55fe7243f55da03a3
2273 F20101107_AACCXW reckdenwald_a_Page_012.txt
4106dd0d4aef71bfa4fbecdfef48b3e4
4967caca2948f24879d23e7d4c13617935735447
2701 F20101107_AACDEF reckdenwald_a_Page_027thm.jpg
a9a73829509ff184b2f23384d5ee38e9
98f9bae78c72e33379717d72340ba9ef2f0a90d4
8894 F20101107_AACDDQ reckdenwald_a_Page_019thm.jpg
eed069702de23bc3cbafc8abb04cfad9
82a1d5142299783aea54c38f37328ecc0db53393
487 F20101107_AACCYK reckdenwald_a_Page_026.txt
078c1273561cfc3a28c3253d6d0dd772
99f70c60f12caeb46743cf91641d0737f8f8605c
2149 F20101107_AACCXX reckdenwald_a_Page_013.txt
a30c9d7d89290c296a3fbb5ccfce9993
9f6312f5ac177bb647151772b950743d976e2acb
34673 F20101107_AACDEG reckdenwald_a_Page_028.QC.jpg
3c84dc8022b575a11794b8dbcd26fc86
4e1200333ddbbe67c027cae890d28fdcd8ece134
2080 F20101107_AACCZA reckdenwald_a_Page_042.txt
d4e9059d6dff15a0fa6635c221dc6daf
a271ea0854bbe58d277842bba132e945c6d115fe
31004 F20101107_AACDDR reckdenwald_a_Page_020.QC.jpg
fcf3381d35383db5cc3e25327fd15c6d
d4866bd1df9df3d2f8c1f56f596a4acdab1734b9
554 F20101107_AACCYL reckdenwald_a_Page_027.txt
50c71522c8d9d19d1d609849a721d550
da880d8f322d0d497af1845a26bab32b3c6a1683
2136 F20101107_AACCXY reckdenwald_a_Page_014.txt
c6ac4c075fac7716d54fdee75b3a822d
8754d813b80392d5d5f24f26d04704cdf17f2f14
8600 F20101107_AACDEH reckdenwald_a_Page_028thm.jpg
3ab4ad75550d439243ea0d47cd9a67b8
2dba0261c905401b3f56193c5bdfebd4ee2520fa
2154 F20101107_AACCZB reckdenwald_a_Page_043.txt
8b0b20c60e322de9749dad885006f927
e857e5147acae11d1f9f615ae62da75d0fb6131f
7721 F20101107_AACDDS reckdenwald_a_Page_020thm.jpg
3be2d2747afc64566c0ed7e4ef446740
0e6fa690b4964ad0d00c5f9493a1a3ebfac5c103
2165 F20101107_AACCYM reckdenwald_a_Page_028.txt
17b2c8757c7ed9fe0369eb1c677d6b6b
10d099c4ffa423427afb3e26b9d0080fdd26e1b8
2242 F20101107_AACCXZ reckdenwald_a_Page_015.txt
e58d11bac60273ed726d19b75a112ce2
765df8fa651fe3313dce56b8813fac4bac18790e
33371 F20101107_AACDEI reckdenwald_a_Page_029.QC.jpg
2d99477e945712f7e2f5404faeae98ca
ec11a2ab5f5c8e4716060a6c4ef179053ffb7822
2036 F20101107_AACCZC reckdenwald_a_Page_044.txt
88515e483eb61ba6352062d4d21fc6e3
fb7fc1128055f5c400aeca8caa9e322f7cf7fdbe
33542 F20101107_AACDDT reckdenwald_a_Page_021.QC.jpg
cca1a7e60c48d095e3cce299ba330971
f586308c543ff40e4d9f597bff2b5e21ad7be396
2023 F20101107_AACCYN reckdenwald_a_Page_029.txt
019f78594bd11cfb22cda4f226c745f2
92e94dfc4dcb70c5d4dd52d69d42d60df6b47ebb
8235 F20101107_AACDEJ reckdenwald_a_Page_029thm.jpg
8643016548390ce7691f799201a93a83
0b6d5ef316fcdd41b0882265fd7d61df52964d33
2046 F20101107_AACCZD reckdenwald_a_Page_045.txt
6e9134d8d9733934869cad98890af707
86b69a905c7807ebe48e5be000c6ce263033bbbe
8557 F20101107_AACDDU reckdenwald_a_Page_021thm.jpg
98fa0f2f6b960adfeb3cb5bf767a17f1
02c19c4f8c8aaf77eedf270414d1cb7240fd7e2a
F20101107_AACCYO reckdenwald_a_Page_030.txt
9c3cc9213fa2bdda3033dec8f37f0974
7e0b2ba743aeb3ae08d744f1a31b343c3a209c4e
35288 F20101107_AACDEK reckdenwald_a_Page_030.QC.jpg
bd9aed365f0877fcb80d35fb02d4bb58
37d25b1d7cd8c978c00cb2f1197bfb1c0e37c993
37002 F20101107_AACDDV reckdenwald_a_Page_022.QC.jpg
a3750b9992604881102d3a5193fe0e3a
977dd763af5a5132eed63891e4a0cbb08c82e67d
2066 F20101107_AACCYP reckdenwald_a_Page_031.txt
164db28c62494e1b1b9b7b980e72cde8
f809dd32889bf2bcd4db58376ac6c56f9ff96de9
8454 F20101107_AACDEL reckdenwald_a_Page_030thm.jpg
7bdd5b2bbba4fc3c26080e2a2b1e1638
6d3aba4b3e165dec2a15f36f84007e320993746c
F20101107_AACCZE reckdenwald_a_Page_046.txt
508715fb27c5c0dc591ca1d0f2e19e86
b5592926ff4910856d8d75fbf0ab16f40dfe8f5a
14325 F20101107_AACDDW reckdenwald_a_Page_023.QC.jpg
75479130611fa4c8691b9a955bb13996
67ace51aa18d2eaef5a70d9846790f34c3e02a48
2140 F20101107_AACCYQ reckdenwald_a_Page_032.txt
3624660e272d4643d1fec5facbf894d4
5b31419501e62e9c218f6639b3c0e1d7b70b5280
34540 F20101107_AACDEM reckdenwald_a_Page_031.QC.jpg
29436e8a96c8c9f3e8cbe556eec79451
241aacf9e43eb704204d89882bb4f4199f3d65c7
1439 F20101107_AACCZF reckdenwald_a_Page_047.txt
a78ca53bbb11e9ee149fe525dffccc49
d53a4687fcf026f10d5014c195ce79ec2eb8802f
3750 F20101107_AACDDX reckdenwald_a_Page_023thm.jpg
e49c9de4543626800ebbf2cf1e8f06e8
fd11786df52877c1c453cbac98f2370ad2fb5a79
2224 F20101107_AACCYR reckdenwald_a_Page_033.txt
7d6902a18cb8ed6c4aa41ed73fb5a0d0
d5b702a4fb77235d544613dcfc2f25e608171c61
35743 F20101107_AACDFA reckdenwald_a_Page_038.QC.jpg
b788a4d57150cf728e12e3069427cd45
62626b67c2d10eb10993e1c5da9c17374f2c3dcd
8868 F20101107_AACDEN reckdenwald_a_Page_031thm.jpg
dd1e07f95d5c311934e72c58616bf3c9
1bba41d4870b46106319641ac63acda353fff5c1
716 F20101107_AACCZG reckdenwald_a_Page_048.txt
0c20d156cd32345496aa4c835b2efa2c
bf28317fb2f9e0facab5b490e9112f06afb94633
8157 F20101107_AACDDY reckdenwald_a_Page_024.QC.jpg
3b9db31c30df454da0d9b4b0f3c619e0
e8cbc32496721a98f0c492587d3ed4a3791961fa
2248 F20101107_AACCYS reckdenwald_a_Page_034.txt
fffb5525c441dfa83d5d2c37fffcf875
d0e1f77b5685a681bf2fae36bc4c44a832a83699
8725 F20101107_AACDFB reckdenwald_a_Page_038thm.jpg
8ec060184d2033f33d7a4d558b352e39
567482eff578a2408b6f0f90a87b19621d9dae72
F20101107_AACCZH reckdenwald_a_Page_049.txt
8c07c21dd7dd7114876b92ec48a2124b
7bbb6904d898f5a28f83348148861a8341634541
2789 F20101107_AACDDZ reckdenwald_a_Page_024thm.jpg
98d051cc368287ede34a17d208f0f7d7
290649a53feb56d5047e2ee5e62b758361da48d3
2148 F20101107_AACCYT reckdenwald_a_Page_035.txt
8f5aba46bfe4797d9e5815282af50efa
c360dc95aeec73312f21738b904fd12865636b2a
36021 F20101107_AACDFC reckdenwald_a_Page_039.QC.jpg
75d559f2b2830395384ec7f21481485b
23d7115e287e18f061d0fc441c0cc8d775c8524b
36422 F20101107_AACDEO reckdenwald_a_Page_032.QC.jpg
a572a3dba75ce7160c279ef35613b319
985ed612d14435f94a675c74c64517c354afa03c
2096 F20101107_AACCZI reckdenwald_a_Page_050.txt
132607876085ac8e94fee39e26adf27d
6b329a00d17c542ef574b3deca21d699925bb3b6
2013 F20101107_AACCYU reckdenwald_a_Page_036.txt
c23db2f2c31fe676f41ae9104647b399
e992a6446d88c170267dfeab3f0346c5f6a90c24
9042 F20101107_AACDFD reckdenwald_a_Page_039thm.jpg
39e7508801065bafb3b0eca309b3bdbf
cf8ab5b64eb0020f0bf2611d3f3ef4e454d15212
9008 F20101107_AACDEP reckdenwald_a_Page_032thm.jpg
a6ac8f47c8ad20394081521b3b6d9807
65b0bbac9cf5e4edc528f49b6868280153bef42c
1886 F20101107_AACCZJ reckdenwald_a_Page_051.txt
84025675754ae47945658b5a354d440b
aa4adb62a24197a006c7a8a9b759e78809499e87
2243 F20101107_AACCYV reckdenwald_a_Page_037.txt
3be53c9813d2b754593264b0631c49e3
cff12273242f91adc041d2a52c1a3a7695b515c0
32860 F20101107_AACDFE reckdenwald_a_Page_040.QC.jpg
e771b5db563914a5dafc17ae9ba1442b
32e1c2dac5b503f3634b3b840a13bcf16621a98c
37360 F20101107_AACDEQ reckdenwald_a_Page_033.QC.jpg
f54e7756a3d87e2cd726aeee8f50e22b
d65a25afc449717cc49e7db990ba34a6c110a264
2000 F20101107_AACCZK reckdenwald_a_Page_052.txt
abd374432d823feb23fde5e0203789cf
1264c50e95ba162af1ed2f86ebd434a06e32ad8d
2163 F20101107_AACCYW reckdenwald_a_Page_038.txt
730e9a7c0468c1be1b27d84e99ea2762
f3e93b09942db798f71dc3c89b22a5c67abdfc05
8012 F20101107_AACDFF reckdenwald_a_Page_040thm.jpg
6dc3e52580b9a9d71afd85e76142f8f8
d71bf443ec24d616caa1becf7dceee0fdde17997
9478 F20101107_AACDER reckdenwald_a_Page_033thm.jpg
1a02dab77a4f9e7e2ddfaa64580e7af1
5f9d6a8575d7b67c45d450029893fbcc09594d05
1903 F20101107_AACCZL reckdenwald_a_Page_053.txt
2812ccf5ad8aeaa7fee13e4e0a0f391c
9ca76d4b25c37aeb32a896509557b5c5a126507e
F20101107_AACCYX reckdenwald_a_Page_039.txt
7dea13ec679d86f50ca43555c69c2a9e
bdd5d2ff7dfa60c1d79299842c1765644eee6d73
33918 F20101107_AACDFG reckdenwald_a_Page_041.QC.jpg
e139aadf840f42072194e8f008b0e2ac
a4760fb5349755308a711a18f16ea0a4fd474306
36599 F20101107_AACDES reckdenwald_a_Page_034.QC.jpg
cd055ab1073234bdc81b8c3c3e921e40
176ed0cbb75a881a3df0ab93714ef396d1caffe9
2204 F20101107_AACCZM reckdenwald_a_Page_054.txt
ae9f68f0bea84413bc7bcb2c7ebbc08b
735a797447210754ebfc93614e56983e10f97c0a
1874 F20101107_AACCYY reckdenwald_a_Page_040.txt
f9e1924ba3b92a4ce1316d1b198437c4
35667a0fa6403e68bbe31d7683c4eef2aa015b7c
8589 F20101107_AACDFH reckdenwald_a_Page_041thm.jpg
40d1b31263fddcb1c46cb6206bff63ea
2aa253f217680e6ce7b1c9b3910c87c7beb279fa
8892 F20101107_AACDET reckdenwald_a_Page_034thm.jpg
c9a046d8f677ada6b841cddf7fd691bb
9de0d1cd44f3526b446bd4a4f419f2587131706a
1964 F20101107_AACCZN reckdenwald_a_Page_055.txt
a2fd456167b878d80d7464e7537e7ba7
0da948329df27b04055c8e58973ecbc1f177babc
2206 F20101107_AACCYZ reckdenwald_a_Page_041.txt
0691a05936739a7ed004a6b7aec302eb
bd3b75bf3fb2d970ca7d6d5ae8acca6aad1a8341
34621 F20101107_AACDFI reckdenwald_a_Page_042.QC.jpg
07c4eddabf682f16ec83eb51c7b94492
81e36a5282b3b8caa6b5fa2efc18787486e98756
35227 F20101107_AACDEU reckdenwald_a_Page_035.QC.jpg
8bf12fd8497c714c8ffd3b6462f9eae7
18ef09f55bae24c7c1818d51f23627930062d2bc
F20101107_AACCZO reckdenwald_a_Page_057.txt
39e497b5deca7a5768666ea4485ed72e
ef5f603ff3a79f651ba6356a5c555bd36cdd03d0
8490 F20101107_AACDFJ reckdenwald_a_Page_042thm.jpg
23c73f6756afcb4e6cb31a364ae6cdb0
c9af3f22ffa5919587a8b98c1716bf202b2c41c7
8719 F20101107_AACDEV reckdenwald_a_Page_035thm.jpg
0d5080b9794fc3204b55b1a7cea2674c
4eab6d0b19b8f47ec1da1dec5ee7e2261fccccbe
F20101107_AACCZP reckdenwald_a_Page_058.txt
25f4cbd8792c9c3519f9ac45d34f3700
2d1e3ec4c6a673aaf1e326a1378005ca588cd43a
36572 F20101107_AACDFK reckdenwald_a_Page_043.QC.jpg
4ec8d90f4f45a51fc72f62ec5b643266
41eba1db53201edc605ed399ab8fcd9a8b4b2db6
33733 F20101107_AACDEW reckdenwald_a_Page_036.QC.jpg
4e875a17b9cbba55d656c94ccdfb3cbf
b009b95478cb73613b960e1d5d52ab9b67361ee1
F20101107_AACCZQ reckdenwald_a_Page_059.txt
e89b057c764f3c37bb7ca71b50d509e4
f97b4676ad65a6472e3d0c93b2867eb8d11a7832
9061 F20101107_AACDFL reckdenwald_a_Page_043thm.jpg
f319ff386eaf96f3e4bf2f367dc8214e
58cb200383c7785d07b8fba785d0508238e64da7
8627 F20101107_AACDEX reckdenwald_a_Page_036thm.jpg
8c5fe05ccd2452c4a46cf640b4947fbd
ed7c150d86a543e55788c42fd63b873824ef3fd7
2050 F20101107_AACCZR reckdenwald_a_Page_060.txt
8eafd0ae08e0000f8316460289e3d870
5cebf1e57b403f63ef2bd3c62e5f0c4f31070c06
32005 F20101107_AACDGA reckdenwald_a_Page_051.QC.jpg
d140077734f415d1b23bbfdc341af98c
13a854dc998292b2a9d174ca9259c5c9e4decde4
34226 F20101107_AACDFM reckdenwald_a_Page_044.QC.jpg
fc41c38f9cf9c712c99434470b3138bd
907998cb134110c845b688522e7fca189310dabc
37846 F20101107_AACDEY reckdenwald_a_Page_037.QC.jpg
f7544059d2542995cb6d113a791bde4e
8f8c81196720620fae85580c16e46c37146c9964
2157 F20101107_AACCZS reckdenwald_a_Page_062.txt
3e7c0a237d88d50f643055ea45c656b3
b2b5c4720041eed498244f0847793c7124b44181
7864 F20101107_AACDGB reckdenwald_a_Page_051thm.jpg
128b39d68f947c6f936e0c2140ee6749
2abb373f55e133836f81553191dc6eb49e242999
8241 F20101107_AACDFN reckdenwald_a_Page_044thm.jpg
40bbf775257c2c7f737248a91e29e882
5db31dbb4772078155c87c708ea70e4c69c3e6c7
9066 F20101107_AACDEZ reckdenwald_a_Page_037thm.jpg
14dd280ee91efbffce1e7a74048a9089
cad809a15a0b3d1771abe87bed12f3aaa43f64e1
2618 F20101107_AACCZT reckdenwald_a_Page_063.txt
3a9479290e0b1822be9aca730c4b97ba
3c6ed779e4462816876f124cf9d1e82e22d80aa3
F20101107_AACCDA reckdenwald_a_Page_029.jp2
d128e0ff6e71f81808876816f4d47400
3c923c4a8b2cc047ccf6af0342d24843b1ce5f3b
34704 F20101107_AACDGC reckdenwald_a_Page_052.QC.jpg
5ffb79c45425d3725684a71becc363d9
e7074a807b43054843e178c5e42547f74f87bea3
34494 F20101107_AACDFO reckdenwald_a_Page_045.QC.jpg
653d9964be1464f0ca156f2918de0f0e
e4c7e5c9d22a584745963fe1ed144760de6bc57e
1988 F20101107_AACCZU reckdenwald_a_Page_064.txt
f7445ec35f71278a0cf6aa0a3e8d7861
2bca2ecdc8854ffb2d37c15cdd20e7931ef2ef49
26621 F20101107_AACCDB reckdenwald_a_Page_001.jpg
5907c85edd295cb65156453567c4e7c7
7516652c852790df643e7ab1371df376e84a809f
F20101107_AACDGD reckdenwald_a_Page_052thm.jpg
661c4c71c785b5c883cca2f81cab5772
16c6cdca2f16f87c3502373198c3c9fb0276e30a
2045 F20101107_AACCZV reckdenwald_a_Page_065.txt
5645ee6274c93902866ee6b2428bc8e3
3c6f60df7af72e42b7cf54bdfe884f935658ed95
473556 F20101107_AACCDC reckdenwald_a_Page_023.jp2
3c8902745b63321c3db148eed1b567df
61cdb1835df27a85e8499cab8134d053583fae70
30957 F20101107_AACDGE reckdenwald_a_Page_053.QC.jpg
5f84f68d45485ed19b16cb761823d8bf
2f621268978c9be8386351696e515260f97dc53e
8092 F20101107_AACDFP reckdenwald_a_Page_045thm.jpg
2f58c7f247aae88d981c2249dfe3c8bb
5738fd3510a897a3e4233f79dc4029458ecf60be
F20101107_AACCZW reckdenwald_a_Page_066.txt
472f0453e86340ad17e7e069d635d694
ddc4a928eee7f29b590293f5184861cb25a9c173
2358 F20101107_AACCDD reckdenwald_a_Page_073thm.jpg
ae1312497faa1fe76d6f2e26e33e6115
bc4cf6426758e5abd3306fd0e02b95143ff9c360
8038 F20101107_AACDGF reckdenwald_a_Page_053thm.jpg
848c08cf4af239752230a377a6c980d7
a94132a2c342cfb927c3b830e6936b033daa267c
33911 F20101107_AACDFQ reckdenwald_a_Page_046.QC.jpg
1c6dbd0ae92e74a8d5ec56825ffe678e
daef403590695913bcbb895acbd5f4902c397559
2170 F20101107_AACCZX reckdenwald_a_Page_067.txt
c0679a019e2a7c645e5de79397af61d3
0e171338c8c4899184f951bcbe4a3a6f98774ad5
387699 F20101107_AACCDE reckdenwald_a_Page_025.jp2
1f38979cc8737f1f3d9ac72821b7276b
038a0f8b1542b882db500b0f583c830f5e31509a
33276 F20101107_AACDGG reckdenwald_a_Page_054.QC.jpg
394f5733aa500b2e7ab6609b22a149c0
1f76c3c3b9871e51f804c3b95768f588609346ce
8088 F20101107_AACDFR reckdenwald_a_Page_046thm.jpg
aaa8d7e487ecb92e7d66d872afb57636
732b600f75f285e0e2ffef9274168307c2fbb4b4
1637 F20101107_AACCZY reckdenwald_a_Page_068.txt
992b7b3e15cbc5aae607bcb427f2526b
c232a5ce106bcbaec9e32837466f6f5549a352fc
2454 F20101107_AACCDF reckdenwald_a_Page_119.txt
8884746792189b91d5804034d55161de
4684733ce037685f29dd7ff92003248f111c6c9c
8097 F20101107_AACDGH reckdenwald_a_Page_054thm.jpg
f517b040a7a4e47a3435033a7dc67c68
b8b7f617105d1ea36d46a63b5258cb1dfae5ee32
24003 F20101107_AACDFS reckdenwald_a_Page_047.QC.jpg
fdd8fe059b98a67ccd5145f11d552764
6f4d9e6f3e9f21cdb3d60ee98dcf8b914124187a
1479 F20101107_AACCZZ reckdenwald_a_Page_069.txt
1bedaa74489eb020b9ea540b47bd2b36
f8c9e5498b3e503596373d0f5cedd4902fc1e5c2
F20101107_AACCDG reckdenwald_a_Page_109.tif
dc0dac53599769e1a9454a452c89eb9e
01d30e8d3bd9727fb8e3a0353cb61dac75c03780
31591 F20101107_AACDGI reckdenwald_a_Page_055.QC.jpg
45aeeb098f3ca48f9f6f94a31cc3420a
1ce32f671b42e9375ca1990f4f04e62820f6ce16
6063 F20101107_AACDFT reckdenwald_a_Page_047thm.jpg
3429e77c03da68e6a64ce66d0cd492b8
af1639586d5897c35a7439ebf96f3cb2b023ec5a
52170 F20101107_AACCDH reckdenwald_a_Page_104.pro
130d8906b70d3a5dcafd0af9b6d3515c
64734a16f0b318bb379edace2c089cc24368a32c
7857 F20101107_AACDGJ reckdenwald_a_Page_055thm.jpg
731660fc550d024fc1d720bc82d962c6
9505c2f8dad81f6f215ceea2e57783fed60f0efb
11106 F20101107_AACDFU reckdenwald_a_Page_048.QC.jpg
0fd4f0f6aeb7186b3c25e089c55e372c
2dcd6c0563f15b1979d37fe3b616ad269fdd70f8
107390 F20101107_AACCDI reckdenwald_a_Page_014.jpg
a0b6e77891543e5d79b0d5c8afcd7ce8
e6d1e9be8679ae71a97939c8df58695d555f8a01
32834 F20101107_AACDGK reckdenwald_a_Page_056.QC.jpg
85702695e76aa13f6ec8878b03c74b33
ea568ec8e8d8cce2ad913540f44e0ae7c7e22e25
3316 F20101107_AACDFV reckdenwald_a_Page_048thm.jpg
e005ff0a4b4a3adef98ecf8f7218ccdf
51a630575b27a5c793e3d92dfdc4e0f14080b711
F20101107_AACCDJ reckdenwald_a_Page_057.tif
54953afb19ecea4750d5ce421a21bddd
981b5041070959321b93e728fe81b20d3ae60e29
8222 F20101107_AACDGL reckdenwald_a_Page_056thm.jpg
ccbf6dd5642be8493890e4ec740e31dc
d6b086dec0e74c3fc70e286f5830bfd5840138bc
32811 F20101107_AACDFW reckdenwald_a_Page_049.QC.jpg
09852ff9dfd32e50cd37867393ff762c
4b6727e19f5a81a71cc96346fbb17b005f101705
33711 F20101107_AACDHA reckdenwald_a_Page_064.QC.jpg
800039a7a611ff1ad6ba1fc6d42d7adc
6749163660bcc9cfb99da66aa43b725e542c90c3
8682 F20101107_AACCDK reckdenwald_a_Page_076thm.jpg
37e04d01077094342df647ae898198f2
e76bbc5c84a6a2a61c146861831e972117f48f6f
31674 F20101107_AACDGM reckdenwald_a_Page_057.QC.jpg
d589a3e1bff457ec2eccec2191564f12
2cc49490742b444af13637bf949c575e49356445
8203 F20101107_AACDFX reckdenwald_a_Page_049thm.jpg
a83ca4f2442bb84deffeea2f9f09a4ce
8d3abde9f00d07f76a14476fe6a5ac563d568d8f
8477 F20101107_AACDHB reckdenwald_a_Page_064thm.jpg
87225aa509fe4fc99dccc26038d4608b
6358edb48054f99f41037fc68cab6262806f4404
19431 F20101107_AACCDL reckdenwald_a_Page_071.QC.jpg
c6d16e4892feeb88413b10bb0cacd462
310c647dae99e0363b67de6fb439b65e844e03bb
8111 F20101107_AACDGN reckdenwald_a_Page_057thm.jpg
6837d1c336adc32736e0324b5ccb3224
cf7430ee962509e37ff6a9e46298f24ed80198ab
32819 F20101107_AACDFY reckdenwald_a_Page_050.QC.jpg
39a353029fe4241c8fab1b4900890c62
34544ac7a1499c845265a96901ac1c9ed25d349d
13154 F20101107_AACCDM reckdenwald_a_Page_118.QC.jpg
07177b38c99ae5dbc3d8e838d1d2fa2d
1c55e3669d2380b25e0069e64c6a92d0b9cb0841
33721 F20101107_AACDGO reckdenwald_a_Page_058.QC.jpg
5e06551342ac29e141665e6c75371af7
4ff8ffb54cef47c211a7388ecdcb7c811b52ba30
47286 F20101107_AACCCX reckdenwald_a_Page_040.pro
76ed0a165da6f4a3adda3225def74fdd
332bd04dace1dd834cd879c332453f200352b8dd
8139 F20101107_AACDFZ reckdenwald_a_Page_050thm.jpg
a02381c95d60b19adbb0eb856fe7c8d5
4ee0b211f065e90e2759220cfcef6652eae8aa29
4716 F20101107_AACCEA reckdenwald_a_Page_003thm.jpg
3b1e456e032f4128159520a7f57f57f6
690c65d97f1c0a431f3d251c6c26fb4efe6759e7
33643 F20101107_AACDHC reckdenwald_a_Page_065.QC.jpg
64564d397685b2a05e6e5fc7cc23482d
fad16aedda6cbcd27073f5901a2211c3acac6614
34137 F20101107_AACDHD reckdenwald_a_Page_066.QC.jpg
b668256a838bd9b2e5f724775d2d8e97
1d3620935eb7fdbde5c3b6538b392e7aaf09e98c
93874 F20101107_AACCDN reckdenwald_a_Page_055.jpg
9b3f1ffd4069811aaccbda34f5837cdf
64ba902bb58abaddb48d829ff7146e1eb16f795d
7957 F20101107_AACDGP reckdenwald_a_Page_058thm.jpg
a4622afbbc322f0d583acdae4e32aea1
1faee2514dad9eebd38bb0731bbf5a7f2481bda7
110585 F20101107_AACCCY reckdenwald_a_Page_096.jpg
5eb9f43a29e5f729fc4048467f7633c3
8ab62d8ed089703178bbdb878fa68f862d464e72
18442 F20101107_AACCEB reckdenwald_a_Page_088.QC.jpg
0c17ff6c40ad73d598465aa80607f372
c0b835d27925d233cd94d31a0acb7e194cd73b4d
F20101107_AACDHE reckdenwald_a_Page_066thm.jpg
e6c9607b152a169ee706e2cbea69ca37
d9d2e5a9846edc3455a4e9d9503693010e47f164
109294 F20101107_AACCCZ reckdenwald_a_Page_013.jpg
57c1548bdfdb5d166e4f225baf0cdfe9
e27b4dbe53119e1d6baa82800cdec723f54efce7
8798 F20101107_AACCEC reckdenwald_a_Page_022thm.jpg
0c5768e2838366cb9ee94c7890f72379
98b294a16b94f8812312720891d7fa2da916cdfc
F20101107_AACDHF reckdenwald_a_Page_067.QC.jpg
cf04f09f3bc8cfb561c2aedd02953ccf
21acc6604786dd8dc80ff04ab0cebbb4c2ecde9b
54654 F20101107_AACCDO reckdenwald_a_Page_017.pro
5ffe1e44a7b9f63a63015f6f8fe95f90
8b37ba0961ea436bc48fa65f0f0690e1aed6d45a
32670 F20101107_AACDGQ reckdenwald_a_Page_059.QC.jpg
78bfaa7642093ab703ce41c5072e1368
4b6e8a02cb25d8d96e432f492f24f30efaac1624
1032 F20101107_AACCED reckdenwald_a_Page_008.txt
d67b022256e829dd4dc9c7af3c484b00
d4e5925e28d44483b239cc28b1fe8e51933df415
8149 F20101107_AACDHG reckdenwald_a_Page_067thm.jpg
0340fd301b3b471813357a42568ca273
76a8001b60d93a7f926a693836d34407ee90cea7
59028 F20101107_AACCDP reckdenwald_a_Page_112.jpg
1825161f7c6f12ffb31b4dea37936fa0
f96385d269f8768738cc31e2273a3a217cfb7ecc
8204 F20101107_AACDGR reckdenwald_a_Page_059thm.jpg
0c3fe880cb720809e329a19385e90a3e
6f44b36e4c7edefff430e0d4ef0b5c377bb1dc05
133683 F20101107_AACCEE reckdenwald_a_Page_120.jpg
74b74c0cf52612cfe69cac6f053ed253
24fd94eff0ca1ca1f446990aa0549cf24f7dc185
27646 F20101107_AACDHH reckdenwald_a_Page_068.QC.jpg
9b901c9874f7bdf2ab60c77509f8f094
c7f7fad7e2426033e0d7ccd71a79e18444d182e8
250730 F20101107_AACCDQ reckdenwald_a_Page_001.jp2
cbb656c87710edce5f10370dcc2ea71d
b643710dc3d3dd52009305350ae80f2653a30360
34410 F20101107_AACDGS reckdenwald_a_Page_060.QC.jpg
14ecc107e5907ff5c16386eacaed2808
e18245c3d2a3015af6a8ffcd238335623f879a18
F20101107_AACCEF reckdenwald_a_Page_100.jp2
5bf3f074d92bd537b8dabfffd12eea7d
e8c40608ce03abc6689da71e96d4e6568d5ac58e
6766 F20101107_AACDHI reckdenwald_a_Page_068thm.jpg
169a4b48cf076621dccb08bae0af1c5e
b7cd898a544c0c9f1e25b04e6d91ff221198bf7f
103434 F20101107_AACCDR reckdenwald_a_Page_044.jpg
981c45e78d3dbbf1bae8e7e9845d60a8
0001310252ad10d461d389073db5d1db2aaceb14
8666 F20101107_AACDGT reckdenwald_a_Page_060thm.jpg
a5f864abbbeb025ed6b79b70de95c686
99958034eff1deaefdb5e2c6752960f24df97329
34841 F20101107_AACCEG reckdenwald_a_Page_095.QC.jpg
394e43b734607279f178d8863235a2e7
d3c14d0e5e6de129db7e70af9d9601948f5f6849
18246 F20101107_AACDHJ reckdenwald_a_Page_069.QC.jpg
f4406e19e313c4eab4ae95939be6c2a4
d39ff35761650596eabf320cdb5019356fd9001c
49021 F20101107_AACCDS reckdenwald_a_Page_075.pro
5ed0436c2f3717108d87fc2c9f96ab7e
7e3996329126c1e78116d9d39757d7225128973c
36554 F20101107_AACDGU reckdenwald_a_Page_061.QC.jpg
feb3a31c6f5aff9bf7d0243f17b1c2f3
fa8d5971ce47a8b34ded24ff5234d424c8c53b90
50836 F20101107_AACCEH reckdenwald_a_Page_036.pro
f9361adad6e99d5dcd99a634f654376c
c8d5f085aabccb68ad8e379c9133ba1c874ff73d
4876 F20101107_AACDHK reckdenwald_a_Page_069thm.jpg
7bd96d8e7427763c789c0e00a73dcf9c
a4d560dec2bdaab487c388634196c9f4de8bc202
F20101107_AACCDT reckdenwald_a_Page_002.tif
fbe18329233c8a8ec1c079c50b3d5621
18e75048fb93d88f85ee3f082add10fef7ae4d3a
8697 F20101107_AACDGV reckdenwald_a_Page_061thm.jpg
978c7e313be98219ab8cb28e1f5d3fa9
474653312aa8b7c50bf8824e7f288eff386e16ab
F20101107_AACCEI reckdenwald_a_Page_074.tif
410d4064f196171cdc9e71828584abde
b1527052f2750bfa7b7c7d4f127ede6b4035e74d
21273 F20101107_AACDHL reckdenwald_a_Page_070.QC.jpg
c32a8762d04f34115d124e854edcb095
61e79c939e34774111be67819b9a8556b0b04508
35292 F20101107_AACCDU reckdenwald_a_Page_096.QC.jpg
467ca481a453828fa7446167152087fa
7121209b1a747b49ea8ad2fc069a9b8a548cc79d
35660 F20101107_AACDGW reckdenwald_a_Page_062.QC.jpg
883892b8bb6e56fcf429dc51ce1d80cd
6149f3fd3334d4230d751b5be909b46de3801153
112056 F20101107_AACCEJ reckdenwald_a_Page_022.jpg
93f40834233b2380de257916fdca1ad0
7bf3c6746b02c5943c0ef1a5931931b91abc926b
35205 F20101107_AACDIA reckdenwald_a_Page_079.QC.jpg
013966a4009f9b5e8539f7e1a09ec66b
3def9290d1f72dd35cedbba3bf02eccfedf77fa8
5534 F20101107_AACDHM reckdenwald_a_Page_070thm.jpg
7acde9d36d7d9ec4460ba18a58aeed4b
6d343ae27f850fffa319fe926b5cdf3948b4812b
290966 F20101107_AACCDV reckdenwald_a_Page_086.jp2
996788e9bbc45272d6cc3816b455e16d
075c382e4b54c46022b205949df04c70735d1031
8875 F20101107_AACDGX reckdenwald_a_Page_062thm.jpg
ee1d65f05d26f4f1a5c9917c89267849
ca28951eaaa0b275a485dbb1be86e14e8457eb70
608377 F20101107_AACCEK reckdenwald_a_Page_108.jp2
411910c2b2bdb62625719dd1adba2b28
105b5cd9403abb8646d317a77cb3e505047cae84
8788 F20101107_AACDIB reckdenwald_a_Page_079thm.jpg
3fa284effe6810244f75c718dedc4fe9
10455997a85b8d3a8c01c7061aaf561f4caee186
5167 F20101107_AACDHN reckdenwald_a_Page_071thm.jpg
700b7917579cb5c5cd61492dbb238faa
02f9ff31cf128c54ff44bdabb3039f722df2b0d1
34877 F20101107_AACCDW reckdenwald_a_Page_018.QC.jpg
3b0c1c1ae88c8f4861e03f5fe9ab03c8
760dc86c90938ef44337ffb33f069ea6315eaa7b
36286 F20101107_AACDGY reckdenwald_a_Page_063.QC.jpg
2ec5d7ad4b133ed1d287fd048e4e1181
6aa1bde5c84d6532c1c926c3227a619a3a94057c
F20101107_AACCEL reckdenwald_a_Page_040.jp2
f1e1ba5155092692b049f8828a466c83
f3300d5ad664de5801f5280eceb1d408486ce029
35268 F20101107_AACDIC reckdenwald_a_Page_080.QC.jpg
ecd305e820b66ec97be9c34c23f91294
3753c46fb6c706b25ef2d5c5439d26bd73fb8986
7903 F20101107_AACDHO reckdenwald_a_Page_072.QC.jpg
5b751142ab4d3cd5a79a4db95826887e
be6d50158209f0db2809d60921de5e971910f8a3
55195 F20101107_AACCDX reckdenwald_a_Page_116.pro
a2f701149fc818483aac5079a34230da
1227457d6774764d414770348d9cf1f57c1df306
8693 F20101107_AACDGZ reckdenwald_a_Page_063thm.jpg
4d06fbdda1ba7e9c7a035cbb6183fbdd
fc98a9ef47eb534c4e2926340438b868155a6c4a
8420 F20101107_AACCEM reckdenwald_a_Page_065thm.jpg
1dcbeb218c6632a38132c1f0450fdf99
08c248b20ebf2995f7ca18afb8fe4048aee6f010
8814 F20101107_AACDID reckdenwald_a_Page_080thm.jpg
865983f63e5d9377dd8c93743c647c00
36a5c4fd760f80eb8e2e3d77e8d1fdeaa18b9bc0
2359 F20101107_AACDHP reckdenwald_a_Page_072thm.jpg
a7c702e7d2084f17d32220c7ad1eb832
54c498cc03cc0fa2ea404b9944a32c80a46a1046
496881 F20101107_AACCDY reckdenwald_a_Page_117.jp2
07eba5d2db9db24a4126b81016f4ddde
61cdc812769a4230bb1228842360aa8af452c480
4103 F20101107_AACCFB reckdenwald_a_Page_002.jpg
ff43a3f8b3c242b38da95c645d8fffee
69cc8560e7f1ba7222b25a3727fdaec018f71600
41807 F20101107_AACCEN reckdenwald_a_Page_128.pro
35625c55dc8d6ef4d7271be698746917
ca3faa84763c913f6a4936f1163c827fe6b3adb1
35664 F20101107_AACDIE reckdenwald_a_Page_081.QC.jpg
813b4b262dd67f67b2f50291bfd844b2
bc7eef379d894c7541fb74c96729aaa0ec9f7adc
7927 F20101107_AACDHQ reckdenwald_a_Page_073.QC.jpg
340e0c268ecec7688ad831a941622d49
3c6a0464367ef768db02bde0c8121d841ac21b15
57389 F20101107_AACCDZ reckdenwald_a_Page_034.pro
bed4e6caf91fa73e853dfaa586d65112
d7fa069f7dcdbba23beca6e937f33d8b051a86ec
54889 F20101107_AACCFC reckdenwald_a_Page_003.jpg
36d8d4aa20dda2aa5c7fd69fbe5d30c0
8deea5d5b268dc91159d6ad5be08ab6aebf06648
40911 F20101107_AACCEO reckdenwald_a_Page_010.pro
df9b2d834dc0db1e7829dce584b8c548
f3c13320d641ed6811855f054ee8149930a95134
8925 F20101107_AACDIF reckdenwald_a_Page_081thm.jpg
ca7fbce457923f6239877ec9c21142c4
a653f85d27fe7810ed793f84ce39ef7c1d5d3460
117777 F20101107_AACCFD reckdenwald_a_Page_004.jpg
2d69ac52a1db1a9b6bc9073aa7bb1271
6bef76313dd40c8e5bc07bbd3090a1435036b739
35230 F20101107_AACDIG reckdenwald_a_Page_082.QC.jpg
3f318c50bb3fa5601d84a44d0dbe79dc
6a541f77cf985d61e217d17e3fd516c315d2b6fc
32623 F20101107_AACDHR reckdenwald_a_Page_074.QC.jpg
cccda0ddf3eb35ef6e08bf58a6b53beb
9b1ab9c355dbcad68670a2f9f74c700af0cf2402
134130 F20101107_AACCFE reckdenwald_a_Page_005.jpg
8607fe233d463fcc86781f6fea0d0199
1f73b88a2561f4cf24ccfdaa2e41a481e9a6783d
2074 F20101107_AACCEP reckdenwald_a_Page_056.txt
0e8dc275e702c3d1351b506020b36320
676fb73a5e3037ee85ee4ff815209192da765719
8873 F20101107_AACDIH reckdenwald_a_Page_082thm.jpg
b839eaa42d7a53d0932c0b12ad9188ed
73504d8e54176adf4b56ba04a6225df7704f550f
8152 F20101107_AACDHS reckdenwald_a_Page_074thm.jpg
64c9d6bc0d9fc0f73850b9e6fc8a547e
0e6305836d485feac18a90fb5b90ae5fb3863730
18255 F20101107_AACCFF reckdenwald_a_Page_006.jpg
3f04ef16b22ed9b68f65eada7ed37ff4
097ac15e08a4e96509b8992860b6e41ed8316d48
F20101107_AACCEQ reckdenwald_a_Page_049.tif
f9ec9354bfed93d104f1564b851d95eb
b4fca421dbaef98fcd9cc92d807b867042eab838
3143 F20101107_AACDII reckdenwald_a_Page_083.QC.jpg
0c35aa1be5b42938a554e4f19af95326
ce3f7bf4392ba1b3ff65afd1e400a091c521bd82
32439 F20101107_AACDHT reckdenwald_a_Page_075.QC.jpg
d15b4dd978755db7a108ab0ac96f244c
a6a74869cb1dd144746fb066285c75aa8fba7338
82060 F20101107_AACCFG reckdenwald_a_Page_007.jpg
7950de65ac2e24ba777ce7ea08061df4
314fd1d20d608721ff92465897e4d284cfc8cce2
65402 F20101107_AACCER reckdenwald_a_Page_124.pro
8be05f459b2bb9a7ae170dde63f0c15d
199d3bd89660285dcda9bf4953892ade0db863fc
917 F20101107_AACDIJ reckdenwald_a_Page_083thm.jpg
316bf2b6a36aac6836988ba199f68abf
adb9af3e2fa940de065ce3400e11eafe5f41ee63
7869 F20101107_AACDHU reckdenwald_a_Page_075thm.jpg
0fd78ecd9694bdaa646ee7485f2bd0c6
1378ac5b1b92ba98ce572673f56ef466e35926d3
51850 F20101107_AACCFH reckdenwald_a_Page_008.jpg
e7d6643698a1e1641d32935ef5d76985
fe0d9cc5c3214c1fd381e4d4ea5f2f2de6389e7a
F20101107_AACCES reckdenwald_a_Page_016thm.jpg
05fc1aeea733e11c717a32a5bf7171f2
ac8f2e140ff0919888cc094bbbfc735e9d9337af
18435 F20101107_AACDIK reckdenwald_a_Page_084.QC.jpg
dbf78ce5e29db26dc00e439967c80007
b11b7584854c51a7ae8639e2f4adf8467aefc0ba
36584 F20101107_AACDHV reckdenwald_a_Page_076.QC.jpg
d067ed4743d57d72ab996b8171685881
9247396a8665aeddd40b9d57c0a6d6f130bbb951
98819 F20101107_AACCFI reckdenwald_a_Page_009.jpg
a1d92510e4ae1ace88379f62eef7833a
676488eea616d487c8fc9cfbf22d3b1176b785ea
1071 F20101107_AACCET reckdenwald_a_Page_003.txt
3aa1b1809a9c84e0f502ea9e08d0f6fa
c21ee21a4a01836a746cff6c705418cec4cfe371
5013 F20101107_AACDIL reckdenwald_a_Page_084thm.jpg
63a49630e7b8501a5c4baf742409f01f
9a669cdc3f9bfec45f6bc9f9932ba037759dd668
37714 F20101107_AACDHW reckdenwald_a_Page_077.QC.jpg
f4651f9fd82ac64e311efee9550531a2
15081944f019e1412f11af8081972d092950f091
83301 F20101107_AACCFJ reckdenwald_a_Page_010.jpg
cf3d50c5adaa6e271848cc39e4a6a5d3
8ebfdb73a120c7a7c67edad3ee83ddc8fadf76c2
2211 F20101107_AACCEU reckdenwald_a_Page_061.txt
8f5ed4b3306e3b1a7f4f6b3309551cf4
ca47bd6dd42d2c5f5704dd81339a2fb20d3f05cc
8630 F20101107_AACDJA reckdenwald_a_Page_092thm.jpg
61f94d70f28498bd86978b9aa5844a25
c8523ebd3dc1f7ad0bec06972400318c37caa746
18560 F20101107_AACDIM reckdenwald_a_Page_085.QC.jpg
54934e3adb0d19640f25178bb58f4f76
314c32710ccb5c0aee2d3295c9cc81f9382b932e
8967 F20101107_AACDHX reckdenwald_a_Page_077thm.jpg
5152af7dba77bd3cc45c96828c01a52a
b09fccad29f982310118126ec0ff8ab42f295872
116762 F20101107_AACCFK reckdenwald_a_Page_011.jpg
b841fe1b3d4443bbe799b82d4970c284
cb627b69154fa4f300372b2c01237c93fa048b42
54874 F20101107_AACCEV reckdenwald_a_Page_082.pro
99520fd8b7940ceb068b75a0ad1532f7
33ba6ab1c8de222a3b6eee93d620fdba24cac337
35196 F20101107_AACDJB reckdenwald_a_Page_093.QC.jpg
f0bff6051f974d37cf93d01fb546538b
f6cf69ab44730cb92648b8b1c4af268fc088985d
5045 F20101107_AACDIN reckdenwald_a_Page_085thm.jpg
7ab808d013c5b3d6459d05269c8a9709
2782a2c1dce20c7ceb5d6267f1896a588a7b0c18
36486 F20101107_AACDHY reckdenwald_a_Page_078.QC.jpg
c0968b90011d1b9c3686ef47d5d13caf
25afb2a450f56ca01a3fdf3c39834cfd73efe114
115529 F20101107_AACCFL reckdenwald_a_Page_012.jpg
1f775389c2e875e066096cfadb428d0a
84a443f819aba35ac2655089d696ddf14fcc1c0a
556304 F20101107_AACCEW reckdenwald_a_Page_071.jp2
62a50a261bec6b6630ba82ec952fea9c
d429e7d9c22af5a8f7a55e06ef4b6010fa879634
8737 F20101107_AACDJC reckdenwald_a_Page_093thm.jpg
510caa5e239b3d92390d434455a2f28d
8c3193bfc143bc7b90b582d462b2aecde6ba3630
9675 F20101107_AACDIO reckdenwald_a_Page_086.QC.jpg
f166ce02d7b1363cf1c7d08425fb242a
fca3cfa9e0aa53e39bfe146a8ea6e441c3051a92
9191 F20101107_AACDHZ reckdenwald_a_Page_078thm.jpg
ecdb8ea646af6616d2fb7c3b3c682f17
fa0e76cfbb4a7e7288a75e3fec8437b945d556c7
104053 F20101107_AACCGA reckdenwald_a_Page_030.jpg
c5db67d2b223c53418194432c022450a
e5f049f359f5206fbb7355f822fae6d417b1749b
113405 F20101107_AACCFM reckdenwald_a_Page_015.jpg
dca38b03ab654f4eb685f824de1e189e
e16ccada0998e83fdd39abdc9ff2b69927e6d954
314212 F20101107_AACCEX reckdenwald_a_Page_026.jp2
81d23ec9e29ee9100f17b4e8bd349084
442d9ce5e36dba96f24f065be7b1f1ae78daf2c2
33770 F20101107_AACDJD reckdenwald_a_Page_094.QC.jpg
a7a984b4d8ab1bc40c8c69da82fd8ac3
107c4ed29e901d76e1c9b3315933c8e9cb50eb12
2897 F20101107_AACDIP reckdenwald_a_Page_086thm.jpg
55bc49047251a460fa080e9b2bdbf742
28c87548b3389375ae5e07104220773a546a2e2b
107615 F20101107_AACCGB reckdenwald_a_Page_031.jpg
8cd297fd7fa3802a8be7fc9cc9b41888
44089cf8f9198b3b2a21ca753b0f99d340f53d81
108215 F20101107_AACCFN reckdenwald_a_Page_016.jpg
8b57a3f38a1eae4763bffd4ce8548595
a3d9a49a535c768487159c7903574b3fce02da85
198102 F20101107_AACCEY UFE0022561_00001.xml
725d2cb65f65b5a0299e249300ca3a9f
b139ba3149accf702036832be914bac8374a66c5
8319 F20101107_AACDJE reckdenwald_a_Page_094thm.jpg
9343844e409eeb81b2faa5df5201c83b
56343b6eb2c4268bc777024be877ce7ece99aecb
18672 F20101107_AACDIQ reckdenwald_a_Page_087.QC.jpg
ee5833926123ddb310d83342bb782113
79cba6a2010112ed599f272f8f13c43481f5dcab
110804 F20101107_AACCGC reckdenwald_a_Page_032.jpg
fdd2e5dcc26199653666d1f9e9a15eb3
002e26cd3a376499274500260b3dced9c38e99fa
110999 F20101107_AACCFO reckdenwald_a_Page_017.jpg
6e9e84b2b730cd745af212e1f3ee4c94
4d2b5abbe7a5fd8caf9aca7fc00c1b5ce035fe38







EXAMINING CHANGES IN MALE AND FEMALE INTIMATE PARTNER HOMICIDE
OVER TIME, 1990-2000

















By

AMY RECKDENWALD


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008


































2008 Amy Reckdenwald










ACKNOWLEDGMENTS

First and foremost, I would like to thank all of my family and friends for all their love and

encouragement through my many years of college. I honestly would never have finished without

their endless support and I know they are as happy as I am to see my graduate school career

finally come to an end. I would also like to thank my entire dissertation committee, Dr. Karen

Parker, Dr. Lonn Lanza Kaduce, Dr. Alex Piquero, and Dr. Barbara Zsembik, for all their

support and suggestions in the writing of this dissertation. I especially want to thank Dr. Karen

Parker for all her help and guidance through my 4 years at UF as well as all the assistance she

gave in the completion of this dissertation. She is a mentor as well as a good friend. I would also

like to thank Dr. Richard Hollinger for sitting in on my defense last minute and giving me many

good suggestions to improve this dissertation and Dr. Larry Winner for all his help with the

statistical analysis.









TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ...............................................................................................................3

LIST OF TA BLES .............. ......... ....................................................... 7

LIST O F FIG U RE S ................................................................. 8

ABSTRAC T ..........................................................................................

CHAPTER

1 IN TR O D U C T IO N ............................................................................ ............. .. 11

D decline in Intim ate Partner H om icide ........................................... ................... ...... .... 11
Importance of Examining Intimate Partner Homicide.........................................................12
Overview of Theoretical Perspectives ............................ ..... ..... ....................... ..... 14
Importance of Structural Changes and Domestic Violence Awareness .............................17
Research Questions........... .......... .......... ... .. .... ...... .... .......... 20
Significance of Study ....................................................... ............ .. ............ 20
S u m m ary ................... ...................2...................2..........

2 L ITE R A TU R E R E V IE W ........................................................................ .. .......................28

Theoretical Perspectives ..................................... .................... .......... ........ .... 30
Exposure R education H ypothesis........................................................... ............... 30
F actors that R educe E exposure .............................................................. .....................32
Exposure R education Predictions ............................................................ ............37
B acklash or R retaliation H ypothesis........................................... .......................... 38
B acklash P reductions ....................... ........................................ .. ........ ......... 4 1
Economic Deprivation and M arginalization ....................................... ............... 41
Economic Deprivation and Marginalization Predictions .........................................45
Sum m ary ........................................................................... .........................4 5
H y p o th e se s ..... ..................................................................... .......................................4 5
E exposure R edu action .......... ..................................................................... ........ .. .... 46
B ack lash .................... .......... .. ...................................4 6
Economic Deprivation and M arginalization ....................................... ............... 47
Change Models ............... ......... ...........................................47

3 D A TA A N D M ETH O D O LO G Y ........................................ ............................................49

Unit of Analysis ............ .............................. ...............................49
S o u rc e s o f D ata ................................................................................................................. 5 0
H om icide D ata ......... .. .................. ................................................ 50
D om estic V violence Service D ata ......... .... .......................... .............. ........... ...... 51
Measures ........................................ .................. 52


4










D dependent V variable ....................................................... ..... .............. 52
Independent V ariables .................. .................................... .. ........ .... 52
Exposure reduction m easures........................................................ ............... 53
B backlash m measures ............................................................... ....55
Economic deprivation and marginalization measures................... .... ........... 56
C o n tro ls ...................................................................5 8
M issin g D ata ........................................... ............................... 5 9
M u ltic o llin e a rity ............................................................................................................... 6 1
M e th o d o lo g y ..................................................................................................................... 6 3
D e scrip tiv e S statistic s ................................................................................................. 6 3
A n a ly tic a l P la n .................................................................................................................. 6 5
C ross-Sectional A analyses ...................... .... .................... ....................65
Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation .....................66

4 R E S U L T S ..........................................................................7 4

Cross-Sectional Analysis............................................ ......... 74
Female-Perpetrated Intimate Partner Homicide 1990 .....................................74
Male-Perpetrated Intimate Partner Homicide 1990 ............................................... 76
Female-Perpetrated Intimate Partner Homicide 2000 ..................................... 77
Male-Perpetrated Intimate Partner Homicide 2000 ..................................................... 79
Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation .....................81

5 DISCUSSION AND CONCLUSIONS .................................................90

D iscu ssion .. . ...................... ................................................................9 0
E x p o su re R edu ctio n ................................................................................................... 9 1
B ack lash .................... ......... .. ...................................9 3
Economic Deprivation and Marginalization .................................. ...............95
Summary ............ ........................... ........................................95
Changes Over Time ............... ......... ................. 97
L im station s ............. .. ............... ................. ................................... 9 8
F utu re R research ................................................................99
C o n clu sio n ......... .... ................................................. ...........................10 0

APPENDIX

A G lo ssary ......... .... .................................................. ............................10 3

B Correlation Matrix for Explanatory and Outcome Variable 1990 .............................. 109

C Correlation Matrix for Explanatory and Outcome Variables 2000. ................................1. 11

D Simplified Correlation Matrix for Explanatory and Outcome Variables 1990. ................... 113

E Simplified Correlation Matrix for Explanatory and Outcome Variables 2000. ................ 115

F Variance Inflation Factors for Simplified Models ................................... ...................117









G Correlation Matrices for Domestic Violence Service Variables. ............................118

R E F E R E N C E S ......... ..... ........................................................ .................................119

B IO G R A PH IC A L SK ETCH ........................................................................................ 129



















































6









LIST OF TABLES


Table page

3-1 Means, standard deviations (in parentheses) for all variables, 1990 and 2000 (N=
178). 69

3-2 T-test scores for change from 1990 to 2000 (N=178). .....................................................70

3-3 Variance inflation factors for all variables included in the models (N=178). .................71

3-4 Principal components analysis after varimax rotation 1990 (N=178). ............................72

4-1 Zero-inflated negative binomial regression equations with coefficients (and Z-
Scores) for gender-specific intimate partner homicide 1990........................ ...............84

4-2 Zero-inflated negative binomial regression equations with coefficients (and Z-
Scores) for gender-specific intimate partner homicide 2000 .........................................85

4-3 Summary of cross-sectional results, 1990 and 2000.................................. ... ................ 86

4-4 Fixed-effects regression coefficients for the relationship between changes in
predictor variables and changes in logged female intimate partner homicide counts
for 143a U S cities, 1990-2000 ........................ ........ ................................. ............... 87

4-5 Fixed-effects regression coefficients for the relationship between changes in
predictor variables and changes in logged male intimate partner homicide counts for
174a U S cities, 1990-2000 ............................................ ................. .. ...... 88

4-6 Summary of change model results, 1990 to 2000...................................... .................... 89









LIST OF FIGURES


Figure page

1-1 Total homicide rate per 100,000 residents for large cities, 1989-2001 ............................24

1-2 Gender-specific intimate partner homicide rates for large cities, 1989-2001 ..................25

1-3 Proportion of all homicides involving intimate partners, 1989-2001..............................26

1-4 Summary of percent change from 1990 to 2000 in total homicide rate, female-
perpetrated intimate partner homicide rate, and male-perpetrated intimate partner
h o m ic id e rate ...................................... ................................... ................ 2 7

2-1 Detailed theoretical and conceptual model on exposure reduction, backlash,
economic deprivation and marginalization and male- and female-perpetrated intimate
partner homicide. ....................... .......... ... .. .. ...... ........... 48









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


EXAMINING CHANGES IN MALE AND FEMALE INTIMATE PARTNER HOMICIDE
OVER TIME, 1990-2000

By

Amy Reckdenwald

August 2008

Chair: Lonn Lanza-Kaduce
Cochair: Karen F. Parker
Major: Criminology, Law and Society

Research on intimate partner homicide has increased in the recent years. This is partially

due to the dramatic decline witnessed over the last couple of decades in these types of homicides

as well as the growth that has occurred in public awareness and policy responses toward

domestic violence. Recent intimate partner homicide research has predominately focused around

two perspectives to explain the relationship between intimate partner homicide and domestic

violence resources the exposure reduction hypothesis and the backlash or retaliation

hypothesis, with results that support both (Dugan, Nagin, and Rosenfeld, 1999; 2003). The

exposure reduction hypothesis proposes that domestic violence resources that reduce the

exposure or contact between intimate partners should decrease the probability of intimate partner

homicide, while the backlash hypothesis suggests the opposite. That is, domestic violence

interventions may have unintended consequences and increase the risk of intimate partner

homicide if they threaten male dominance and control over their partners. The contradictory

results from these studies have frustrated advocates and have made them question their efforts to

make females safer. Societal remedies to lower intimate partner homicide seem to be addressing

male-perpetrated intimate partner violence and homicide insufficiently. It is important to gain a









better understanding of what factors are truly influencing gender-specific intimate partner

homicide, during a time that is marked by significant transformations in domestic violence

legislation and response toward domestic violence with the enactment of the Violence Against

Women's Act of 1994. This research helps to address the contradictory findings while

controlling for a number of structural factors that have shown to be important in the homicide

literature. The current study examines both the arguments of exposure reduction and backlash, in

addition to economic deprivation and marginalization to explain the observed patterns in male-

and female-perpetrated intimate partner homicide over time, something that has not been done in

the literature to date. Measures of key concepts will be collected for 2 decennial points (1990 and

2000). Supplemental homicide files for this time period and census data will be utilized. Poisson

regression models will be used to investigate which theoretical perspective is associated with

male- and female-perpetrated intimate partner homicide in 1990 and 2000. Pooled cross-

sectional fixed effect time series regression will be used to determine whether changes from

1990 to 2000 in key structural indicators influence trends in male- and female-perpetrated

intimate partner homicide during this same time period. Overall, results suggest that key

theoretical indicators do in fact influence the trends in female-perpetrated intimate partner

homicide, but not male-perpetrated intimate partner homicide.









CHAPTER 1
INTRODUCTION

Only in the area of partner homicides do women's perpetration rates approach that of
men. (Browne, Williams, and Dutton 1999: 150)


Decline in Intimate Partner Homicide

Research on intimate partner homicide1 has increased in recent years. Increased attention

toward intimate partner homicide has resulted from the dramatic decline in perpetration that has

been witnessed in these types of homicides. This decline coincided with the remarkable decline

that was observed in total homicides rates that still puzzles researchers. Beginning in the early

1990s and extending into the 21st century, Americans witnessed an unprecedented drop in

violent crime rates, specifically a drop in homicide rates. This drop was witnessed across the

U.S., but was seen in large cities2 in particular. From 1990 to 2000, homicide rates in large cities

dropped roughly 42%3. See Figure 1-1 for an illustration of the decline in total U.S. homicide

rates in large cities4. Interestingly, the decline in adult homicide rates has been recorded for all

victim-offender relationship categories, but the sharpest decline in homicides have been observed

in the family category for both males and females (Rosenfeld, 2000), with most of those

homicides generally occurring between spouses, ex-spouses, boyfriends, and girlfriends (Bureau

of Justice Statistics, 2006; Durose et al., 2005).

Statistics reveal that U.S. intimate partner homicide counts have been declining for over

two decades (Bureau of Justice Statistics, 2006; Greenfield et al., 1998). See Figure 1-2 for


1 Intimate partners include spouses, ex-spouses, or boyfriends and girlfriends.
2 Large cities have a population of 100,000 residents or more.
3 See Figure 1-4 for summary of percent change from 1990 to 2000.
4 Homicide rates per 100,000 population were calculated by dividing the homicide counts by the population and then
multiplying that number by 100,000 to obtain a rate. Denominators were obtained from the 1990 and 2000 U.S.
Census Bureau. Annual population estimates were calculated for between census years by subtracting the 1990
population from the 2000 population and then dividing that difference over time by 10 (i.e., number of years). The
annual population estimate was then subtracted from the 1990 population to obtain a population estimate for 1989
and added to 1990 and each year after to obtain a population estimate for each of the following years.









gender-specific total intimate partner homicide rates for large cities, 1989-20015. Surprisingly,

male- and female-perpetrated intimate partner homicide rates appear to follow a similar pattern,

though rates of male-perpetrated intimate partner homicide are higher than female-perpetrated

rates. That is, both male and female intimate partner homicide rates fluctuate but are fairly stable

from 1989 until 1992. From 1992 to 1994 a dramatic spike in the rates of intimate partner

homicide is apparent. After 1994, intimate partner homicide rates level off. Female-perpetrated

intimate partner homicide rates remain fairly stable from 1995 to 2001, while male-perpetrated

intimate partner homicide rates decline slightly. Though, the rates of male and female intimate

partner homicide show a similar pattern over time, a greater decline is observed in the rate of

female-perpetrated intimate partner homicide. Specifically, the rate of male-perpetrated intimate

partner homicide decreased 33% from 1989 to 2001, whereas the rate of female-perpetrated

intimate partner homicide decreased 45% during this same time. The reason for the more

substantial decrease in the rate of female-perpetrated intimate partner homicide compared with

male-perpetrated intimate partner homicide is unknown and remains an interest to researchers.

Importance of Examining Intimate Partner Homicide

There are many reasons that make intimate partner homicide an important topic for

research. First and foremost, as stated earlier, intimate partner homicide has decreased in recent

years with no concrete explanation. Furthermore, females are more likely to be killed by an

intimate partner than any other offender. Among female homicide victims, roughly 1 in 3 is

killed by an intimate partner, whereas among male homicide victims, only 5% are killed by an

intimate partner. Moreover, although intimate partner homicide rates have been decreasing, the

proportion of female homicide victims killed by an intimate partner has gradually increased from

1989 to 2001 (27.7% and 31.4%, respectively), whereas the proportion of male homicide victims

5 See footnote 4 for the rate calculation.









killed by an intimate has decreased during this same time period (5.3% and 2.8%, respectively).

This is evident in Figure 1-3. Specifically, from 1989 to 2001 there has been a 13% increase in

the proportion of females killed by an intimate partner compared to a 47% decrease in the

proportion of males killed by an intimate partner. In addition, the intimate partner is the only

victim-offender relationship category where female-perpetrated homicide rates approach that of

men (Browne, Williams, and Dutton, 1999). This is apparent when examining the difference

(i.e., the gap) between male- and female-perpetrated intimate partner homicide rates.

Typically, males commit roughly 90% of all homicides occurring in the United States,

making males 10 times more likely than females to commit murder (Bureau of Justice Statistics,

2006). Males are more likely than females to be perpetrators for every victim-offender

relationship category (i.e., intimate partners, family members, acquaintances, and strangers).

However, when females kill they are more likely to kill someone that is close to them, such as an

intimate partner or a family member, making the intimate partner the only victim-offender

relationship where female-perpetration approaches that of men (Browne, Williams, and Dutton,

1999).

The gap in the difference between male- and female-perpetrated intimate partner

homicide rates for large cities has narrowed in recent years. This is apparent when examining

Figure 1-2. The gap between the rates of male- and female-perpetrated intimate partner homicide

was fairly stable from 1989 until 1992. In 1992, the rate of intimate partner homicide increased

for both males and females, but far more drastically for males killing their female-partners. This

difference in the rise in the rates of intimate partner perpetration resulted in a widening of the

gap between the rates of male- and female-perpetrated intimate partner homicide. As the rates of

male- and female-perpetrated intimate partner homicide plummeted and then leveled off after









1994, the gap in perpetration rates began to consistently narrow. The rate of male-perpetrated

intimate partner homicide gradually decreased while the rate of female-perpetrated intimate

partner homicide remained fairly stable, resulting in a gradual narrowing of the gap in

perpetration rates that continued through 2001.

Figure 1-4 summarizes the decline in the rate of total U.S. homicides in large cities, as

well as the decline in the rates of male- and female-perpetrated intimate partner homicide from

1990 to 2000. The total rate of homicide decreased 42% during this time period. Both male- and

female-perpetrated intimate partner homicides decreased as well. The rate of female-perpetrated

intimate partner homicide had the largest decrease of 52% compared to 34% for male-

perpetrated intimate partner homicide. No confirmed explanation is available for the greater

intimate partner homicide rates among males compared to females or the gender-specific

differences that have been observed over time.

Overview of Theoretical Perspectives

Three theoretical perspectives are of interest in this study the exposure reduction

hypothesis, the backlash hypothesis, and economic deprivation and marginalization. Previous

research has supported each of these perspectives; therefore all three perspectives are believed to

influence intimate partner homicide uniquely.

The exposure reduction hypothesis proposes that the reduction in the time spent (i.e.,

exposure) between couples in violent relationships results in less intimate partner homicide,

because couples have less contact with one another and therefore have less opportunities to kill

their intimate partner. A number of factors have been considered to cause exposure reduction

effects on intimate partner homicide, such as declining marriage rates, increasing divorce rates,

greater economic and educational status of females, and the availability of domestic violence

resources and services (Dugan et al., 1999; 2003; Rosenfeld, 1997).









The backlash hypothesis proposes that the increased status of females threatens males'

control and dominance over females and will result in more intimate partner homicide, because

males will use more violence to dominate and control their female partners. Factors that have

been thought to produce backlash effects overlap with some of the factors that are considered to

produce exposure reduction effects. Some of the factors considered include: greater economic,

employment, and educational status of females and the availability of domestic violence services

when they fail to adequately reduce contact between violent partners (Dugan et al., 1999; 2003).

Research supports these two main theoretical perspectives to explain intimate partner

homicide (Dugan et al., 1999; 2003). However, the contradictory results from these studies have

frustrated advocates. For instance, Dugan et al. (1999) reported an exposure reduction effect of

domestic violence resources on male intimate partner homicide victimization, but not on female

intimate partner homicide victimization, suggesting that the movement to prevent domestic

violence only really made men safer and not women. In addition, Dugan et al. (2003) reported

that some domestic violence services actually create a backlash effect and seem to put some

women in more danger of intimate partner homicide victimization (i.e., prosecutors' willingness

to prosecute). Both of these perspectives need to be examined more closely to determine which

factors are truly influencing gender-specific intimate partner homicide.

Due to the contradictory findings from the above two perspectives it is important to also

consider other factors. This study draws from the economic deprivation and marginalization

literature to gain a better understanding of intimate partner homicide. Economic deprivation and

marginalization have been shown to explain homicide in general and therefore seems reasonable

to include in the present research. Ideas behind economic deprivation and marginalization come

from both the strain and feminist literatures. Strain theorists relate the opportunity structure to









violence. Merton, ([1949] 1968: 185-248) argues that deviance is a result of an unequal

distribution of opportunities among classes in society. According to Merton, the distribution of

opportunities in the social structure determines which classes are most likely to be involved in

crime and deviance. Males or females that lack certain opportunities because of deprivation (i.e.,

poverty and unemployment) may be frustrated or strained and this strain may influence intimate

partner homicide perpetration. On the other hand, feminists scholars tend to focus on inequalities

between men and women. Feminist literature examining inequalities tend to concentrate on the

economic marginalization hypothesis, which focuses on the economic disadvantage of women

relative to men (Heimer, 2000). The economic marginalization hypothesis argues that the

economic disadvantage of women (i.e., female unemployment and poverty) is a significant

predictor of female crime (Box and Hale, 1983, 1984; Heimer, 2000; Heimer, Wittrock, and

Unal, 2005; Hunnicutt and Broidy, 2004; Steffensmeier and Streifel, 1992). Extant homicide

research has incorporated the ideas behind economic deprivation and marginalization (Bailey,

1984; Blau and Blau, 1982; Land, McCall, and Cohen, 1990; Parker, McCall, and Land, 1999;

Steffensmeier and Haynie, 2000a; 2000b; Whaley and Messner, 2002); however limited research

has included indicators of economic deprivation and marginalization to explain intimate partner

homicide (for an exception see Reckdenwald and Parker, 2008). Economic deprivation and

marginalization in terms of poverty or unemployment may influence male and female intimate

partner homicide perpetration.

All three theoretical perspectives have received support in the literature; though at times

some of the findings are mixed and even contradictory. This study hopes to determine which

theoretical factors significantly contribute to male- and female-perpetrated intimate partner









homicide and whether or not similar theoretical factors influence male and female intimate

partner homicide.

Importance of Structural Changes and Domestic Violence Awareness

Declines in both male- and female-perpetrated intimate partner homicide rates were not

the only marked changes witnessed during the 1990 and 2000 time period. Changes were also

seen in domesticity, the status of women, levels of economic deprivation and marginalization,

and the availability of domestic violence resources. Any of these structural changes may have

had an impact on the observed patterns in intimate partner homicide.

From 1990 to 2000, there were significant changes in marital and non-marital

domesticity. Marriage rates plummeted and divorce rates increased, while the percent of

unmarried partner households increased from 3.5% in 1990 to 5.2% in 2000 (US Census Bureau,

2003). Declining marriage rates and increasing divorce rates influence intimate partner homicide

in terms of opportunity. With fewer people entering marriage and more getting divorced there is

less opportunity for violent married intimate partners to kill each other. However, while rates of

marriage and divorce are changing so are the rates of unmarried partners living together. With

more and more individuals choosing to live together and postpone marriage, the opportunity

increases for intimate partner homicide between non-married intimate partners.

Furthermore, the inherent dynamics of this type of relationship is different from a marital

relationship and may have an effect on the rate of intimate partner homicide.

Moreover, the status of females increased from 1990 to 2000. Females' employment

status, income, and educational attainment all improved. For instance, the percentage of females

16 and over that were employed in 1989 was 54.3%, compared with 57.4% in 1999 (Bureau of

Labor Statistics, 2008a). Improvements in the status of females would afford females the

opportunity to be able to leave an abusive relationship before the violence resulted in death to









themselves or their partners; thus, resulting in a lowering of intimate partner homicide. However,

these improvements in females' status may have unintended consequences on intimate partner

homicide. Some males may feel threatened by an increase in the status of females and may react

in a violent and possibly lethal manner towards their female partners to gain back the perceived

loss of control.

In addition to changing domesticity and improving status of females, the 1990s were a

time of economic prosperity. The percent of individuals that reported living in poverty decreased

from 13.1% in 1989 to 12.4% in 1999 (U.S. Bureau of the Census, 2003). In addition to this, the

unemployment rate decreased as well. In 1989, the percentage of the population that was

unemployed was 5.3%, compared with 3.7% in 1999 (Bureau of Labor Statistics, 2008b).

Economic deprivation may influence the dynamics within the household. Less stress and

frustration over lack of employment and living in poverty may result in a decrease in violence

towards intimate partners.

Domestic violence service and resource availability increased in the 1970s, but more so

after the enactment of the 1994 VAWA. During this time, domestic violence policies and

programs that offered services for females and even males expanded considerably. The

availability of domestic violence resources and services has been considered a key factor in the

explanation of the decline in intimate partner homicide (Dugan et al., 1999; 2003). Before the

1990s, many jurisdictions did not allow police to arrest on a misdemeanor charge unless they had

witnessed the act. The battered Women's Movement pressed for redefining the act of domestic

violence from a domestic disturbance (i.e., something that is seen as a private family matter) to

defining it as unacceptable and a criminal offense that will not be tolerated (Schechter, 1982). By

the mid 1970s resources began to become available to abused women. However, it could be









argued that domestic violence against women was not seen as a national social problem until the

enactment of the 1994 VAWA. The VAWA significantly increased funding for domestic

violence services, "and supported domestic violence specialization in local police departments

and prosecutor offices" (Dugan et al., 2003: 171). Some examples of the VAWA's programs

include policies to encourage the prosecution of abusers, victim services, and domestic violence

prevention programs. Although this act specifically acknowledges violence against women,

violence, both lethal and non-lethal, should be acknowledged against both men and women.

Domestic violence services should influence intimate partner homicide by limiting the

opportunity for contact between violent intimate partners and thus limit the opportunities for

partners to kill each other.

Despite the reports that the increased availability of domestic violence services and

resources is a significant factor in the explanation of the decline in intimate partner homicide

over the past couple of decades (Browne and Williams, 1989; Dugan et al., 1999; 2003),

resources have historically been directed at females and more so married females (Browne and

Williams, 1993; Browne, Williams, and Dutton, 1999). This gives females options besides

killing their partners and may in fact be a main reason that male intimate partner homicide

victimization has decreased significantly. However, this explanation does not hold for female

intimate partner homicide victimization. That is, female intimate partner homicide victimization

has not seen as drastic of a reduction as male intimate partner homicide victimization.

Taken together, there have been many changes witnessed over the last couple of decades

in intimate partner homicide as well as in domesticity, the status of women, economic

deprivation, and in the availability of domestic violence resources. It is important to determine if

these changes can account for the significant decline in intimate partner homicide. Examination









reveals that rates of female-perpetrated intimate partner homicide from 1990 to 2000 had a

greater decline than male-perpetrated intimate partner homicide. It is important to determine

what factors account for this decline as well as possible reasons why the decline was not as

drastic for male-perpetrated intimate partner homicide.

Research Questions

One of the main purposes of this study is to examine gender-specific intimate partner

homicide and determine which structural correlates contribute to the observed patterns seen in

1990 and 2000. In addition, this study will examine the changes between covariates in 1990 and

2000 to determine if any of the changes can account for the changes in male- and female-

perpetrated intimate partner homicide over time.

The following research questions will be addressed:

1. What is the impact of structural measures on male- and female-perpetrated intimate

partner homicide in 1990 and 2000?

2. Will criminological theories concerning domestic violence help explain male- and

female-perpetrated intimate partner homicide in 1990 and in 2000?

3. Does the availability of domestic violence services influence intimate partner

homicide in 1990 and 2000?

4. What is the impact of change over time (1990-2000) in key structural variables on

male- and female-perpetrated intimate partner homicide trends?

Significance of Study

This study expands on the intimate partner homicide literature in several ways.

First, this study draws from three theoretical perspectives in an effort to explain intimate partner

homicide over time. The domestic violence literature has acknowledged both the importance of

reducing the exposure between violent partners as well as backlash or retaliation by male









partners as a response or consequence to interventions designed to reduce violence between

couples. However, this research has found results that support both perspectives. This study

draws from two domestic violence perspectives while also controlling for a third theory of

economic deprivation and marginalization in attempt to more clearly understand trends in

intimate partner homicide.

This study examines both male- and female- perpetrated intimate partner homicide and

compares their trends over time. This study not only documents change but also explains

changes in intimate partner homicide trends over time and gives a better understanding of what is

correlated with this observed change. Examining changes in key indicators on intimate partner

homicide over time has not been done in the literature. The dynamic nature of key indicators will

be examined to determine whether or not the changes from 1990 to 2000 in key indicators

correlates with the change in male- and female-perpetrated intimate partner homicide over time.

For example, it is of interest to determine if the change in the percent divorced in cities from

1990 to 2000 is related to the changes in intimate partner homicide during this same time period.

Focusing on changes over time will give a better understanding of the differential trends in male-

and female-perpetrated intimate partner homicide.

The time period being examined in this study depicts a time where there has been

significant changes in homicide itself as well as domesticity (i.e., living arrangements), the status

of females, and economic deprivation and marginalization. Moreover, during this time, domestic

violence was being seen as a national social problem and funding for domestic violence services

were increasing significantly with the enactment of the 1994 Violence Against Women's Act

(VAWA). This study controls for the influence of various domestic violence services while









examining other structural factors in an effort to better understand male- and female-perpetrated

intimate partner homicide over time.

Summary

This is a macro-level study that proposes to build on existing literature by further

considering the effects of key indicators that represent the exposure reduction and backlash

perspectives, while also including a third theoretical perspective of economic deprivation and

marginalization. Data will be collected to capture these theoretical perspectives in an effort to

further the research on intimate partner homicide. Guided by these three perspectives there are

two main research questions that this study hopes to address. First, this research will determine

what structural factors correlate with male- and female-perpetrated intimate partner homicide

cross-sectionally (1990 and 2000) before examining changes overtime. Next, this research will

document changes between key structural factors to determine if these changes correlate with the

changes in male- and female-perpetrated intimate partner homicide over time.

Intimate partner homicide has gone through some remarkable changes over the past

couple of decades. Both male- and female- perpetrated intimate partner homicide have witnessed

drastic declines. However, gender-specific trends in intimate partner homicide rates show that

female-perpetrated intimate partner homicide has declined more drastically than male-

perpetrated intimate partner homicide, thus lessening the gap between male and female intimate

partner homicide perpetration rates. Recent research has predominately used two main ideas to

explain the over two decade decline in intimate partner homicide the exposure reduction

hypothesis and the backlash hypothesis. Research results favor both perspectives. Research also

points to the importance of the emergence of domestic violence services and programs with the

Battered Women's Movement in the 1970s and the VAWA of 1994. This study is designed to

examine these two theoretical perspectives as well as a third theoretical perspective of economic









deprivation and marginalization, which has shown to be very important in the homicide literature

(Bailey, 1984; Blau and Blau, 1982; Land, McCall, and Cohen, 1990; Parker, McCall, and Land,

1999; Steffensmeier and Haynie, 2000a; 2000b; Whaley and Messner, 2002), in an attempt to

gain a better understanding of the differential gender-specific trends in intimate partner

homicide. Structural factors and the availability of domestic violence services are of particular

importance.

This dissertation begins with Chapter 2 detailing the three theoretical perspectives that

are of interest in this study. In Chapter 3 the data and methodology are discussed. Chapter 4

provides the results of the multivariate analysis. Chapter 5 contains conclusions and directions

for future research.















0 Total Homicide Rate




10 0









15


0
















1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Year


Figure 1-1.Total homicide rate per 100,000 residents for large cities, 1989-2001.
I 1---------------------------------------





















Figure 1-1. Total homicide rate per 100,000 residents for large cities, 1989-2001.





















9 -
0t Female-Perpetrated Intimate Partner
I Homicide
8


C 7

IX
h 6

I 1


SX I
= 4

3 3







1




1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year



Figure 1-2.Gender-specific intimate partner homicide rates for large cities, 1989-2001.
















35.00%


- -


a m m


f f l rll
~IaiL


- Male
- Female


1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001


Year


Figure 1-3. Proportion of all homicides involving intimate partners, 1989-2001.


30.00%


25.00%




, 20.00%




15.00%




10.00%


5.00%




0.00%














1990 2000 % Change

Total Homicide Rate 20.9827 12.1792 41.9

Female-Perpetrated IPH Rate 1.520151 0.730003 51.9

Male-Perpetrated IPH Rate 5.066648 3.325714 34.36

Figure 1-4. Summary of percent change from 1990 to 2000 in total homicide rate, female-
perpetrated intimate partner homicide rate, and male-perpetrated intimate partner
homicide rate.









CHAPTER 2
LITERATURE REVIEW

While research on intimate partner homicide is growing, the majority of research on

intimate partner homicide is directed at describing victimization and/or offending patterns (Block

and Christakos, 1995; Browne and Williams, 1993; Gallup-Black, 2005; Puzone et al., 2000;

Reidel and Best, 1998; Shackelford and Mouzos, 2005) rather than analyzing these patterns by

controlling for the impact of key factors that may have an influence on intimate partner homicide

over time (Block and Christakos, 1995; Browne and Williams, 1993; Puzone et al., 2000; Reidel

and Best, 1998). This research has found that the trends in intimate partner homicide during this

time period vary not only by the gender of the victim and offender, but also by the type of

relationship between the victim and offender.

For instance, Puzone et al. (2000) examined U.S. homicide victimization from 1976-1995

and found that the decline in homicide victimization rates were more pronounced for husbands

(75%) than wives (41%), boyfriends (68%) more than girlfriends (21%)6, and ex-husbands

(88%) more than ex-wives (41%). Specifically, Puzone et al. (2000) found that the risk of being

killed by a spouse in 1995 was 2.6 times greater for a wife than a husband, 2 times greater for a

girlfriend than a boyfriend, and for the combined period of 1992-1995 was 3.8 times greater for

an ex-wife than an ex-husband. Earlier work examining male- and female-perpetrated intimate

partner homicide supports these findings (Browne and Williams, 1993). That is, Browne and

Williams (1993) examined Supplemental Homicide Report data from 1976-1987 for all states

and found that both married male-perpetrated (i.e., males killing their female spouses) and

female-perpetrated (i.e., females killing their male spouses) intimate partner homicide declined,

however the decline was more pronounced for the rate of wives killing husbands than the rate of


6 However, this decline was not statistically significant.









husbands killing wives. In addition, Browne and Williams (1993) showed that while marital

intimate partner homicide has been decreasing over the years, male-perpetrated non-marital

intimate partner homicide has actually been increasing. Browne and Williams (1993) concluded

that the rate of unmarried females being killed by their intimate partner increased significantly

during this period, while the rate that unmarried males were killed by their intimate partner

"show no clear trend during the 1976-1987 period" (87). Similarly, in Rosenfeld's (1997)

examination of intimate partner homicide victimization in California between 1987 and 1996,

rates of marital intimate partner homicide decreased while rates of non-marital intimate partner

homicide increased. Rosenfeld (1997) attributes this marked change to an increase in the

proportion of younger people who are not married. Along these lines, Riedel and Best (1998)

also examined intimate partner homicide in California for the years 1987 to 1996 and found that

intimate partner homicide is more common in common-law relationships compared with spousal

and boyfriend/girlfriend relationships. They suggest that a reason for this may be because of the

meaning that is attributed to the relationship (Makepeace, 1997) and the lack of stability

(Makepeace, 1989).

In summary, research describing recent trends in intimate partner homicide has found

some interesting results. It appears that while total intimate partner homicide is decreasing, once

one disaggregates by gender and relationship type different trends emerge. For instance, overall

patterns show that males have experienced a greater decline in intimate partner homicide

victimization than females. In addition, married intimate partner homicide has decreased, while

non-married intimate partner homicide has increased and this increase is especially pronounced

for non-married female intimate partner victimization.









Though research has not come to a definitive explanation for these trends, research that

has controlled for the impact of key factors has tended to attribute the recent patterns to the

social responses toward domestic violence; specifically, the availability of domestic violence

resources (Browne and Williams, 1989; Dugan et al., 1999; Dugan et al., 2003) and have found

that the availability of legal (i.e., presence of statutes pertaining to domestic violence) and extra-

legal services (i.e., number of shelters and other programs) to be related to the decline in the rates

of female-perpetrated intimate partner homicide, but not associated with the rates of male-

perpetrated intimate partner homicide (Browne and Williams, 1989). However, research has also

shown that some domestic violence resources (i.e., prosecutors' wiliness to prosecute) that were

intended to help women have the unintended consequence of making them more at risk for

intimate partner homicide victimization (Dugan et al., 2003). Increasing divorce rates, declining

marriage rates, and an improved economic status of women have also shown to be important in

explaining intimate partner homicide (Dugan et al., 1999; 2003; Rosenfeld, 1997). These factors

will be examined more closely as I describe empirical findings for the theoretical perspectives of

interest in this study.

Theoretical Perspectives

Three theoretical perspectives are of interest to this study. The following pages will

explain each perspective the exposure reduction hypothesis, the backlash or retaliation

hypothesis, and economic deprivation and marginalization. The section will end with detailed

hypotheses for the current study.

Exposure Reduction Hypothesis

According to the exposure reduction hypothesis, limiting the exposure or contact of

intimate partners to one another should decrease the probability of intimate partner homicide,

because there is less exposure to a violent partner and more opportunities to exit the relationship.









Ideas behind the exposure reduction perspective stem from three main findings from previous

research.

Domestic violence research shows that intimate partner homicide is commonly the result

of prolonged violence in the relationship (Browne, 1987; Campbell, 1992; Goetting, 1995;

Smith, Moracco, and Butts, 1998; Totman, 1978) or a prior history of violence (Browne, 1987;

Chimbos, 1978; Daniel and Harris, 1982; Totman, 1978). For instance, Campbell (1992)

estimated that in Dayton, Ohio 64% of female intimate partner homicide victims were abused

physically by their offender before they were killed. Also, they estimated that 79% of male

intimate partner homicide victims had abused their offender previously during the relationship.

In addition, self-report studies have shown that abused women make many attempts to try to get

outside help before actually committing the intimate partner homicide (Sherman and Berk,

1984). For instance, in reviewing police records in Detroit and Kansas City, Sherman and Berk

(1984) discovered that in 90% of the intimate partner homicide cases police had been called at

least 1 time during the 2 years prior to the homicide. Furthermore, they found that in 54% of the

intimate partner homicide cases, police had been called 5 or more times. Other research has

shown similar results (Goetting, 1995).

Moreover, research has shown that in violent relationships women are more likely to be

seriously injured (Berk et al., 1983; Browne, 1993; Brush, 1990; Crowell and Burgess, 1996;

Fagan and Browne, 1994; Langan and Innes, 1986; Schwartz, 1987; Stark, Flitcraft, and Frazier,

1979; Stets and Straus, 1990; Stout and Brown, 1995; Straus, 1993). In addition, victim-

precipitation is more likely in female-perpetrated intimate partner homicide than male-

perpetrated intimate partner homicide (Goetting, 1995; Rosenfeld, 1997; Silverman and

Mukherjee, 1987; Wolfgang, 1958). Victim-precipitation means the victim initiated the incident









that later resulted in their own death by their intimate partner. Rosenfeld's (1997) results support

this idea as do a number of other research studies (Campbell, 1992; Cazenave and Zahn, 1992;

Goetting, 1995; Jurik and Winn, 1990; Silverman and Mukherjee, 1987; Wilson and Daly,

1992a).

To summarize, intimate partner homicide is more likely to occur in a relationship where

there is a history of violence. In these violent relationships, women are more likely to be

seriously injured than men. Furthermore, victim-precipitation is more likely in female-

perpetrated intimate partner homicide. These findings suggest that women have very different

motives for killing their intimate partners and women are more likely to kill intimate partners out

of protection or self-defense (Bernard et al., 1982; Block and Christakos, 1995; Browne, 1985;

1987; Campbell, 1992; Daly and Wilson, 1988; Dobash et al., 1992; Dugan et al., 1999;

Goetting, 1988; Peterson, 1999; Stout, 1991; Stout and Brown, 1995; see Archer, 2000 and

Saunders and Browne, 2000 for a review). That is, when women kill intimate partners it is likely

a result of attempting to prevent their partners from hurting their children or themselves, during a

violent attack by their partner, or in an attempt to prevent a future attack (Browne, 1986; 1987;

Dugan et al., 1999). According to Wilson and Daly (1992b) "Unlike men, women kill male

partners after years of suffering physical violence, after they have exhausted all available sources

of assistance, when they feel trapped, and because they fear for their own lives" (206).

Factors that Reduce Exposure

Availability of domestic violence resources.

A main factor that has been shown to influence female-intimate partner homicide by

reducing the exposure of violent partners is the availability of domestic violence resources.

Browne and Williams (1989) was the first study to examine the influence of the availability of

domestic violence resources (i.e., number of shelters; wife abuse programs other than shelters)









and domestic violence legislation (i.e., statutes providing civil injunction relief for victims of

abuse; statutes providing temporary injunction relief during divorce, separation, or custody

proceeding; statutes defining the physical abuse of a family or household member as a criminal

offense; statutes permitting warrantless arrest based on probable cause in domestic violence

cases; statutes requiring data collection and the reporting of family violence by agencies that

serve these families; and statutes providing funds for family violence shelters) on female-

perpetrated intimate partner homicide. Following prior work on homicide and intimate partner

homicide, specifically, (Browne and Flewelling, 1986) Browne and Williams (1989) examined

the rate of intimate partner homicide in U.S. states for the years 1976-1979 and 1980-1984 and

found that indeed, domestic violence resources have the ability to provide abused females with

alternatives besides killing their male partners. Specifically, they found a negative relationship

between both domestic violence resources and domestic violence legislation and intimate partner

homicide, however once controls were included in the analysis only the domestic violence

resources remained significant. Other research has also supported this hypothesis and shown that

social response to domestic violence may be important in explaining intimate partner homicide

(Browne and Williams, 1989; Dugan et al., 1999; Dugan et al., 2003).

According to Browne and Williams (1989) domestic violence resources provide the

means for women "(1) to seek protection for themselves and their children, such as emergency

restraining orders against abuse and harassment or police removal of the abusive mate; (2) to

employ more direct avenues of escape, such as shelters where they can hide from the abuser for

short periods; (3) to utilize third-party interventions, such as support groups, crisis counseling,

and legal aid, which give advise and encouragement in identifying effective nonviolent means of

responding to threat or danger; and, in some areas, (4) to receive the benefits from court-









mandated treatment programs that work directly with the abuser on his problem with violence"

(76).

Dugan et al. (1999) examined the ideas behind exposure reduction as well, looking

specifically at the availability of domestic violence resources, as well as declining domesticity

(i.e., decrease in the marriage rate), divorce (i.e., increase in the divorce rate) and the improved

economic status of women (i.e., females' educational attainment relative to males' and females'

income relative to males'). They hypothesized that these three factors would reduce the exposure

between violent partners. Results are supportive of the exposure reduction perspective; however

these factors contributed more to the decline in male intimate partner homicide victimization

rather than female intimate partner homicide victimization. That is, they found that domestic

violence resources (i.e., legal advocacy) reduced homicides of married males; however they did

not find that domestic violence resources reduced homicides of married females.

Dugan et al.'s (2003) work was an extension of their previous (1999) work. They again

examined the relationship between intimate partner homicide and domestic violence resources,

but they extended their research by including more places (i.e., 48 large cities with a population

over 250,000), a longer period of time (i.e., 1976-1996), and utilize a richer source of domestic

violence resource measures. As with their previous research they examined the exposure

reduction effects7. They found that aggressive arrest policies have an exposure reduction effect

such that more aggressive arrest policies in cities is associated with less unmarried intimate

partner homicide. In addition, they also showed an exposure reduction effect such that a decrease

in Aid to Families with Dependent Children benefits is related to an increase in male victims

killed by their girlfriends, specifically African-American males. They suggest that public

assistance provided to poor women with children may cushion the financial impact of leaving a

7 Dugan et al. (2003) also examined backlash effects of domestic violence resources.









violent partner (Allard et al., 1997), but reductions in AFDC benefits would limit opportunities

for females to leave their abusive partners and live on their own; thus resulting in more

opportunities for females to kill their abusive partners. Other variables that were shown to

produce an exposure reduction effect include: marriage rates, legal advocacy, warrentless arrest

laws, and mandatory arrest laws.

Declining domesticity.

Other common factors that have been considered to produce exposure reduction effects

are increasing divorce rates, and decreasing marriage rates. Rising divorce rates would result in

fewer married couples living together and would therefore reduce the exposure between violent

couples. The same idea is behind falling marriage rates, which would reduce the exposure of

violent couples because fewer individuals would be getting married and living together.

Rosenfeld (1997) examined intimate partner homicide trends in St. Louis and found that 30% of

the decline in African American spousal homicides was attributable to falling marriage rates and

rising divorce rates.

Dugan et al. (1999) found that domesticity does in fact influence intimate partner

homicide. They found that the decline in marriage rates is related to the decline in married male

and female intimate partner victimization. That is, they found that the higher the marriage rate in

cities, the higher the rates that wives kill their husbands and husbands kill their wives.

Conversely, a high divorce rate corresponded to fewer wives killing their husbands8.

Specifically, they found that "for every reduction of 10,000 women entering marriage, 2.3

married women's lives are saved. The same decrease in marriage among men saves the lives of

3.8 married men... and... as 10,000 additional men are granted divorce, 3.8 married male

homicides are avoided" (203). Dugan et al. (2003) found similar results. Specifically, they found

8 However, declining marriage rates was related to an increase in unmarried males killing their female partners.









that lower marriage rates are related to fewer husbands being killed by their wives. Rosenfeld

(1997) suggests that "much of the decline in intimate partner homicide is a function of change in

the rate of marriage with the age groups at highest risk of homicide victimization and offending"

(73).

However, decreasing marriage rates may mean that more individuals are cohabitating

together without getting married. Cohabitation has been shown to be an important risk factor in

intimate partner homicide (Shackelford and Mouzos, 2005). Research suggests that males and

females in cohabitating relationships are at a higher risk of intimate partner homicide

victimization compared to males and females in married relationships (Daly and Wilson, 1988;

Shackelford, 2001; Wilson, Daly, and Wright, 1993; Wilson, Johnson, and Daly, 1995). For

instance, Wilson, Daly and Wright (1993) and Wilson, Johnson and Daly (1995) found that

females that cohabitate with their partner are 9 times more likely to be killed by their intimate

partner than are married females. In addition, Shackelford (2001) found that men that are

cohabitating with their female partners are 10 more likely to be victims of intimate partner

homicide compared to men in married relationships. Rodriquez and Henderson (1995) suggest

that "individuals in cohabitating relationships, having invested heavily in such relationships, may

be more likely to resort to violent retaliation (i.e., homicide) than those in dating or non-

cohabitating relationships because they do not perceive themselves as having legal 'protection'

associated with marriage" (48).

Improved economic status of women.

Improved economic status of women has also been considered to produce exposure

reduction effects. Improved economic status of women in terms of educational attainment,

income, and employment increases the access of women to opportunities that may give them









other options besides resorting to killing their male partners. Also, increased status would reduce

females' dependence on males. Dugan et al. (1999) did indeed find that females' improved status

was associated with intimate partner homicide victimization, particularly male intimate partner

homicide victimization. That is, the increase in females' relative income is associated with a

decline in married female-perpetrated homicide. Furthermore, an increase in females' relative

educational attainment is associated with a decline in non-married male victimization. They

suggest that "more educated women are better able, and perhaps more willing, to exit violent

relationships and thus avoid killing their partner" (204-205).

Exposure Reduction Predictions

In summary, it appears that women in violent relationships may be looking for other

options, but without other options than remaining in the relationship available may turn to lethal

methods. Ideas behind the exposure reduction perspective suggest that limiting the exposure or

contact of intimate partners to one another should decrease the probability of intimate partner

homicide, because there is less exposure to a violent partner and more opportunities to exit the

relationship. Empirically, if exposure reduction efforts were available to intimate partners before

either partner took the violence to a lethal level intimate partner homicide should decrease.

Based on these ideas it seems likely that factors that reduce contact or exposure between intimate

partners in violent relationships should have a significant impact on female-perpetrated intimate

partner homicide, because certain factors may give women other options than having to resort to

violence, lethal or not, to protect themselves.

The availability of domestic violence resources and the possible effect on intimate partner

homicide has gained increase attention. Results have been supportive of the exposure reduction

perspective; however this is only true of female-perpetrated intimate partner homicide. It appears

that domestic violence services that are designed to make females safer are actually making









males safer instead. Divorce rates have also proved to be a significant factor influencing intimate

partner homicide; as well as cohabitation of intimate partners, and improved status of females. It

is believed that measures of exposure reduction will have a significant effect on female-

perpetrated intimate partner homicide by creating other options for females besides killing their

male partner.

Backlash or Retaliation Hypothesis

Although research has shown the importance of reducing the exposure between intimate

partners in violent relationships, it is well known that the highest risk for homicide is when the

victim leaves the relationship and this is especially true for females being killed by their male

partners (Block and Christakos, 1995; Block, 2000). Research has suggested the possibility of

retaliation by the abusive partner from domestic violence interventions (Campbell, 1992;

Goetting, 1995). Dugan et al. (2003)'s findings are supportive of this statement. Dugan et al.

(2003) found a retaliation effect where domestic violence resources actually increased homicide

between intimate partners because they failed to effectively reduce exposure between intimate

partners. In fact, the prosecutor's willingness to prosecute violators of protection orders, though

intended to reduce exposure between violent intimate partners, actually caused a retaliation

effect where homicide increased for married and unmarried white females and African-American

unmarried males. They concluded that "being willing to prosecute without providing adequate

protection may be harmful" (192).

Contrary to exposure reduction predictions, research has hypothesized that increased

economic status of females would make it easier for females to exit an abusive relationship and

may possibly threaten males' control over their partners. Dugan et al. (1999) found that "net of

other changes, more educated women are better able, and perhaps more willing, to exit violent

relationships and thus avoid killing their partner" (205), however improved status appears to









increase the female intimate partner homicide victimization rate. Dugan et al. (1999) suggests

that some men are threatened by their loss of power over their female partners as their status

increases.

The backlash perspective theorizes that increased male violence towards their intimate

partners is due to a perceived loss of power or control because of women's improving economic

conditions (Browne, 1987; Vieraitis and Williams, 2002). This hypothesis was first developed by

Williams and Holmes (1981) and Russell (1975) as a warning of the potential consequences of

women's increased economic conditions or gender equality during the women's movement. That

is, women's improving economic conditions (i.e., increased income, employment, and

educational attainment) would make women less dependent on males, which would threaten men

and actually cause increased violence towards females by males. For instance, Avakame (1999)

states "initial reductions in gender inequality might cause increases in violence because it would

frustrate males into intensifying their use of violence to reassert their diminishing patriarchal

power and authority" (927). Though measures of gender inequality have been identified by

feminist scholars as a cause of violence against women, research has shown conflicting findings

about the direction of the relationship. Some research has been supportive of this view (Yang and

Lester, 1988; Whaley and Messner, 2002), while others have not been (Gartner et al., 1990).

Whaley and Messner (2002) hypothesized and found that greater gender equality increased male

killings of females, results directly consistent with the backlash perspective. They concluded that

"greater gender equality is threatening to male dominance, and as such, it increases male

violence against women..." (199). In addition, other research has shown that when male

unemployment is high relative to females, wives killed by their husbands' increases (Bailey and

Peterson, 1995). However, other research shows contradictory results. Gartner et al. (1990)









examined females' and males' risk of homicide victimization for the period 1950 to 1980 for 18

developed democracies (i.e., 13 western European nations, Canada, the U.S., New Zealand,

Australia, and Japan) and found a relationship between higher female education and lower

female homicide. They conclude that "the fact that the relationship between higher female

education and lower female homicide increases over time suggests that the status gained from

education acts through incremental changes in policies and norms proscribing violence against

women" (608).

Attempts have been made to test the backlash perspective on other forms of violence,

such as rape (Avakame, 1999; Baron and Straus, 1987; Ellis and Beatie, 1983), however mixed

findings have resulted. For instance, Avakame (1999) examined rape data from the National

Crime Victimization Survey (1992-1994) to determine if females' labor force participation had

an increase on females' rape victimization due to backlash, but did not find support. Instead,

Avakame (1999) found that unemployed women were at a greater likelihood of being raped.

However, Ellis and Beattie (1983) showed that gender equality (i.e., in terms of male/female

differences in earnings, educational attainment, and employment) is related to an increased

likelihood of rape. Surprisingly, Baron and Straus (1987) found the opposite results more

gender equality, lower likelihood of rape.

Individual-level research has looked at status incompatibilities between husbands and

wives and the influence on physical and emotional abuse. Kaukinen (2004) examined the effect

that men and women's relative economic contributions to the family have on husband-wife

physical and emotional abuse. Status compatibility was broken down into three statuses -









traditional status, status parity, and status reversal.9 Kaukinen (2004) found that there was no

effect of status compatibility on physical violence, but there was an effect for emotional

violence. Specifically, women in status reversal relationships are at risk for emotional abuse by

their partner. It seems that men in relationships where their female partners had a higher status

exerted their control by emotionally abusing their partners.

Backlash Predictions

Despite the mixed results for the backlash perspective, research has suggested that power

differences between partners increase the risk for abuse among women (Anderson, 1997). In

addition, men who kill their intimate partner do so to dominate and control their female partner

(Wilson, 1989; Wilson and Daly, 1992a, 1992b). It is therefore important to include measures of

backlash in the study of male- and female-perpetrated intimate partner homicide to determine if

these factors influence intimate partner homicide in 1990 and 2000 and whether the changes in

these factors contribute to the changes in intimate partner homicide over time. It is expected that

measures of backlash will have a significant impact on male-perpetrated intimate partner

homicide, because males will use violence to exert control over their female partners in

conditions where female status is high or in conditions where they feel a perceived lack of

control over their female partners.

Economic Deprivation and Marginalization

The importance of economic deprivation as a factor influencing crime has been

documented in both the strain and feminist literatures and it is consistent with the other two

perspectives of exposure reduction and backlash. The strain literature posits opportunity

structures to be related to crime rates (Merton [1949], 1968). Merton ([1949], 1968: 185-248)


9 In traditional status the economic status between the woman and the partner favored the partner. In status parity the
economic status between the woman and the partner was equal. In status reversal the economic status between the
woman and the partner favored the woman.









argues that deviance is a result of an unequal distribution of opportunities among classes in

society. According to Merton, the distribution of opportunities in the social structure determines

which classes are most likely to be involved in crime and deviance. Blocked or unavailable

opportunities have been suggested to increase offending (Cohen, 1955; Cloward and Ohlin,

1960; Merton, 1968). Caused from differentials in access to opportunities, economic inequalities

in particular have been suggested to increase violence levels (Braithwaite, 1979; Blau and Blau,

1982). Specifically, Blau and Blau (1982) found that relative racial socioeconomic inequalities

have a significant influence on criminal violence. That is, "aggressive acts of violence seem to

result not so much from lack of advantages as from being taken advantage of, not from absolute

deprivation but from relative deprivation" (126). Other research has examined the influence of

economic inequalities on homicide (for reviews see Messner and Rosenfeld, 1999). Research has

shown that, when economic deprivation is measured as an index including many measures of

deprivation it is an important predictor of homicide rates (Land, McCall, and Cohen, 1990;

Parker, McCall, and Land, 1999).

Feminist scholars are interested in a particular form of inequality, particularly gendered-

inequality and its influence on gendered-violence. Feminist literature signifies the patriarchal

system of male domination and control over women as a cause of gender inequalities in society

and ultimately the oppression of women in society (Walby, 1986). "Feminist scholars argue that

domestic violence is rooted in gender and power and represents men's active attempts to

maintain dominance and control over women" (Anderson, 1997: 655). When examining gender

inequalities a majority of research tests the economic marginalization hypothesis, which

attributes female crime to the economic disadvantage of women relative to men (Heimer, 2000).









Specifically, the economic marginalization hypothesis suggests that as the economic

disadvantage of women increases, so will female crime. Research has found support for this

hypothesis in regards to female perpetrated offenses including intimate partner homicide (Box

and Hale, 1983; 1984; Steffensmeier and Streifel, 1992; Dugan et al., 2003; Reckdenwald and

Parker, 2008). To test the economic marginalization hypothesis, research has measured the

economic disadvantage of females in terms of female poverty and joblessness. For example, Box

and Hale (1983; 1984) found that female unemployment was positively associated with female

violent and property offenses. Importantly, research has shown that there is an overlap between

factors that effect male and female offending (Boritch and Hagan, 1990; Steffensmeier and

Haynie, 2000a; Steffensmeier and Haynie, 2000b). For instance, Steffensmeier and Haynie

(2000a) examined the effects of structural disadvantage (i.e., poverty, unemployment, income

inequality, female-headed households, and percent black) on male and female offending rates for

multiple offenses (i.e., homicide, robbery, aggravated assault, burglary, larceny-theft).

Steffensmeier and Haynie (2000) concluded that structural disadvantage variables influenced

female and male offending rates in similar ways. Surprisingly, the effects of the structural

variables were stronger for male offending.

The majority of studies have examined gender inequalities in relation to the killings of

females (Bailey and Peterson, 1995; Gauthier and Bankston, 1997; Avakeme, 1999), but

examining the impact of gender inequality on male homicides is important as well (Whaley and

Messner, 2002). For instance, Whaley and Messner (2002) explored the effects of both gender

inequality and economic deprivation on gendered-homicides (i.e., male offender-male victim,

male offender-female victim, female offender-female victim, and female offender-male victim)

in large cities for 1990 to 1994. Whaley and Messner measured gender inequality as the ratio of









male to female median income, the percent of males aged 16 and over who were employed

relative to the percent of females aged 16 and over who were employed, the percent of males

employed in the labor force, the percent of managers, executives, and administrators who were

male, and the ratio of males aged 25 or more with 4 or more years of college education to

females aged 25 or more with 4 or more years of college education. They measured economic

deprivation as the percentage of black males, percentage of black females, percentage of poor

males, percentage of poor females, percentage of unemployed males, percentage of unemployed

females, and the gini index of income inequality10. Results suggested that gender equality is

positively related to rates of males killing other males and males killing females in southern

cities. In addition, they found that economic deprivation had a significant positive relationship

with females killing males, females killing other females, males killing females, and males

killing other males.

Consistent with this research, other research has found that poverty and economic

inequality influence male and female homicides (Gartner, Baker, and Pampel, 1990; Smith and

Brewer, 1992). Also, research has shown that married females are at a higher risk of intimate

partner homicide victimization in cities where there are high levels of educational attainment and

employment inequalities between men and women (Bailey and Peterson, 1995). Furthermore,

research has found that women's economic power is significantly related to their involvement in

intimate partner homicide. Specifically, it has been shown that as women's economic resources

increase their involvement in intimate partner homicide decrease (Gauthier and Bankston, 1997).







10 The gini index represents an index of income inequality. The index can range from zero, representing perfect
equality, to 1, representing perfect inequality.









Economic Deprivation and Marginalization Predictions

Although economic deprivation and marginalization is talked about differently in the

strain and feminist literatures, it is still applicable to the current study and should have an impact

on gender-specific intimate partner homicide. Economic deprivation is expected to increase both

male- and female-perpetrated intimate partner homicide, either in terms of differential

opportunity structures or marginalization.

Summary

In summary, three theoretical perspectives are examined to determine their ability in

explaining male- and female-intimate partner homicide. Each theoretical perspective is believed

to have a unique influence on male- and female-perpetrated intimate partner homicide, based on

findings from prior literature. The exposure reduction perspective and the backlash perspective

have shown to be important in understanding intimate partner homicide. Economic deprivation

and marginalization has shown to be an important explanation of homicide offending, and is

assumed to have an influence on intimate partner homicide as well. Many of the factors believed

to represent the theoretical perspectives overlap in many ways. For instance, in prior research

inequality has been used to represent economic deprivation and/or marginalization (Whaley and

Messner, 2002; Reckdenwald and Parker, 2008), as well as exposure reduction (Dugan et al.,

1999); however in this study inequality represents a measure of backlash. I have done my best to

break up the variables based on ideas from other research as well as my research hypotheses for

this study.

Hypotheses

Based on the previous literature on intimate partner homicide and homicide in general, it

is expected that indicators of exposure reduction, backlash, and economic deprivation and

marginalization will significantly impact intimate partner homicide for both males and females









for the years 1990 and 2000. Figure 2-1 contains a detailed theoretical and conceptual model of

the theoretical perspectives on male- and female-perpetrated intimate partner homicide. As

illustrated in Figure 2-1, it is believed that measures of exposure reduction will significantly

decrease female-perpetrated intimate partner homicide, by reducing the exposure between

violent partners and reducing the chance that females will have to kill their partners in self-

defense. In addition, it is hypothesized that measures of backlash will significantly increase

male-perpetrated intimate partner homicide, due to males' perceived loss of control and

dominance over their female partners. Furthermore, it is believed that measures of economic

deprivation and marginalization will increase both male- and female-perpetrated intimate partner

homicide, whether it is from the lack of opportunities or marginalization. Also, it is expected that

changes in key indicators from 1990-2000 will impact changes in male- and female-perpetrated

intimate homicide from 1990-2000. Examining these changes will give a clearer picture of the

differential declines witnessed for male- and female-perpetrated intimate partner homicide. As

depicted in Figure 2-1, I propose the following hypotheses:

Exposure Reduction-

H1: Measures of exposure reduction will have a significant negative impact on female-

perpetrated intimate partner homicide for both 1990 and 2000, based on research that suggests

that some females are more likely to kill their intimate partner in self-defense and that exposure

reduction efforts are mainly geared towards females.

Backlash-

H2: Measures of backlash will have a significant positive impact on male-perpetrated intimate

partner homicide for both 1990 and 2000, based on the idea that an increase in the status of

females causes some males to kill their female partners in an effort to exert control and

dominance over them.









Economic Deprivation and Marginalization-

H3: Measures of economic deprivation and marginalization will have a significant positive

impact on both male- and female-perpetrated intimate partner homicide for both 1990 and 2000,

due to the linkage in the literature of the role that economic deprivation and marginalization has

on crime.

Change Models-

H4: The change from 1990 to 2000 in the theoretical measures of exposure reduction will

correlate with the change in female-perpetrated intimate partner homicide from 1990-2000.

H5: The change from 1990 to 2000 in the theoretical measures of backlash will correlate with the

change in male-perpetrated intimate partner homicide from 1990-2000.

H6: The change from 1990 to 2000 in the theoretical measures of deprivation and

marginalization will correlate with the change in both female- and male-perpetrated intimate

partner homicides from 1990-2000.

Data is collected for the years 1990 and 2000 to address these hypotheses. The purpose of

this study is to address the contradictory findings of previous research on intimate partner

homicide and reach a better understanding of the trends in intimate partner homicide that have

been witnessed over this time period. To do this, this study draws on the exposure reduction and

backlash perspectives, as well as ideas behind economic deprivation and marginalization.












Exposure Reduction
% Divorced
% Unmarried Households
Shelter Rate
Legal Assistance Rate
Batterers Counseling Rate
Referral Rate



Economic Deprivation
% Living in Poverty
% Unemployed
% HHs on Public Assistance








1+


Female-Perpetrated
Intimate Partner Homicide





Male-Perpetrated
Intimate Partner Homicide


Figure 2-1.Detailed theoretical and conceptual model on exposure reduction, backlash, economic
deprivation and marginalization and male- and female-perpetrated intimate partner
homicide.


Backlash (higher levels of
equality)
Ratio F/M Median Income
Ratio F/M Education
Ratio F/M Employment









CHAPTER 3
DATA AND METHODOLOGY

This section details the methodological strategy used in this study. To begin, the data

sources and the unit of analysis are discussed. Next, all measures are operationalized. After

examining the data three methodological issues are apparent. First, there is a number of missing

cases in the dataset in regards to offender information and domestic violence resource information.

Techniques for dealing with this will be presented. Second, there is evidence of collinearity

between predictors included in this study. To address this concern, principal components

(specifically principal axis factoring) analysis is employed. Third, Ordinary Least Squares

regression estimations are inappropriate due to the skewed nature of the dependent variables.

Poisson-based regression techniques will be utilized for the cross-sectional models, because the

dependent variables are based on discrete counts of rare events and have a skewed distribution.

These issues will be outlined in the following pages.

Unit of Analysis

For this study, the unit of analysis is cities with a population of 100,000 or more in 1990.

This definition of urban cities is consistent with the Bureau of Justice Statistics definition of

large cities (Bureau of Justice Statistics 2006). The resulting urban sample contains 200 cities

that meet that criterion. Of those cities 178 cities were used in the analysis. Chattanooga, TN,

Citrus Heights, CA, East Los Angeles, CA, Kansas City, KS, Lowell, MA, Metairie, LA,

Omaha, NB, Overland Park, KS, Paradise, NV, Springfield, IL, Sterling Heights, MI,

Worchester, NY, and all cities in the state of Florida were dropped from the analysis due to not

reporting at all during either the period 1989-1991 or 1999-2001. Chicago, IL was the final city

dropped due to being an extreme outlier. The average city population in 1990 is 323,445 (a =









629,308), ranging from 100,217 to 7,322,564. The average city population in 2000 is 355,578 (a

= 686,124), ranging from 95,658 to 8,008,278.

Sources of Data

Multiple data sources were used for the current study. The dependent variables are based

on data from the Supplemental Homicide Files for the years 1989-1991 and 1999-2001 (Fox,

2005). U.S. Bureau of Census summary files 111 and 312 for 1990 and 2000 are the sources of

data for the independent variables. Uniform Crime Reports: Police Employee (LEOKA) Data are

the source of data for police force size (consistent with work by Steffensmeier and Haynie

2000a; 2000b). Also, the 1991 and 1999-2000 Domestic Violence Service Directory, collected

by the National Coalition Against Domestic Violence, is the source for the data on the

availability of domestic violence services. These years were chosen because they were the years

published by the National Coalition Against Domestic Violence and represented 1990 and 2000

closest.

Homicide Data

The homicide data used in this study were obtained from the Supplemental Homicide

Reports (SHR). The SHRs are part of the Uniform Crime Reporting Program (UCR). In addition

to monthly criminal offense information compiled for UCR purposes, law enforcement agencies

submit supplemental data to the Federal Bureau of Investigation on homicide incidents. The

SHRs contain detailed, incident-level data on nearly all murders and nonnegligent manslaughter

that have occurred in the United States for a given year. SHRs contain information for each





1 Summary File 1 presents 100-percent population and housing data for the total population, for 63 race categories,
and for many other race and Hispanic or Latino categories. See Appendix A for a more detailed definition.
12 Summary File 3 presents data on population and housing long-form subjects, such as income and education. See
Appendix A for a more detailed definition.









homicide incident, including information on victim and offender's gender as well as victim and

offender relationship information.

Domestic Violence Service Data

The data on domestic violence services were obtained from the 1991 and 1999-2000

Domestic Violence Service Directory, collected by the National Coalition Against Domestic

Violence. Unfortunately, the National Coalition Against Domestic Violence does not collect data

ever year. These years were chosen because they represented 1990 and 2000 the closest. The

information about each domestic violence program for each city is provided by the domestic

violence programs themselves. For 1999-2000, the National Directory includes information on

whether 16 service categories are provided by the domestic violence program in each city. These

service categories include: shelter, safe homes, 24-hour hotlines, TTY/TDD (ability to

communicate by telephone with women who are hearing impaired or deaf),

counseling/advocacy, non-resident support groups, children's counseling/programs, rape/sexual

assault services, emergency transportation, legal services/advocacy, wheelchair accessible, age

limit for male children, counseling for batterers, shelter capacity, maximum length of stay, and

languages other than English spoken. For 1991, the National Directory includes information on

whether 14 service categories are provided by the domestic violence program in each city. These

service categories include: shelter, safe homes, 24-hour hotlines, accept collect calls,

counseling/advocacy, non-resident support groups, children's counseling/programs, fees for

service, transportation provided, legal services/advocacy, counseling for batterers, wheelchair

accessible, deaf services, and languages spoken besides English.









Measures


Dependent Variable

The Supplemental Homicide Reports (SHR) are the data source for the two dependent

variables male- and female-perpetrated intimate partner homicide counts of arrest for the years

1990 and 2000. To determine the counts of male- and female-perpetrated intimate partner

homicide for a given city and a given year, the counts were aggregated to the city-level for each

year by the offender's gender and intimate partner relationship. Consistent with other work

(Messner et al., 2005; Villarreal, 2004) arrest counts for each city are based on three year sums

(1989, 1990, 1991; and 1999, 2000, 2001) to control for any fluctuations in reporting and relative

low frequency of offending. For the current study only murders and nonnegligent manslaughter

with single victim and single offender were included. This is consistent with other research

(Flewelling and Williams, 1999; Williams and Flewelling, 1987). Also, intimate partners include

spouses, common-law spouses, ex-spouses, and boyfriends/girlfriends. Homosexual relationships

were excluded for the purpose of this study, because the male-female dynamic is of particular

importance and because of the small number of cases (Dugan et al., 2003).

Independent Variables

U.S. Bureau of Census data for the years 1990 and 2000 were used as information for the

majority of the independent variables. Police Employee (LEOKA) Data were the source of data

for the officer rate. For simplicity, in the following section the independent variables are broken

down into 4 categories of measures: exposure reduction, backlash, economic deprivation and

marginalization, and control variables. Gender-specific indicators are utilized when appropriate.

It is acknowledged that many of these measures may overlap and could possibly fall within

another category of measures. For the purpose of this study the measures are divided with the









purpose of capturing the three theoretical perspectives of exposure reduction, backlash, and

economic deprivation and marginalization.

Exposure reduction measures

To assess whether domesticity and domestic violence service availability had an exposure

reducing effect on intimate partner homicide, eight measures of exposure reduction are included

in this study. These measures include: percentage of divorced males, percentage of divorced

females, percentage of unmarried partner households, the number of shelters offered per 100,000

females, the number of legal services offered per 100,000 females, the number of referrals

offered per 100,000 females, and the number of male batterers counseling services offered per

100,000 males.

Percentage of divorced males. The first measure of exposure reduction is the percent of

males that are divorced. This was calculated by dividing the number of males aged 15 and older

who are divorced by the total number of males aged 15 and older. This calculation was then

multiplied by 100 to obtain a percent.

(No. Males > 15 years of age that are divorced)l 00
(No. Males > 15 years of age)

Percentage of divorced females. The percent of females that are divorced was calculated

the same as the percent of males divorced. The number of females aged 15 and older who are

divorced was divided by the total number of females aged 15 and older and then multiplied by

100 to obtain a percent.

(No. Females > 15 years of age that are divorced) x100
(No. Females > 15 years of age)

Percentage of unmarried partner households. Another measure of exposure reduction is

the percent of unmarried partner households. This measure captures cohabitation by unmarried









partners. Unmarried partners include male householders living with a female partner and female

householders living with a male partner. This measure was calculated by dividing the number of

unmarried partner households by the total number of households and then multiplying this by

100 to obtain a percent.

(No. Unmarried Partner Households) x100
(Total no. Households)

Shelter rate. To measure the rate of the number of shelters available to females a measure

of the number of shelters for each city and year per 100,000 females 15 years and older was

computed. The number of shelters was divided by the total number of females aged 15 years and

older and then multiplied by 100,000 to obtain a rate. This is similar to Dugan et al. (1999)'s

calculation for the availability of hotlines, counseling, and legal services.

[ (Number of shelters for each city and year) x
x 100,000
(Number of females aged 15 years and older)

Legal service rate. The legal service rate is designed to measure the availability of legal

assistance, such as in obtaining restraining orders, court accompaniment, legal clinics or

advocacy. To measure the rate for the number of legal services available to females a measure of

the number of legal services for each city and year per 100,000 females 15 years and older was

computed. The number of legal services was divided by the total number of females aged 15

years and older and then multiplied by 100,000 to obtain a rate.

(Number of legal services for each city and year)x
x 100,000
(Number of females aged 15 years and older)

Number of referrals rate. Many cities may not have offered a particular service (i.e.,

shelters, hotlines, counseling, children's counseling, and legal services) but were able to refer

individuals to other programs that offered the service. To capture this, a measure for the rate of









the number of referrals per 100,000 females aged 15 and older for a particular city offered was

calculated. The number of referrals was divided by the total number of females aged 15 years

and older and then multiplied by 100,000 to obtain a rate.

[ (Number of referrals for each city and year) x
x 100,000
(Number of females aged 15 years and older)]

Male batterers counseling rate. To measure the rate for the number of male batterers

counseling services available to males a measure of the number of male batterers counseling

services for each city and year per 100,000 males 15 years and older was computed. The number

of batterers counseling services was divided by the total number of males aged 15 years and

older and then multiplied by 100,000 to obtain a rate.

(Number of batterers counseling services for each city and year) 1
-x 100,000
(Number of males aged 15 years and older)


Backlash measures

Three ratio measures of female status relative to men were included in this study to

capture the ideas behind backlash. These measures have been used in previous research (Whaley

and Messner, 2002). They include: the ratio of male to female median income, the ratio of the

percentage of male employment to the percentage of female employment, and the ratio of the

percentage of male education to the percentage of female education. A value of 1.0 on the ratio

measures would represent equality between males and females. Values greater than 1.0 represent

greater disadvantage of females to males.

Ratio of male to female median income. The first backlash measure is the ratio of male to

female median income. This measure was calculated by dividing the male median income by the

female median income.









S(Male median income
(Female median income)

Ratio of the percentage of male employment to the percentage of female employment.

The second backlash measure is the ratio of the percentage of male employment to the

percentage of female employment. This measure was calculated by dividing the percent of males

aged 16 or older who were employed in the civilian labor force by the percent of females aged 16

or older who were employed in the civilian labor force.

(% of males aged 16 or more who are employed in the civilain labor force)
(% of females aged 16 or more who are employedin civilian labor force)

Ratio of the percentage of males with college education to the percentage of females with

college education. The final backlash measure is the ratio of the percentage of males with a

college education to the percentage of females with a college education. This measure was

calculated by dividing the percent of males aged 25 and older with 4 or more years of college

education by the percent of females aged 25 and older with 4 or more years of college education.

(% of males aged 25 and older with 4 or more years of college education)
(% of females aged 25 and older with 4 or more years of college education)

Economic deprivation and marginalization measures

Two gender-specific measures were included in the models to capture economic

deprivation and marginalization. These measures included: percentage of males living in

poverty, percentage of females living in poverty, percentage of males unemployed, and the

percentage of females unemployed. A third measure of the percentage of households on public

assistance was also included to measure economic deprivation.

Percentage of males living in poverty. The first measure is the percentage of males living

in poverty. It was calculated by dividing the number of males below the officially defined









poverty threshold in either 1989 or 1999 by the total number of males for whom poverty status

was determined in 1989 or 1999 and then multiplied by 100 to obtain a percent.

(No. Females for whom poverty status was determined in 1989/1999 with incomes below poverty level)100
(Total Females for whom poverty status was deterermined in 1989/1999)

Percentage of females living in poverty. The next measure is the percent of females living

in poverty. It was calculated by dividing the number of females below the officially defined

poverty threshold in 1989 or 1999 by the total number of females for whom poverty status was

determined in 1989 or 1999 and then multiplied by 100 to obtain a percent.

(No. Males for whom poverty status was determined in 1989/1999 with incomes below poverty level)100
(Total Males for whom poverty status was deterermined in 1989/1999)

Percentage of males unemployed. Another measure representing economic deprivation

and marginalization is the percent of males unemployed. This measure was calculated by

dividing the number of males 16 yrs and older who are unemployed by the total number of males

aged 16 and older and then multiplying this by 100 to obtain a percent.

(No. Males > 16 years of age that are unemployed) x100
(No. Males > 16 years of age)

Females unemployed. Another measure representing economic deprivation and

marginalization is the percent of females unemployed. This measure was calculated by dividing

the number of females 16 years and older who are unemployed by the total number of females

aged 16 and older and then multiplying this by 100 to obtain a percent.

(No. Females > 16 years of age that are unemployed) x100
(No. Females > 16 years of age)

Percentage of households on public assistance. The final measure representing economic

deprivation and marginalization is the percent of households on public assistance income. This









measure was calculated by dividing the number of households on public assistance income by

the total number of households and then multiplying this by 100 to obtain a percent.

(No. of Households on public assistance) x100
(No. of Households)

Controls

Four control measures were included in this study. They include: residential mobility,

officer rate, log of the percent Hispanic, and the south.

Residential mobility. Residential mobility (i.e., percent of the population who have lived

in current household for 5 years or less) was included in this study to represent a standard social

disorganization measure. It was measured by dividing the number of people in the population 5

years and older that had lived in a different house 5 years prior by the total population 5 years

and older and then multiplying this by 100 to obtain a percent. This variable has been suggested

to influence intimate partner homicide (Browne and Williams, 1989; Williams and Flewelling,

1988), particularly female-perpetrated intimate partner homicide as a result of male aggression.

Browne and Williams (1989) found that disruptions in life in the form of population mobility

were associated with greater female-perpetrated intimate partner homicide. They suggest that

mobility may isolate females from support networks that would give females other avenues when

conflicts arise instead of resorting to lethal violence.

(Population 2 5 years of age that lived in a different house in 1995/1985)100
(Population 2 5 years of age)

Officer rate. The officer rate was calculated by dividing the total number of officers for

each city by the total population and then this was multiplied by 1,000 to obtain a rate of officers

per 1,000 persons in the population. This calculation is consistent with other work (DeWees and

Parker, 2003; Steffensmeier and Haynie, 2000a).









(Total number of police officers)l
1x 1,000
(Total population in 1990/2000) 1

Log Percent Hispanic. Another control variable is the percent Hispanic population. This

variable is included in the models to account for the fact that research has shown that Hispanics

are less likely to be involved in intimate partner homicide (Paulsen and Brewer, 2000). In their

analysis of spousal sex ratios of killing (i.e., the number of female intimate partner homicide

perpetrators for every 100 male intimate partner homicide perpetrators), Paulsen and Brewer

(2000) found that the largest disparities between male-perpetrated intimate partner homicide and

female-perpetrated intimate partner homicide was for Hispanics. One reason could be that

domestic violence and homicide is viewed differently in the Hispanic culture. To calculate the

percent Hispanic the total Hispanic population was divided by the total population and multiplied

by 100 to obtain a percent. The natural logarithm of the percent Hispanic was then calculated due

to the skewed nature of the independent variable.

Ln (Total Hispanic population) x100
nI (Total Population)

South. A dummy coded variable for the south is used in the present analysis to control for

any regional differences in intimate partner homicide offending. Research has shown that

homicide rates are higher in the south (Nisbett and Cohen, 1996) and that partner homicide is

higher in the south as well (Browne and Williams, 1989; Nisbett and Cohen, 1996). This

measure is based on Census Bureau definitions of southern regions (i.e., coded as 1 for cities

located in the southern region and 0 for cities not located in the southern region).

Missing Data

Missing information in the homicide data was problematic. For the most part this is

caused by the fact that participation in the Supplemental Homicide Reports program is









completely voluntary. Therefore, some law enforcement agencies fail to report their homicide

incidents to the FBI. According to the Bureau of Justice Statistics (2006) the SHRs are just over

90% complete. Though the coverage is high, there are still a number of homicides that go

unaccounted for. This underreporting by law enforcement agencies was corrected with an

adjustment factor based on the total number of homicide incidents reported to the Uniform Crime

Reporting program (see Fox, 2004 for discussion on weight calculation; Williams and

Flewelling, 1987). That is, SHR records were adjusted so that State and national total counts of

murder and nonnegligent manslaughter matched UCR estimates. It is important to note that this

weighting process assumes that the missing records are not systematically different from those

that have been reported to the FBI. In addition, there is a problem with using offender data

because there is a growing problem with unsolved murders. Ignoring homicides with missing

offender information understates homicide offending. A weighting strategy based on available

information about the victims murdered in both solved and unsolved homicides is used in the

present study to adjust for missing offender data (see Fox, 2004 for discussion). In the current

analysis all cases are weighted to reflect UCR estimated U.S. homicide counts and imputed

offender data.

Missing data were also an issue with the domestic violence service data. There were

instances where a program in a city did not report whether or not a particular service was offered.

In these cases, the mean of that particular service across cities was imputed. For the domestic

violence service variables included in the model, a dummy variable for whether the mean was

imputed for a city was included in the final models (i.e., 1 = mean imputation, 0 = no mean

imputation for a city) to determine if the missing values were missing at random. Results showed









that the missing values were missing at random and therefore the mean imputations were

appropriate to use.

Multicollinearity

There are two ways to assess mulitcollinearity: (1) examining bivariate correlations

between the variables and, (2) examining variance inflation factors. The bivariate correlation

matrices for all variables in 1990 and 2000 are presented in Appendix B and Appendix C. After

examination of the correlation matrices it is evident that there is collinearity and partiality among

the regressors. While there is no clear standard, researchers generally accept a correlation greater

than .500 to mean that there may be problems with mulitcollinearity. Examination shows that

there are correlations between the independent variables that are above .500. For instance, in

1990 and 2000 the percent of female unemployment and the percent of females living in poverty

are highly correlated (r =.782 and r = .816 respectively). The percent of females living in poverty

and the percent of households on public assistance are correlated (r = .786 and r = .751). The

percent of females unemployed and the percent of households on public assistance income are

also correlated (r = .793 and r = .751). Furthermore, the percent of male unemployment and the

percent of males living in poverty are highly correlated (r=.786 and r = .788 respectively). The

percent of males living in poverty and the percent of households on public assistance are

correlated (r = .766 and r = .757). The percent of males unemployed and the percent of

households on public assistance income are also correlated (r = .863 and r = .816).

To be sure that multicollinearity is not a problem variance inflation factors were

calculated. Generally it is accepted that VIFs greater than 4 represent multicollinearity (Fisher

and Mason, 1981:109). However, some have suggested that a value of 10 or greater indicates

multicollinearity (Ott and Longnecker, 2001:652; Pindyck and Rubinfeld, 1998). Variance

inflation factors for all independent variables are presented in Table 3-3. For the female-









perpetrated intimate partner homicide model in 1990, the percent of females living in poverty,

the percent of females unemployed, and the percent of households on public assistance have a

variance inflation factor greater than the value of 4 (5.465, 4.887, 5.192, respectively). None of

the other independent variables have a variance inflation factor greater than 4. As in the female-

perpetrated intimate partner homicide model, the same three variables have variance inflation

factors greater than four the percent of males living in poverty, the percent of males

unemployed, and the percent of households on public assistance (4.697, 6.970, 4.871,

respectively). Again, none of the other independent variables have a variance inflation factor

greater than 4.

For the female-perpetrated intimate partner homicide model in 2000, the percent of

females living in poverty and the percent of households on public assistance have variance

inflation factors that are greater than the value of 4 (4.827 and 4.496 respectively). In addition,

the percent of females that are unemployed has a variance inflation factor that is very close to the

value of 4 (3.834). None of the other independent variables have a variance inflation factor close

to or greater than 4. For the male-perpetrated intimate partner homicide model in 2000, the

percent of males living in poverty, the percent of males unemployed, and the percent of

households on public assistance all have variance inflation factors greater than 4 (4.040, 4.494,

and 4.865 respectively). As with the female-perpetrated intimate partner model, none of the other

independent variables have a variance inflation factor close to or greater than 4.

Due to relatively high variance inflation factors, the percent living in poverty, the percent

unemployed, and the percent of households living on public assistance were included in a

confirmatory factor analysis to determine if these variables loaded together. An index was

developed using principle axis factoring with varimax rotation. The principal components









analysis for 1990 is displayed in Table 3-4 and the analysis for 2000 is presented in Table 3-5.

The measures of percent unemployment, percent poverty, and the percent of household living in

poverty loaded together in both the male-perpetrated and female-perpetrated models for both

time periods. An index was created that is referred to as the male economic deprivation index

and the female economic deprivation index. The simplified correlation matrices for 1990 and

2000 are presented in Appendix C and Appendix D. In addition, variance inflation factors for the

simplified models are displayed in the Appendix F.13

The bivariate correlation matrices for the domestic violence service variables clearly

show that many of the individual domestic violence services are highly collinear with each other

(See Appendix G). Four variables were of particular interest because they represented service

information that were collected for both time periods and contained the most information. These

variables included whether or not shelters, legal services, male batterers counseling, and referral

services were offered for a city. 14 In the final analyses composite measures of these variables are

utilized. Examining variance inflation factors revealed that collinearity was not an issue between

these variables.

Methodology

Descriptive Statistics

The means and standard deviations for all the variables in 1990 and 2000 can be seen in

Table 3-1. Examining the dependent variables shows that both male- and female-perpetrated


13 The variance inflation factors for the simplified male-perpetrated and female-perpetrated models in 1990 and 2000
reveal that after principal components is considered and the index made, the variance inflation factors do not exceed
4.0.
14 Initially an index was created that measured the availability of domestic violence services, which included the
number of programs per 100,000 females offered for each city and the number of different domestic violence
services offered (i.e., whether the domestic violence program provided shelters, a 24-hour hotline,
counseling/advocacy, children's counseling, and legal/services/advocacy) per 100,000 females. The index for the
availability of domestic violence services did not obtain significance in the models. However, once the services were
measured separately individual services reached significance. Therefore, individual services were included in the
final analysis, instead of the combined index.









intimate partner homicides decreased from 1990 to 2000. The mean of female-perpetrated

intimate partner homicides (i.e., 3 year sum of counts) in 1990 is 6.553 (C = 10.864). In

comparison, the mean of female-perpetrated intimate partner homicides in 2000 is 2.970 (a =

4.779). The mean of male-perpetrated intimate partner homicides in 1990 is 11.569 (a = 18.378)

and the mean in 2000 is 8.770 (a = 13.929).

For the indicators of exposure reduction, both the percent of males divorced and females

divorced decreased from 1990 to 2000, whereas the percent of unmarried partner households

increased. In addition, the rate of the number of shelters available to females and the rate of the

number of legal services available to females increased from 1990 to 2000 (.354 to 1.49 and .313

to 1.42, respectively). In contrast, the rate of the number of male batterer's counseling programs

and the rate of the number of referral services offered decreased from 1990 to 2000 (.746 to .552

and .089 to .075, respectively).

Examining the indicators of backlash shows that the disparities between male and female

educational attainment and employment decreased (1.33 to 1.14 and 1.24 to 1.19, respectively),

whereas the disparity between male and female income increased from 1990 to 2000 (1.42 to

1.55). This suggests that over time, the gap between males' and females' educational attainment

and employment are becoming more narrower, while at the same time the gap between males'

and females' median income is widening.

For the indicators of economic deprivation and marginalization, both the male and female

economic deprivation indices decreased from 1990 to 2000. In 1990 the mean of the female

economic deprivation index was 29.38 (a = 11.97) and in 2000 it was 15.88 (o= 6.22). In 1990

the mean of the male economic deprivation index was 26.97 (o= 11.28) and in 2000 was 14.88

(a= 5.88).









Table 3-2 reports the paired sample t-test values to show whether the variable means are

significantly different between 1990 and 2000. All the variable means are significantly different

except for the mean of the number of referral services per 100,000 females. In summary,

statistics show that both male- and female-perpetrated intimate partner homicide decreased from

1990 to 2000. In addition, for both males and females, economic deprivation and the percent

divorced decreased from 1990 to 2000, as well as the ratio of male to females' educational

attainment, the ratio of male to female employment, and the rate of male batterers counseling

programs. In contrast, the ratio of male to female median income, percent of unmarried

households, the rate of the number of shelters, the rate of the number of legal services, percent

Hispanic, residential mobility, and the officer rate increased from 1990 to 2000.

Analytical Plan

Cross-Sectional Analyses

Cross-sectional regression models are used to analyze the impact of the theoretical

concepts at two points in time (1990 and 2000). Poisson regression models will be used for the

current study, because the dependent variables are based on discrete counts of rare events and

have a skewed distribution. In addition, the dependent variables include a number of cities with

zero counts, all of which make OLS inappropriate to use (Osgood, 2000; Osgood and Chambers,

2000).

The poison regression model is the simplest regression model for count data. However, in

the poison regression model the variance is forced to equal the mean, which is often not the case

with count data. After examining the mean and variance of the current distribution it is clear that

the mean and the variance are not equal (Cameron and Trivedi, 1998). The negative binomial

regression model is more appropriate to use because it allows for overdispersion of the data (i.e.,









where the variance exceeds the mean) (Gardner et al., 1995; Osgood, 2000). In the negative

binomial distribution the mean and the variance are as follows:

E(Y) = u Var(Y) = / + (,/2 /k) [1/k is the dispersion parameter]

Though negative binomial models take into account overdispersion, they usually under

predict the number of zero counts in the dependent variable. Since both male- and female-

perpetrated intimate partner homicides are rare events and therefore result in a number of zero

counts across cities, zero-inflated negative binomial regression is more appropriate and will be

utilized. Furthermore, the vuong test produced a z-value of 14.55 (p = .000) for the male-

perpetrated intimate partner homicide model and a z-value of 15.05 (p = .000) for the female-

perpetrated intimate partner homicide model. A significant value means that the zero-inflated

model is a better fit of the data. Excess zero counts may cause overdispersion and zero-inflated

models take this into consideration (Cameron and Trivedi, 1998). In both models the intimate

partner homicide count was converted into an equivalent of a rate by including the logged

gender-specific population size multiplied by a factor of 3 (i.e., to account for the 3 year pooled

dependent variable counts) as an exposure variable in the model (Agresti, 1996).

Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation

Using STATA, version 9.1, pooled cross-sectional time series analyses with fixed effect

estimation are used to estimate the impact of key theoretical constructs on changes from 1990 to

2000 in male- and female-perpetrated intimate partner homicide. An Ordinary Least Squares

fixed effects regression15 was employed as follows



15 Negative binomial fixed effects and random effects regression models were estimated, due to the large number of

zero counts and skewed distribution for the dependent variables. However, the model for the female-perpetrated

intimate partner homicide data would not allow for the estimation of the Hausman test because the data failed to









yit = xit + vi + + it i = 1,...,N; t =1,..., T

where xit is the vector of covariates for both cities (i) and time periods (t), with dummy variables

controlling for city (vi) and time trends (Xt). A time variable for year 1990 was included in the

analysis to control for period effects, with the year 2000 as the reference period. There are

advantages to this type of analysis. For instance, pooled cross-sectional time series analysis

increases the number of observations (i t) and also allows one to model time and space and

generalize between them.

Due to the skewed nature of the dependent variables both the female-perpetrated intimate

partner homicide count over 3 years and the male-perpetrated intimate partner homicide count

over 3 years was transformed using a logarithmic transformation. An issue arose with the

logarithmic transformation when the count of intimate partner homicide equaled zero. The log of

a zero equals -co. To fix this problem, all zero counts were also coded as zero for the logarithmic

transformation of the intimate partner homicide count. To account for this a dummy variable was

included in the models that represented zero counts for intimate partner homicide (See Hausman,

Hall, and Griliches, 1984 for rationale). Also, the log of the population 15 and older multiplied

by a factor of 3 was included in the models to control for the population size.

Both fixed-effects and random-effects models are appropriate to use for pooled cross-

sectional analysis, however for the current study the fixed effect estimation was used for a



meet the asymptotic assumptions of the Hausman test. The assumptions of the Hausman specification test are often

hard to meet. The Hausman test assumes that one of the estimators is efficient (i.e., has minimal asymptotic

variance). This assumption is violated if observations are clustered or pweighted or if the model is misspecified

(STATA 2008). For comparison purposes, both female-perpetrated and male perpetrated intimate partner homicide

were modeled using OLS fixed effects regression.









variety of reasons: (1) it expresses variables only in terms of change within a unit (Parker, 2004),

which is appropriate because the current study seeks to determine the relationship between

change in key theoretical constructs and change in the dependent variable for a given city; (2) it

controls for all omitted and included time invariant variables (Allison, 1994; Johnson, 1995); and

(3) as described below, the Hausman specification test indicated that the fixed effects model was

more appropriate to use than the random effects model.

The results of the Hausman specification test (1978) revealed that fixed effects estimation

was more appropriate than random effects estimation for both the male-perpetrated and female-

perpetrated intimate partner homicide multivariate models. The Hausman specification test

examines the null hypothesis of no difference between the parameter estimates of the more

efficient random effects model and the fixed effects model, whereas the alternative hypothesis

states that there is a significant difference between the random effects model's parameters and

the fixed effects model's parameters. This difference indicates bias in the random effects

estimation and thus the fixed effects estimation is preferred (Greene, 2000; Hausman et al., 1984;

Hsaio, 1986; Maddala, 1983). The Hausman specification test produces a Chi-square statistic,

such that a significant p-value indicates that fixed effects estimation is more appropriate and an

insignificant p-value indicates that random effects estimation is more appropriate.











Table 3-1. Means, standard deviations (in parentheses) for all variables, 1990 and 2000 (N= 178).

1990 2000


Female-perpetrated intimate partner homicide

(counts summed over 3 years)

Male-perpetrated intimate partner homicide
(counts summed over 3 years)

Percent female divorced

Percent male divorced

Percent unmarried households

Male batterers counseling per 100,000 males

Shelters per 100,000 females

Legal service programs per 100,000 females

Referral services per 100,000 females

Ratio of male to female education

Ratio of male to female income

Ratio of male to female employed

Female economic deprivation Index

Male economic deprivation index

Percent hispanic

Residential mobility

Officer rate per 1,000


6.553(10.864) 2.970(4.779)


11.569(18.378)


13.31(4.24)

9.92(3.12)

3.91(1.54)

.746(1.09)

.354(.276)

.313(.289)

.089(.243)

1.33(.140)

1.42(.173)

1.24(.106)

29.38(11.97)

26.97(11.28)

13.40(15.67)

50.22(7.88)

1.98(.910)


.303(.461)


8.770(13.929)


12.25(2.08)

9.31(2.03)

5.25(1.068)

.552(.738)

1.49(.909)

1.42(1.167)

.075(.187)

1.14(.098)

1.55(.180)

1.19(.086)

15.88(6.22)

14.88(5.88)

18.61(18.17)

51.40(6.71)

2.22(1.04)

.303(.461)


South











Table 3-2. T-test scores for change from 1990 to 2000 (N=178).
Mean 1990 Mean 2000 Mean Diff


Female-perpetrated intimate partner homicide

Male-perpetrated intimate partner homicide

Percent female divorced

Percent male divorced

Percent unmarried households

Male batterers counseling per 100,000 males

Shelters per 100,000 females

Legal service programs per 100,000 females

Referral services per 100,000 females

Ratio of male to female education

Ratio of male to female income

Ratio of male to female employed

Female economic deprivation Index

Male economic deprivation index

Percent hispanic

Residential mobility

Officer rate

+ p < .10 p<.05 p< .01


6.55(10.86)

11.57(18.38)

13.31(4.24)

9.92(3.12)

3.91(1.54)

.75(1.09)

.35(.28)

.31(.29)

.09(.24)

1.33(.14)

1.42(.17)

1.24(.11)

29.38(11.97)

26.97(11.28)

13.40(15.67)

50.22(7.88)

1.98(.910)


2.97(4.78)

8.77(13.93)

12.25(2.08)

9.31(2.03)

5.25(1.07)

.55(.74)

1.49(.91)

1.42(1.17)

.08 (.19)

1.14(.10)

1.55(.18)

1.19(.09)

15.88(6.22)

14.88(5.88)

18.61(18.17)

51.40(6.71)

2.22(1.04)


-3.58

-2.80

-1.06

-.61

1.34

-.20

1.14

1.11

-.01

-.19

.13

-.05

-13.5

-12.09

5.21

1.18

24


-5.83**

-4.32**

-3.45**

-2.76**

12.17**

-2.99**

18.11**

13.29**

-.695

-28.94**

12.09**

-5.97**

-13.64**

-24.85**

14.03**

2.76**

4 74**










Table 3-3. Variance inflation factors for all variables included in the models (N=178).
F-perp90 M-Perp90 F-perp00 M-perp00

Percent female divorced 1.228 1.464

Percent male divorced 1.204 1.666

Percent unmarried households 1.366 1.360 2.830 2.940

Male batterers programs per 100,000 males 1.245 1.241 1.405 1.399

Shelters per 100,000 females 2.198 2.138 2.392 2.387

Legal service programs per 100,000 females 2.101 2.129 2.874 2.861

Referral services per 100,000 females 1.179 1.163 1.300 1.323

Ratio of male to female median income 1.883 1.818 2.110 2.159

Ratio of male to female education 1.580 1.550 2.080 2.120

Ratio of male to female employment 1.443 1.468 2.446 2.514

Percent female unemployment 4.887 3.834

Percent male unemployment 6.970 4.494

Percent females in poverty 5.465 4.827

Percent males in poverty 4.697 4.040

Percent of households on public assistance 5.192 4.871 4.496 4.865

Percent hispanic 1.591 1.463 2.359 2.302

Residential mobility 1.976 2.219 1.696 1.718

Officer rate per 1,000 1.796 1.845 2.098 2.056

South 1.720 1.629 2.246 2.063










Table 3-4. Principal components analysis after varimax rotation 1990 (N=178).
VARIABLES Female Model Male Model

Percent below Poverty .881 .836

Percent Unemployed .888 .941

Percent of Households on Public Assistance .893 .917


Eigenvalue 2.574 2.611
% Variance Explained 85.81% 87.037%

Note: only factor loadings greater than 0.5000 are reported










Table 3-5. Principal components analysis after varimax rotation 2000 (N=178).
VARIABLES Female Model Male Model

Percent below Poverty .903 .856

Percent Unemployed .903 .921

Percent of Households on Public Assistance .832 .886


Eigenvalue 2.546 2.574
% Variance Explained 85.851 85.809%

Note: only factor loadings greater than 0.5000 are reported









CHAPTER 4
RESULTS

The following chapter presents the results of the cross-sectional analyses and the pooled

cross-sectional time series analyses.

Cross-Sectional Analysis

Tables 4-1 and 4-2 present the analysis from the zero-inflated negative binomial

regression equations for male- and female-perpetrated intimate partner homicide counts in 1990

and 2000. Table 4-3 summarizes the findings. Some interesting findings emerge regarding

exposure reduction, backlash, and economic deprivation and marginalization on male- and

female-perpetrated intimate partner homicide. The following section will begin by explaining the

findings from 1990 for both the female-perpetrated intimate partner model and the male-

perpetrated intimate partner model and then will end with the cross-sectional results from the

2000 models.

Female-Perpetrated Intimate Partner Homicide 1990

After examining the female-perpetrated intimate partner homicide model for 1990 it is

clear that the first hypothesis is not supported. The only exposure reduction measure that showed

slight significance was the percent of females divorced in cities (p < .1). It was found that as the

percent of females divorced in cities increases, the counts of female-perpetrated intimate partner

homicide increase as well. Specifically, a one standard deviation increase in the percent of

females divorced results in a 13% increase in female-perpetrated intimate partner homicides

{[exp (.029 x 4.24) = 1.130]}. However, this is not in the predicted direction. According to the

exposure reduction perspective, reducing the exposure between violent couples (i.e., an increase

in female % divorced) should reduce the likelihood of female-perpetrated intimate partner

homicide. Surprisingly, none of the other exposure reduction indicators, including the measures









for domestic violence service availability, are significantly related to female-perpetrated intimate

partner homicide.

Results are consistent with the second hypothesis. None of the backlash indicators are

significantly associated with female-perpetrated intimate partner homicide in 1990 when

controlling for a number of other structural factors. It was hypothesized that as females' status in

terms of income, employment, and educational attainment increased male-perpetrated intimate

partner homicide would increase as well. Backlash indicators were not predicted to be related to

female-perpetrated intimate partner homicide.

Consistent with the third hypothesis, economic deprivation and marginalization is

significantly related to female-perpetrated intimate partner homicide in the predicted direction.

That is, the higher the female economic deprivation index in large cities, the higher the counts of

female-perpetrated intimate partner homicide. Specifically, a one standard deviation increase in

the economic deprivation index results in a 40% increase in female-perpetrated intimate partner

homicides {[exp (.028 x 11.97 = 1.398]}. Just as predicted, it was hypothesized that economic

deprivation and marginalization would have a significant impact on female-perpetrated intimate

partner homicide, such that increases in economic deprivation and marginalization would be

related to higher female-perpetrated intimate partner homicide in cities.

In addition, despite findings from previous research none of the control variables are

significant in the model. Residential mobility, the officer rate, log of the percent Hispanic, or

whether or not the city was located in the south are not significantly related to female-perpetrated

intimate partner homicide in 1990. However, all of the coefficient signs are in the predicted

direction.









Male-Perpetrated Intimate Partner Homicide 1990

After examining the male-perpetrated intimate partner homicide model for 1990 it is

clear that contrary to the first hypothesis, a number of the exposure reduction measures are

related to male-perpetrated intimate partner homicide. It was hypothesized that exposure

reduction indicators would impact female-perpetrated intimate partner homicide by reducing the

contact between violent partners. However, three exposure reduction indicators are associated

with male-perpetrated intimate partner homicide. The number of legal services per 100,000

females is negatively associated with male-perpetrated intimate partner homicide. Specifically, a

one standard deviation increase in the number of legal services per 100,000 females results in a

15% decrease in male-perpetrated intimate partner homicide counts { [exp (-.578 x .289 = .846]}.

Moreover, the percent of males divorced is positively related to male-perpetrated intimate

partner homicide. That is, a one standard deviation increase in the percent of males divorced

results in an 18% increase in male-perpetrated intimate partner homicide counts { [exp (.053 x

3.12 = 1.179]}. This contradicts the ideas behind exposure reduction. One would assume that as

the exposure between partners decreased, particularly through divorce, intimate partner homicide

would decrease as well.

Furthermore, the number of shelters per 100,000 females is positively related to male-

perpetrated intimate partner homicide. That is, a one standard deviation increase in the number of

shelters per 100,000 females results in a 26% increase in male-perpetrated intimate partner

homicide counts { [exp (.828 x .276 = 1.257]}. None of the other exposure reduction indicators

reached statistical significance.

Also surprisingly, there is a lack of support for the second hypothesis. It was

hypothesized that the backlash indicators would be significantly related to male-perpetrated

intimate partner homicide, such that an increase in females' status would be related to an









increase in male-perpetrated intimate partner homicide as males attempt to gain control back in

their relationships. However, none of the backlash indicators are significantly related to male-

perpetrated intimate partner homicide.

Consistent with the third hypothesis, economic deprivation is significantly related to

male-perpetrated intimate partner homicide counts in cities for 1990. The higher the male

economic deprivation index in cities, the higher the counts of male-perpetrated intimate partner

homicide. Specifically, a one standard deviation increase in the male economic deprivation index

results in a 20% increase in male-perpetrated intimate partner homicide counts {[exp (.016 x

11.28 = 1.198]). This is not surprising and is in the predicted direction.

The only control variable that is significant in the model is whether or not the city was

located in the South. According to the results, if the city is located in the South the likelihood of

male-perpetrated intimate partner homicide increases. Other research has found similar results.

For instance, DeWees and Parker (2003) found that female victimization is higher in the South.

In summary, there are clear differences between which structural variables account for

male- and female perpetrated intimate partner homicide in 1990. Exposure reduction measures

influence both male- and female-perpetrated intimate partner homicides. However more of the

measures influence male-perpetrated intimate partner homicide and in particular, a couple of the

domestic violence service indicators. None of the backlash measures influence male- or female-

perpetrated intimate partner homicide in 1990. As predicted, economic deprivation and

marginalization influence both male- and female-perpetrated intimate partner homicide in 1990.

Female-Perpetrated Intimate Partner Homicide 2000

Examining the female-perpetrated intimate partner homicide model for 2000 reveals

some very different results than the 1990 model. In 2000, two of the exposure reduction

indicators are significantly related to female-perpetrated intimate partner homicide. Inconsistent









with the first hypothesis, the number of shelters per 100,000 females in cities is positively related

to female-perpetrated intimate partner homicide. That is, a one standard deviation increase in the

number of shelters per 100,000 females results in a 55% increase in female-perpetrated intimate

partner homicide counts {[exp (.48 x .909 = 1.547]}. Although a somewhat surprising finding,

this was also found in the male-perpetrated intimate partner homicide model for 1990.

In the predicted direction, the number of legal services per 100,000 females in cities is

negatively related to female-perpetrated intimate partner homicide. That is, a one standard

deviation increase in the number of legal services per 100,000 females results in a 28% decrease

in female-perpetrated intimate partner homicide counts { [exp (-.279 x 1.167 = .722]}. This was

also seen in the male-perpetrated model for 1990.

Contrary to the second hypothesis, all three backlash indicators are significantly related

to female-perpetrated intimate partner homicide. Results suggest the higher the ratio of male to

female median income (i.e., females' status relative to males' decreases) in cities, the lower the

female-perpetrated intimate partner homicide. Specifically, a one standard deviation increase in

the ratio of male to female median income results in a 19% decrease in female-perpetrated

intimate partner homicide counts { [exp (-1.156 x .180 = .812]}. Also, the ratio of male to female

employment is negatively related to female-perpetrated intimate partner homicide. That is, the

higher the ratio of male to female employment (i.e., gap widens) in cities, the lower female-

perpetrated intimate partner homicide. Specifically, a one standard deviation increase in the ratio

of male to female employment results in a 23% decrease in female-perpetrated intimate partner

homicide counts { [exp (-3.081 x .086 = .767]}. However, in the opposite direction, the ratio of

males' to females' educational attainment is positively related to female-perpetrated intimate

partner homicide. That is, a one standard deviation increase in the ratio of males' to females'









educational attainment (i.e., gap widens) results in a 22% increase in female-perpetrated intimate

partner homicide counts {[exp (2.042 x .098 = 1.222}.

Not supporting the third hypothesis, the economic deprivation index is not significant in

the 2000 model. This is surprising considering that the economic deprivation index was

significant at the .001 level in the 1990 model (p < .001). However, examining descriptive

statistics does show a significant decrease in the female economic deprivation index from 1990

to 2000. The lack of significance in the economic deprivation index in 2000 may be due to the

differences in the U.S. economy in the 1980s and 1990s. The U.S. experienced a recession in the

late 1980s until the early 1990s. This was followed by a substantial period of economic growth

throughout the 1990s.

The only control variable that is significant in the female-perpetrated model is the officer

rate. That is, the higher the officer rate per 1,000 persons, the lower the count of female-

perpetrated intimate partner homicide. This suggests that more police officers would be able to

react quicker to domestic violence calls and therefore prevent many intimate partner homicides.

Male-Perpetrated Intimate Partner Homicide 2000

The main difference between the male-perpetrated intimate partner homicide model in

2000 compared with 1990 is that the economic deprivation index is not significant. This is

surprising, considering that economic deprivation was significant at the .001 level in 1990

(p<.001). As stated earlier, the lack of significance in the economic deprivation index in 2000

may be due to the differences in the U.S. economy in the 1980s and 1990s. However, all other

results are similar.

Of the exposure reduction indicators, the percent of males that are divorced in cities is

positively related to male-perpetrated intimate partner homicide. That is, the higher the percent

of males divorced in cities, the higher the count of male-perpetrated intimate partner homicide.









Specifically, a one standard deviation increase in the percent of males divorced results in a 12%

increase in male-perpetrated intimate partner homicide counts {[exp (.058 x 2.03 = 1.124]}. This

follows the findings from the 1990 male-perpetrated model.

Also, the number of shelters per 100,000 females is positively related to male-perpetrated

intimate partner homicide. That is, the higher the number of shelters per 100,000 females in

cities, the higher the count of male-perpetrated intimate partner homicide. Specifically, a one

standard deviation increase in the number of shelters per 100,000 females results in a 20%

increase in male-perpetrated intimate partner homicide counts { [exp (.202 x .909 = 1.202]}.

Though contradictory to the exposure reduction perspective, it is not all that surprising

considering that the same finding showed up in the 1990 male-perpetrated model.

Moreover, the number of legal services per 100,000 females is negatively related to male-

perpetrated intimate partner homicide. Specifically, the higher the number of legal services per

100,000 females in cities, the lower the count of male-perpetrated intimate partner homicide.

That is, a one standard deviation increase in the number of legal services per 100,000 females

results in a 25% decrease in male-perpetrated intimate partner homicide counts { [exp (-.244 x

1.167 = .752]}. This too, was a significant finding in 1990.

Again surprisingly, none of the backlash indicators are related to male-perpetrated

intimate partner homicide in 2000. It appears that measures of females' status relative to males'

are not predictive of male-perpetrated intimate partner homicide. Also, none of the control

indicators are significant predictors of male-perpetrated intimate partner homicide.

In summary, many of the structural indicators influence male- and female-perpetrated

intimate partner homicide. Exposure reduction measures influence both male- and female-

perpetrated intimate partner homicides, with two domestic violence service variables









significantly related to both; however both are not in the predicted direction. Contrary to what

was predicted, backlash indicators influence female-perpetrated intimate partner homicide in

2000, but are not related to male-perpetrated intimate partner homicide in 2000. Also opposite of

the predictions, economic deprivation and marginalization did not influence either male- or

female-perpetrated intimate partner homicide.

Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation

Table 4-4 and 4-5 displays the results from the pooled cross-sectional Ordinary Least

Squares fixed-effects regression models. Table 4-6 summarizes the findings. Interestingly,

results suggest that the theoretical constructs explain changes in female-perpetrated intimate

partner homicide, but not male-perpetrated intimate partner homicide. Specifically, changes in

female-perpetrated intimate partner homicide counts are associated with variables used in

exposure reduction explanations, but not with indicators used in backlash explanations. That is,

cities that are characterized by increases in the percent of females that are divorced are also

characterized by decreasing counts of female-perpetrated intimate partner homicide.

Surprisingly, the only domestic violence service variable that is significant in the female-

perpetrated intimate partner homicide model is the number of shelters per 100,000 females and

the relationship between the change in this variable and the change in female-perpetrated

intimate partner homicide is in the opposite direction of what was originally predicted. Cities that

show increases in the number of shelters per 100,000 females also show increases in female-

perpetrated intimate partner homicide counts. No significant relationship is found between

changes in the percent of unmarried partner households, the number of legal services available to

females, the number of batterers counseling programs available to males, or the number of

referral services offered to females and the female-perpetrated intimate partner homicide count.









None of the changes in the indicators for backlash are significantly related to the change in

female-perpetrated intimate partner homicide from 1990 to 2000.

Furthermore, changes in the female economic deprivation index are significantly

associated with changes in female-perpetrated intimate partner homicide counts. This suggests

that intra-city changes in female economic deprivation were strongly associated with within-city

increases in female-perpetrated intimate partner homicide counts in the model. Cities that are

characterized by increases in the percent of females in poverty, the percent of females that are

unemployed, and the percent of households on public assistance from 1990 to 2000 are also

characterized by increases in female-perpetrated intimate partner homicide counts.

Overall, there is mixed support for the changes in the theoretical perspectives on the

changes in female-perpetrated intimate partner homicide counts from 1990 to 2000. A change in

economic deprivation within cities is significantly associated with changes in female-perpetrated

intimate partner homicide counts. Also, changes in two indicators of exposure reduction were

related to changes in female-perpetrated intimate partner homicide counts; however one was in

the opposite direction of what was predicted.

Interestingly, support is not found for the changes in the variables identified by the

theoretical perspectives on the changes in male-perpetrated intimate partner homicide counts

from 1990 to 2000. Male- and female-perpetrated intimate partner homicides are unique events

with different motives that require a unique set of explanations. From this analysis, it appears

that changes in factors that influence female-perpetrated intimate partner homicide from 1990 to

2000 have no influence on the perpetration of male intimate partner homicide from 1990 to 2000.

Research examining these trends has acknowledged the greater support among certain factors in

explaining female-perpetrated intimate partner homicide. For instance, Dugan et al. (1999) find









that indicators of exposure reduction primarily influence male intimate partner homicide

victimization.










Table 4-1. Zero-inflated negative binomial regression equations with coefficients (and Z-Scores)
for gender-specific intimate partner homicide 1990.

Female Model (N=178) Male Model (N=178)


Exposure Reduction
Percent divorced (gender-specific)

Percent unmarried households

Male batterers programs rate

Shelter rate

Legal service rate

Referral services rate

Backlash
Ratio of male to female education

Ratio of male to female income

Ratio of male to female employment

Economic Deprivation
Economic deprivation index (gender-specific)

Controls
Percent Hispanic (log)

Residential mobility

Officer rate

South


Constant
Maximum likelihood R-square
Log likelihood

+ p < .10 p<.05 p


.029 (1.84)+

-.024 (-.62)

-.017 (-.27)

-.015 (-.04)

-.001 (-.00)

-.060 (-.20)


.396 (-.75)

.969 (1.39)

-.514 (-.83)


.028 (3.92)**


-.057 (-1.23)

.010 (.88)

-.069 (-.75)

.255 (2.02)


-12.44**
.831
-350.368**


.053 (3.26)**

-.024 (-.82)

.028 (.59)

.828 (2.72)**

-.578 (-1.99)*

-.202 (-.80)


-.221 (-.51)

-.041 (-.10)

-.256 (-.46)


.016 (2.80)**


-.041 (-1.06)

-.002 (-.23)

-.050 (-.68)

.332 (3.13)**


-10.359**
.821
-462.997**


< .01










Table 4-2. Zero-inflated negative binomial regression equations with coefficients (and Z-Scores)
for gender-specific intimate partner homicide 2000.
Female Model (N=178) Male Model (N=178)


Exposure Reduction
Percent divorced (gender-specific)

Percent unmarried households

Male batterers program rate

Shelter rate

Legal service programs rate

Referral services rate

Backlash
Ratio of male to female education

Ratio of male to female income

Ratio of male to female employment

Economic Deprivation
Economic deprivation index (gender-specific)

Controls
Percent Hispanic (log)

Residential mobility

Officer rate

South


.052 (1.18)

.105 (.79)

.077 (.61)

.48 (3.95)**

-.279 (-2.41)*

.023 (.04)


2.042 (1.72)+

-1.156 (-1.66)+

-3.081 (-1.97)*


.013 (.72)


.001 (.01)

-.020 (-.98)

-.193 (-1.86)+

.218 (.91)


.058 (1.93)+

.112(1.35)

.076 (.94)

.202 (2.69)**

-.244 (-3.32)**

-.565 (-1.55)


.537 (.70)

.019 (.05)

-.040 (-.04)


.009 (.71)


-.045 (-.71)

-.010 (-.84)

.057 (.95)

.085 (.59)


Constant
Maximum likelihood R-square
Log likelihood

+ p < .10 p<.05 p


-8.61**
.783
-245.044**


< .01


-11.99**
.8084
-424.747**










Table 4-3. Summary of cross-sectional results, 1990 and 2000.
1990


Female Model



% divorced (+)


Male Model



% divorced (+)
# shelters (+)
# legal services (-)


Female Model



# shelters (+)
# legal services (-)


Male Model



% divorced (+)
# shelters (+)
# legal services (-)


M/F income (-)
M/F education (+)
M/F employ (-)


EcoDep Index (+)


EcoDep Index (+)


Control NONE South (+) Officer Rate (-) NONE


Exposure
Reduction


2000


Backlash


NONE


NONE


Economic
Deprivation


NONE


NONE


NONE


Control


NONE


South (+)


Officer Rate (-)


NONE










Table 4-4. Fixed-effects regression coefficients for the relationship between changes in predictor
variables and changes in logged female intimate partner homicide counts for 143a US cities,
1990-2000
b t

Exposure Reduction
Female percent divorced -.036+ -1.74
Percent unmarried households .065 1.26
Male batterers program rate .055 .60
Shelter rate .261* 2.15
Legal service programs rate -.035 -.40
Referral services rate -.268 -.90

Backlash
Ratio of male to female education .095 .13
Ratio of male to female income -.184 -.29
Ratio of male to female employment .299 .32

Economic Deprivation
Female economic deprivation index .035* 2.37

Controls
Percent Hispanic (log) -.260+ -1.70
Residential mobility -.023 -1.03
Officer rate -.111 -.99
1990b -.030 -.10
Female dummy log zero' -.754** -5.51
Female population 15 and over (log)d -.534** -2.81
Constant 8.90** 2.65

Within R2 .537
Hausman test .000**

Notes: (a) 70 observations (35 groups) were dropped due to all zero outcomes (i.e. zero count for intimate partner
homicide for both 1990 and 2000); (b) coefficients for period effects are relative to the reference category, 2000;
(c) represents a dummy variable for imputation of zero for log of a zero count; (d) female population is multiplied
by a factor of three to account for the dependent variable multiplied by a factor of three.
+ p < .10 p<.05 ** p< .01










Table 4-5. Fixed-effects regression coefficients for the relationship between changes in predictor
variables and changes in logged male intimate partner homicide counts for 174a US cities, 1990-
2000
b t

Exposure Reduction
Male percent divorced -.009 -.28
Percent unmarried households .034 .66
Male batterers program rate .113 1.40
Shelter rate .035 .33
Legal service programs rate .097 1.16
Referral services rate -.025 -.09

Backlash
Ratio of male to female education .389 .78
Ratio of male to female income .360 .69
Ratio of male to female employment -.834 -.99

Economic Deprivation
Male economic deprivation index .008 .66

Controls
Percent Hispanic (log) -.013 -.09
Residential mobility -.010 -.77
Officer Rate .016 .17
1990b .338 1.16
Male dummy log zero' -1.01** -6.79
Male population 15 and over (log)d .296 .68
Constant -1.99 -.34

Within R2 .303
Hausman Test .008**

Notes: (a) 8 observations (4 groups) were dropped due to all zero outcomes (i.e. zero count for intimate partner
homicide for both 1990 and 2000); (b) coefficients for period effects are relative to the reference category, 2000;
(c) represents a dummy variable for imputation of zero for log of a zero count; (d) male population is multiplied
by a factor of 3 to account for the dependent variable multiplied by a factor of 3.

+ p < .10 p<.05 ** p< .01










Table 4-6. Summary of change model results, 1990 to 2000.
Female Model

Exposure Reduction % divorced (-)
# shelters (+)


Backlash


NONE


Economic Deprivation


Control


EcoDep Index (+)


% Hispanic (-)


Male Model

NONE


NONE


NONE


NONE









CHAPTER 5
DISCUSSION AND CONCLUSIONS

The goal of this study was to examine the trends in both male- and female-perpetrated

intimate partner homicides from 1990 to 2000 to obtain a better understanding of the differential

declines in male- and female-perpetrated intimate partner homicide. Clearly, the decline in intimate

partner homicide over time is a very complex issue. This research suggests that one explanation

will not suffice. This research was based on the theories of exposure reduction, backlash, and

economic deprivation and marginalization. Results suggest a lack of support for these theories.

Cross-sectional analyses are presented to evaluate the ability of the exposure reduction perspective,

the backlash perspective, and ideas behind economic deprivation and marginalization in explaining

male- and female-perpetrated intimate partner homicide counts in large cities at two time periods

(1990 and 2000). Pooled cross-sectional time series analyses are presented to evaluate the

effectiveness of the exposure reduction perspective, the backlash perspective, and economic

deprivation and marginalization in explaining the changes in male- and female-perpetrated intimate

partner homicide in large cities over time.

Discussion

Dramatic changes were occurring in domesticity, the status of females in terms of income,

employment and educational attainment, economic deprivation, and domestic violence resources

from 1990 to 2000 in large cities. For instance, during this time period the percentage of the

population divorced decreased for both males and females, while the percentage of unmarried

households increased. Furthermore, females have made gains in educational attainment and

employment relative to males (i.e., status of females increased). However, this is not true for

females' median income. Females' median income relative to males actually decreased during this

time period (i.e., status of females decreased). Also, economic deprivation in terms of poverty,









unemployment, and public assistance income decreased for both males and females. Moreover, the

availability of shelters and legal services for females increased, whereas the availability of male

batterers counseling programs and referral services decreased. Any of these changes may have an

impact on the gender-specific trends in intimate partner homicide that have been witnessed over

the last couple of decades.

Exposure Reduction

Some of the structural predictors are related to intimate partner homicide perpetration.

According to the exposure reduction perspective, it was hypothesized that measures that reduced

the contact between violent partners would be associated with female intimate partner homicide

perpetration in both 1990 and 2000. For 1990, cross-sectional analyses show that the exposure

reduction measure of female percent divorced is in fact associated with female-perpetrated intimate

partner homicide; however the relationship is in the opposite direction of what was predicted.

Instead of the percent divorced being associated with a decrease in female-perpetrated intimate

partner homicide, as the exposure reduction perspective would predict, the percent of females

divorced is associated with an increase in female-perpetrated intimate partner homicide. Other

research has found similar results (Dugan et al., 1999; 2003). One possible explanation may have

to do with the dynamics that come with the ending of a violent relationship. Research has

suggested that the most dangerous time in a violent relationship is when the relationship is coming

to an end (Campbell, 1992; Goetting, 1995). Some females may be reacting to male partners'

violence over the end of the relationship and may resort to lethal means to protect themselves.

None of the other exposure reduction measures were related to female-perpetrated intimate partner

homicide in 1990.

In 2000, two of the domestic violence service variables reach significance for female-

perpetrated intimate partner homicide, however in opposite directions. The number of shelters per









100,000 females is associated with an increase in female-perpetrated intimate partner homicide.

One reason for this finding may be that although the shelter rate is accounting for the number of

shelters available per 100,000 females in cities, it may not be capturing shelter usage by females

in abusive relationships. It is possible that the availability of shelters across cities is increasing

but females are not taking advantage of them to reduce the exposure between themselves and

their abusive partner. On the opposite end of the spectrum, females may in fact be using shelters

to reduce the exposure between themselves and their violent partners, but once shelter stays end

may find themselves in contact with their violent partners and may have to resort to lethal

violence to protect themselves, if adequate protection is not available. Furthermore, research has

shown that the most dangerous time in an abusive relationship is when a partner is trying to end

the relationship (Goetting, 1995). Some females may be reacting to male partners' violence over

the threat of the end of the relationship and may resort to lethal means to protect themselves.

The second domestic violence service variable associated with female-perpetrated

intimate partner homicide is the number of legal services per 100,000 females. This variable is

associated with a decrease in female-perpetrated intimate partner homicide. It seems that as

females are provided with more opportunities for legal assistance, such as in obtaining

restraining orders, court accompaniment, legal clinics or advocacy the opportunities for females

to kill their male partners decreases.

Interestingly, three of the six exposure reduction measures are associated with male-

perpetrated intimate partner homicides in 1990 and 2000. Similar to female-perpetrated intimate

partner homicide in 2000, the percent of males divorced is positively related to male-perpetrated

intimate partner homicide. Based on suggestions from prior research (Goetting, 1995), it seems









plausible that the end of a relationship may cause some males to use more violence towards their

female partners, resulting in more male-perpetrated intimate partner homicide.

The shelter rate and the legal service rate were also significantly related to male-perpetrated

intimate partner homicide, however in opposite directions. Instead of reducing intimate partner

homicide due to a decrease in opportunity, as the exposure reduction perspective would predict, the

number of shelters per 100,000 females increases male intimate partner homicide perpetration.

Although surprising, this finding is plausible in light of the research that has found that male-

perpetrated homicide is likely to occur when a partner is trying to leave an abusive relationship

(Goetting, 1995). Some males may feel threatened that their partners are leaving and staying at a

shelter and may react in a lethal manner. Moreover, adequate protection may not be provided

when females are ending their stay at a shelter.

In addition, the number of legal services available to females is related to a decrease in

male intimate partner homicide perpetration. It seems reasonable that, as females are provided

with more opportunities for legal assistance, the opportunities for males to kill their female

partners would decrease. Dugan et al. (2003) found similar results. That is, they found that legal

advocacy, measured as an index of the number of agencies with a separate budget for legal

advocacy and the number of agencies that have lawyers on staff, was related to fewer killings of

white married females by their partners.

Backlash

According to the backlash perspective, it was hypothesized that measures that may threaten

male control and dominance would be associated with male intimate partner homicide perpetration

in 1990 and 2000. Interestingly, support was not found for this hypothesis. Although surprising,

backlash explanations may be less relevant in explaining male-perpetrated intimate partner

homicide, because gender roles have changed over the years. Today, it is more acceptable for









women to have jobs outside of the home and pursue higher education. Due to this transformation in

gender roles, females' economic and social status has increased. Furthermore, Felson (2006)

surveyed 10,000 men and women about their spouses' behavior and concluded that males are no

more controlling than females.

Contrary to what was expected, backlash correlates are associated with female-perpetrated

intimate partner homicide in 2000. It appears that, as the gap between male and female

employment and income widens (i.e., females' status relative to males decreases), female-

perpetrated intimate partner homicide decreases. However, as the gap between males' and females'

educational attainment widens, female-perpetrated intimate partner homicide increases. Females

may react to disparities in income, employment and educational attainment just as it is predicted

that males do, just in terms of frustration and not in terms of maintaining control and dominance

in the relationship. Along these lines, females appear to be taking their frustrations about the

widening gap in educational attainment out on their male partners, but not their frustrations about

the widening gap in income or employment. Anderson (1997) suggests that it is possible that

females with lower status relative to males are more likely to engage in violence because they are

less likely to be able to leave a violent relationship. That being said, females with a lower

educational attainment status relative to males may be less likely to be able to remove

themselves from a violent situation, possibly because education affords females opportunities to

learn about services available to reduce contact to prevent violence. Anderson (1997) found that

"women with slightly less education than their male partners have almost 3 times greater odds of

perpetrating domestic violence than women partnered with men processing the same amount of

education" (664).









Economic Deprivation and Marginalization

According to the ideas behind economic deprivation and marginalization, it was

hypothesized that measures of poverty, unemployment and the percent of the population on public

assistance would be associated with both male- and female-perpetrated intimate partner homicides.

Support for this hypothesis was found in 1990. Poverty, unemployment, and having to rely on

public assistance income may significantly impact the dynamics within the home, either in terms

of frustration or strain. Males and females may react to this strain by lashing out in a lethal

manner at their intimate partners.

Summary

The results of this analysis offer support for economic deprivation and marginalization in

explaining both male- and female-perpetrated intimate partner homicide in 1990. Results also

suggest an association between some of the variables representing the exposure reduction

perspective and male- and female-perpetrated intimate partner homicide in both 1990 and 2000. In

general, the three exposure reduction indicators that are consistently significant in 1990 and 2000

are gender-specific percent divorced and two indicators measuring the availability of domestic

violence services the shelter rate and the legal service rate. Importantly, these measures are

related to both male- and female-perpetrated intimate partner homicides in a similar fashion. A one

standard deviation increase in any of these measures results in a significant percentage change in

intimate partner homicide counts. The availability of legal services in cities supports the exposure

reduction hypotheses. That is, an increase in the number of legal services per 100,000 females is

related to an increase in male-perpetrated intimate partner homicide in 1990 and 2000 and female-

perpetrated intimate partner homicide in 2000.

However, not all of the significant exposure reduction measures influence intimate partner

homicide in the predicted direction. For instance, an increase in the percent divorced increases both









male and female intimate partner homicide. Furthermore, results suggest that an increase in the

number of shelters per 100,000 females is related to an increase in male-perpetrated intimate

partner homicide in 1990 and 2000 and female-perpetrated intimate partner homicide in 2000. This

is a main domestic violence service measure that has historically been used to measure domestic

violence services and was expected to have an exposure reduction effect on intimate partner

homicide perpetration. Instead, it appears that this measure is associated with more males and

females reacting out in a lethal manner towards their intimate partner, whether it is for protection

or to maintain control and dominance. Dugan et al. (2003) also found a backlash effect for one of

their measures of domestic violence services. Specifically, they found that prosecutors' willingness

to prosecute violators of protection orders was related to an increase in homicides of married and

unmarried white females and unmarried African-American males. They concluded that, although

prosecutors' willingness to prosecute is advantageous, if this willingness comes without also being

able to provide adequate protection for victims, homicide may result. As stated earlier, if these

services aren't being utilized by females, they will be less likely to have an exposure reduction

effect on intimate partner homicide. Also, given the fact that the likelihood of intimate partner

homicide is extremely high when the relationship is ending, domestic violence resources need to

provide adequate protection when the relationship is ending or all the good intentions of increasing

shelter availability will not prevent intimate partner violence and homicide.

Indicators measuring the backlash perspective do not predict male-perpetrated intimate

partner homicide in 1990 or 2000 as was hypothesized. Interestingly, backlash indicators do reach

significance in the 2000 female-perpetrated intimate partner homicide model. However, the

relationship between these variables and female-perpetrated intimate partner homicide is complex.









A one standard deviation increase in any of these three ratio measures results in a significant

percentage change in the female-perpetrated intimate partner homicide in 2000. An increase in the

ratio of median income and the ratio of male to female employment has a negative effect on

female-perpetrated intimate homicide (i.e., decreases), while an increase in the ratio of male to

female education has a positive effect (i.e., increases) on female-perpetrated intimate partner

homicide.

Changes Over Time

In regards to changes over time, changes in structural variables from 1990 to 2000 were

related to intimate partner homicide perpetration over time, however this is only true for female-

perpetrated intimate partner homicide. Interestingly, although many of the structural predictors

influence male-perpetrated intimate partner homicide cross-sectionally, these perspectives did not

explain male-perpetrated intimate partner homicide over time at all. Results suggest that cities that

are characterized by changes in the percent of females divorced from 1990 to 2000 are

characterized by changes in female-perpetrated intimate partner homicide counts. Also, cities

that show changes in the number of shelters per 100,000 females from 1990 to 2000 also show

changes in female-perpetrated intimate partner homicide counts. Furthermore, cities that are

characterized by changes in the percent of females in poverty, the percent of females that are

unemployed, and the percent of households on public assistance from 1990 to 2000 are also

characterized by changes in female-perpetrated intimate partner homicide counts. Lastly, cities

that are characterized by changes in the logged percent Hispanic population from 1990 to 2000

are characterized by changes in female-perpetrated intimate partner homicide counts.

Overall, there is mixed support for the changes in the theoretical perspectives on the

changes in female-perpetrated intimate partner homicide counts from 1990 to 2000 and

absolutely no support on the changes in male-perpetrated intimate partner homicide counts.









Changes in economic deprivation within cities are significantly associated with changes in

female-perpetrated intimate partner homicide counts. Also, changes in two indicators of

exposure reduction were related to changes in female-perpetrated intimate partner homicide

counts; however the shelter rate was in the opposite direction of what was predicted and

increased shelter availability appears to be associated with more harm than good if shelters are

not being taken advantage of or if shelters are not providing adequate protection.

Limitations

There are some limitations to this research. First and foremost, the data utilized for the

dependent variables was based on Supplemental Homicide Files. The deficiencies of the database

have been well-documented (Junger-Tas and Marshall, 1999). These include underreporting,

errors in the assignment of relationships, failure to identify relationship in a substantial portion of

cases, and other missing information SHFs include information on homicides that were submitted

by law enforcement agencies to the FBI. Submission of homicide information is completely

voluntary and therefore not 100% accurate. In addition, many homicides are coded as having an

unknown victim-offender relationship. A weighting procedure was utilized to account for this lack

of reporting and lack of relationship information.

Moreover, it is acknowledged that actual usage of the domestic violence services was

unable to be captured in cities. Just because shelters, legal services, batterers counseling programs,

and referrals are offered does not mean they are being utilized to the fullest by individuals in

violent relationships. Also, the size of the domestic violence service programs (i.e., service

capacity) was unable to be captured. If domestic violence services are not being utilized or if

programs have limited capacity, domestic violence services would have a limited effect on intimate

partner homicide.









Also, it was difficult to divide the structural factors of interest into the three theoretical

categories. It is possible that many of the structural indicators could have fallen into another

theoretical category. For instance, although for this study females' status relative to males' was

considered to represent the backlash hypothesis, other research has considered females' status

relative to males' as measuring exposure reduction, because an increase in the status of females

affords them more opportunities to leave abusive relationships. In addition, research has used

females' status relative to males' to represent economic deprivation and marginalization. For the

current study, all structural indicators were divided into theoretical categories based on the

developed hypotheses.

Furthermore, data was limited to two time points (1990 and 2000). Although the 1990s

witnessed drastic changes in intimate partner homicide perpetration, changes in intimate partner

homicide have been witnessed since 1976. It would be interesting to expand the data to include

earlier data. Also, data was limited to large cities with a population over 100,000. It would be

useful to examine data in all sizes of cities over time.

Future Research

Future research on intimate partner homicide could take many directions. First, it would be

useful to examine intimate partner homicide as well as intimate partner aggravated assault, in

addition to the access or availability of medical centers. Aggravated assault and homicide are

similar events, with the exception of the outcome (i.e., one ends in death, whereas the other does

not). Harris et al. (2002) argues that research would benefit by focusing on serious assaults, with

only a small percentage that actually end in death. They show that the decline in the homicide rate

parallels the increase in medical technology and support services. It is seems reasonable that the

availability of medical care (i.e., trauma centers) may influence whether an intimate partner

aggravated assault remained an aggravated assault or resulted in death and was categorized as an









intimate partner homicide. Many intimate partner homicides may have been aggravated assaults

that became homicides when adequate medical care was not available. Greater availability of

medical resources may prevent many intimate partner homicides and thus is an important factor

and should be considered when examining intimate partner homicide over time.

Also, it would be important to examine the cities that were excluded from the current

analysis due to lack of reporting. The majority of the cities excluded were located in the state of

Florida. Research has shown that homicide rates are higher in Southern states (DeWees and Parker,

2003; Smith and Brewer, 1992). If intimate partner homicide data could be gathered, it would be

interesting to examine intimate partner homicide in large cities in Florida to see if results were

similar with the results from the current study.

Furthermore, it would be important to include other factors in the analysis especially the

age structure of the cities. Browne et al. (1999) examined age-specific rates of intimate partner

homicide from 1980 to 1995 and found that different trends emerge for different age groups and

for males and females. They found that a decline in female-perpetrated intimate partner homicide

occurred in all age groups, except the age group 13 to 17 years old (rates have remained low). The

greatest decreases occurred in the 25 to 34 and the 35 to 44 year old age groups. For male-

perpetrated intimate partner homicide, increases were found for the 13 to 17 and the 18 to 24 year

old age groups, while decreases were seen in the 25 to 34 and the 35 to 44 year old age groups.

Since differential trends in intimate partner homicide have been witnessed for various age groups

and younger age groups are more likely to participate in intimate partner homicide, it would be

important to control for age in future analyses.

Conclusion

This research offers many contributions to the literature on intimate partner homicide. Most

importantly, this research examines the dynamic nature of a number of different structural









variables on both male- and female-perpetrated intimate partner homicide over time. This is

something that has not been done in the literature to date. The time period covered includes a time

in history where dramatic changes were occurring in domesticity, the status of females, economic

deprivation and marginalization, and the availability of domestic violence resources with the

enactment of the Violence of Against Women Act in 1994.

A main finding of this research speaks to the ineffectiveness of domestic violence services.

The number of shelters per 100,000 females is significantly related to intimate partner homicide in

1990 and 2000 and over time. Despite all the efforts to increase shelter availability to females in

violent relationships, it appears that the increase in availability is actually associated with an

increase in intimate partner homicide. For instance, in 1990 and 2000 the increase in the shelter

rate was related to an increase in male-perpetrated intimate partner homicide. Furthermore, not

only was this finding seen in 2000 for female-perpetrated intimate partner homicide, this finding

was witnessed over time as well. Efforts to prevent domestic violence and homicide need to make

sure that they also provide adequate protection during times that are characterized by increased

violence.

This research adds to the knowledge of the decline in female-perpetrated intimate

homicide, but still leaves explanations behind the decline in male-perpetrated unclear. Research

needs to continue to examine gender-specific intimate partner homicide, but particularly male-

perpetrated intimate partner homicide. Surprisingly, none of the three theoretical perspectives

were associated with male-perpetrated intimate partner homicide over time, whereas some of the

predictors were associated with female-perpetrated intimate partner homicide over time. It is

extremely important to gain a better understanding of predictors of male-perpetrated intimate

partner homicide, given the fact that male-perpetrated intimate partner homicide has not decreased









as sharply as female-perpetrated intimate partner homicide over time despite the increase in

domestic violence resources geared towards preventing violence against females. It is essential to

determine what really has influenced male-perpetrated intimate partner homicide over time. This

may be the only way to for us to see a larger decline in these types of homicides.









APPENDIX A
GLOSSARY

Place Places, for the reporting of decennial census data, include census designated places,

consolidated cities, and incorporated places.

South Region

S.,ntth Atlantic Division:

Delaware, Maryland, District of Columbia, Virginia, West Virginia, North Carolina, South

Carolina, Georgia, Florida

East .Sitll Central Division:

Kentucky, Tennessee, Alabama, Mississippi

West .Snlt Central Division:

Arkansas, Louisiana, Oklahoma, Texas

Employed All civilians 16 years old and over who were either (1) "at work"- those who did

any work at all during the reference week as paid employees, worked in their own business or

profession, worked on their own farm, or worked 15 hours or more as unpaid workers on a

family farm or in a family business; or (2) were "with a job but not at work"- those who did

not work during the reference week, but who had jobs or businesses from which they were

temporarily absent because of illness, bad weather, industrial dispute, vacation, or other personal

reasons. Excluded from the employed are people whose only activity consisted of work around

their own house (painting, repairing, or own home housework) or unpaid volunteer work for

religious, charitable, and similar organizations. Also excluded are all institutionalized people and

people on active duty in the United States Armed Forces.

Unemployed All civilians 16 years old and over were classified as unemployed if they were

neither "at work" nor "with a job but not at work" during the reference week, were looking for









work during the last 4 weeks, and were available to start ajob. Also included as unemployed

were civilians 16 years old and over who: did not work at all during the reference week, were on

temporary layoff from a job, had been informed that they would be recalled to work within the

next 6 months or had been given a date to return to work, and were available to return to work

during the reference week, except for temporary illness.

Civilian labor force Consists of people classified as employed or unemployed in accordance

with the criteria described above.

Labor force All people classified in the civilian labor force (i.e., "employed" and

"unemployed" people), plus members of the U.S. Armed Forces (people on active duty with the

United States Army, Air Force, Navy, Marine Corps, or Coast Guard).

Not in labor force All people 16 years old and over who are not classified as members of the

labor force. This category consists mainly of students, individuals taking care of home or family,

retired workers, seasonal workers enumerated in an off-season who were not looking for work,

institutionalized people (all institutionalized people are placed in this category regardless of any

work activities they may have done in the reference week), and people doing only incidental

unpaid family work (fewer than 15 hours during the reference week).

Household A household includes all of the people who occupy a housing unit. (People not

living in households are classified as living in group quarters.) A housing unit is a house, an

apartment, a mobile home, a group of rooms, or a single room occupied (or if vacant, intended

for occupancy) as separate living quarters

Householder The data on relationship to householder were derived from the question, "How is

this person related to Person 1," which was asked of Persons 2 and higher in housing units. One

person in each household is designated as the householder (Person 1). In most cases, the









householder is the person, or one of the people, in whose name the home is owned, being bought,

or rented. If there is no such person in the household, any adult household member 15 years old

and over could be designated as the householder (i.e., Person 1). Households are classified by

type according to the sex of the householder and the presence of relatives. Two types of

householders are distinguished: family householders and nonfamily householders. A family

householder is a householder living with one or more individuals related to him or her by birth,

marriage, or adoption. The householder and all of the people in the household related to him or

her are family members. A nonfamily householder is a householder living alone or with

nonrelatives only.

Unmarried partner An unmarried partner is a person who is not related to the householder,

who shares living quarters, and who has a close personal relationship with the householder.

Unmarried-Partner Household An unmarried-partner household is a household that includes

a householder and an "unmarried partner." An "unmarried partner" can be of the same or of

the opposite sex of the householder. An "unmarried partner" in an "unmarried-partner

household" is an adult who is unrelated to the householder, but shares living quarters and has a

close personal relationship with the householder. An unmarried-partner household may also be a

family household or a nonfamily household, depending on the presence or absence of another

person in the household who is related to the householder. There may be only one unmarried-

partner per household, and an unmarried partner may not be included in a married-couple

household as the householder cannot have both a spouse and an unmarried partner.

Income of individuals Income for individuals is obtained by summing the eight types of

income for each person 15 years old and over. The characteristics of individuals are based on the

time of enumeration (April 1, 2000), even though the amounts are for calendar year 1999.









Median income The median divides the income distribution into two equal parts: one-half of

the cases falling below the median income and one-half above the median. For households and

families, the median income is based on the distribution of the total number of households and

families including those with no income. The median income for individuals is based on

individuals 15 years old and over with income. Median income for households, families, and

individuals is computed on the basis of a standard distribution (see the "Standard Distributions"

section under "Derived Measures"). Median income is rounded to the nearest whole dollar.

Median income figures are calculated using linear interpolation if the width of the interval

containing the estimate is $2,500 or less. If the width of the interval containing the estimate is

greater than $2,500, Pareto interpolation is used. (For more information on medians and

interpolation, see "Derived Measures.")

Divorced -This category includes people who are legally divorced and who have not remarried.

How Poverty Status is Determined -

The poverty status of families and unrelated individuals in 1999 was determined using 48

thresholds (income cutoffs) arranged in a two dimensional matrix. The matrix consists of family

size (from 1 person to 9 or more people) cross-classified by presence and number of family

members under 18 years old (from no children present to 8 or more children present). Unrelated

individuals and 2-person families were further differentiated by the age of the reference person

(RP) (under 65 years old and 65 years old and over).

To determine a person's poverty status, one compares the person's total family income with the

poverty threshold appropriate for that person's family size and composition (see table below). If

the total income of that person's family is less than the threshold appropriate for that family, then

the person is considered poor, together with every member of his or her family. If a person is not









living with anyone related by birth, marriage, or adoption, then the person's own income is

compared with his or her poverty threshold.

Individuals for whom poverty status is determined Poverty status was determined for all

people except institutionalized people, people in military group quarters, people in college

dormitories, and unrelated individuals under 15 years old. These groups also were excluded from

the numerator and denominator when calculating poverty rates. They are considered neither

"poor" nor nonpoorr." (Census Glossary)

Residence 5 years ago The data on residence 5 years earlier were derived from answers to

long-form questionnaire Item 15, which was asked of a sample of the population 5 years old and

over. This question asked for the state (or foreign country), U.S. county, city or town, and ZIP

Code of residence on April 1, 1995, for those people who reported that on that date they lived in

a different house than their current residence. Residence 5 years earlier is used in conjunction

with location of current residence to determine the extent of residential mobility of the

population and the resulting redistribution of the population across the various states,

metropolitan areas, and regions of the country.

Summary File 1 This Census 2000 file presents 100-percent population and housing data for

the total population, for 63 race categories, and for many other race and Hispanic or Latino

categories. The data include age, sex, households, household relationship, housing units, and

tenure (whether the residence is owned or rented). Also included are selected characteristics for a

limited number of race and Hispanic or Latino categories. The data are available for the U.S.,

census regions, census divisions, states and statistically equivalent entities, counties and

statistically equivalent entities, county subdivisions, places, census tracts, block groups, census

blocks, metropolitan areas, urban areas, American Indian and Alaska Native areas, tribal









subdivisions, Hawaiian home lands, Congressional districts, and ZIP Code tabulation areas. Data

are available down to the block level for many tabulations, but only to the census tract level for

others. Available on CD-ROM, DVD, and American FactFinder.

Summary File 3 This Census 2000 file presents data on population and housing long-form

subjects, such as income and education. It includes population totals for ancestry groups. It also

includes selected characteristics for a limited number of race and Hispanic or Latino categories.

The data are available for the U.S., census regions, census divisions, states and statistically

equivalent entities, counties and statistically equivalent entities, county subdivisions, places,

census tracts, block groups, metropolitan areas, urban areas, American Indian and Alaska Native

areas, tribal subdivisions, Hawaiian home lands, Congressional districts, and ZIP Code

tabulation areas. Available on CD-ROM, DVD, and American FactFinder.

UCR Uniform Crime Reports









APPENDIX B
CORRELATION MATRIX FOR EXPLANATORY AND OUTCOME VARIABLE 1990.











1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1. Female-perp 1.00
IPH


2. % Fem poverty
3. % Fem unempl
4. % HH pub assis.
5. M/F income
6. M/F education
7. M/F % employ
8. % fem div
9. % unmarried
10. Shelter rate
11. Legal serve rate
12. Male batt. rate
13. Referral rate
14. Res. mobility
15. Hispanic(log)
16. Officer rate
17. South

1. Male-perp IPH
2. % Male poverty
3. % Male unempl
4. % HH pub assis.
5. M/F income
6. M/F education
7. M/F % employ
8. % Male div
9. % Unmarried
10. Shelter rate
11. Legal Serv rate
12. Male Batt. rate
13. Referral rate
14. Res. mobility
15. Hispanic(log)
16. Officer rate
17. South


.359 1.00
.300 .782 1.00
.244 .786 .793 1.00
-.221 -.282 -.093 -.265 1.00
-.190 -.393 -.328 -.364 .429 1.00
-.003 .181 .249 .212 .220 .208 1.00
.005 -.106 -.124 -.041 -.076 .050 -.126 1.00
.003 -.020 .053 .124 -.236 -.084 -.105 -.122 1.00
-.281 .024 -.107 -.026 .118 -.039 -.085 .036 -.129 1.00
-.223 .093 .011 .026 .078 -.017 -.068 .079 -.086 .678 1.00
-.078 .052 -.048 .016 -.077 -.178 -.071 .127 -.118 .293 .316 1.00
-.115 -.048 .084 .034 .034 -.046 -.072 .025 .017 .164 -.009 .133 1.00
-.244 -.460 -.431 -.442 .046 .111 -.226 .213 .225 .028 .016 .009 .079 1.00
-.058 -.125 .110 .072 -.159 .032 .152 .109 .198 -.126 -.105 -.045 .077 .279 1.00
.347 .467 .305 .388 -.403 -.345 -.146 -.132 -.004 -.014 -.044 .039 -.049 -.449 -.167 1.00
.183 .189 .045 -.195 .014 -.083 -.041 -.189 -.234 -.008 .010 .074 -.067 -.025 -.262 .105 1.00

1.00
.260 1.00
.199 .786 1.00
.195 .766 .863 1.00
-.250 -.322 -.193 -.265 1.00
-.177 -.405 -.335 -.364 .429 1.00
.006 .192 .218 .212 .220 .208 1.00
.050 .001 .024 .029 -.094 -.009 -.152 1.00
.004 -.021 .063 .124 -.236 -.084 -.105 -.107 1.00
-.280 .011 -.053 -.026 .118 -.039 -.085 .054 -.129 1.00
-.245 .091 .050 .026 .078 -.017 -.068 .079 -.086 .678 1.00
-.039 .062 -.018 .016 -.077 -.178 -.071 .137 -.118 .293 .316 1.00
-.121 -.033 .063 .034 .034 -.046 -.072 .005 .017 .164 -.009 .113 1.00
-.227 -.389 -.537 -.442 .046 .111 -.226 .138 .225 .028 .016 .009 .079 1.00
.018 -.041 .075 .072 -.159 .032 .152 .014 .198 -.126 -.105 -.045 .077 .279 1.00
.323 .426 .444 .388 -.403 -.345 -.146 -.048 -.004 -.014 -.044 .039 -.049 -.449 -.167 1.00
.122 .151 -.160 -.195 .014 -.083 -.041 -.188 -.234 -.008 .010 .074 -.067 -.025 -.262 .105 1.00









APPENDIX C
CORRELATION MATRIX FOR EXPLANATORY AND OUTCOME VARIABLES 2000.





1. Female-perp IPH
2. % Fem poverty
3. % Fem unempl
4. % HH pub assis.
5. M/F income
6. M/F education
7. M/F % employ
8. % fem div
9. % unmarried
10. Shelter rate
11. Legal serve rate
12. Male batt. rate
13. Referral rate
14. Res. mobility
15. Hispanic(log)
16. Officer rate
17. South

1. Male-perp IPH
2. % Male poverty
3. % Male unempl
4. % HH pub assis.
5. M/F income
6. M/F education
7. M/F % employ
8. % Male div
9. % Unmarried
10. Shelter rate
11. Legal serve rate
12. Male batt. rate
13. Referral rate
14. Res. mobility
15. Hispanic(log)
16. Officer rate
17. South


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1.00
.257 1.00
.219 .816 1.00
.147 .751 .751 1.00
-.245 -.256 -.200 -.178 1.00
-.148 -.365 -.365 -.326 .476 1.00
-.124 -.201 -.182 -.122 .396 .535 1.00
.032 -.061 -.146 -.145 -.009 -.038 -.245 1.00
.067 .339 .382 .537 -.294 -.358 -.267 .209 1.00
-.224 .089 .095 .115 -.012 -.079 -.182 .010 .150 1.00
-.197 .030 .033 .072 -.092 -.058 -.168 -.008 .124 .709 1.00
-.100 .018 -.004 -.015 -.021 -.123 -.055 -.021 -.014 .347 .431 1.00
-.137 .047 .065 -.002 .109 -.035 .065 -.034 -.060 .047 -.253 .086 1.00
-.084 -.164 -.155 -.251 -.289 .004 -.122 .162 .117 .008 .058 .019 -.013 1.00
-.058 -.061 .085 .177 -.053 .061 .477 -.345 .133 -.182 -.191 -.008 .108 .153 1.00
.311 .482 .425 .353 -.424 -.224 -.372 -.011 .152 .004 .072 .000 -.103 -.210 -.282 1.00
.096 .099 .001 -.293 -.019 -.023 -.078 .070 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00

1.00
.210 1.00
.126 .788 1.00
.151 .757 .816 1.00
-.226 -.323 -.218 -.178 1.00
-.105 -.370 -.405 -.326 .476 1.00
-.007 -.124 -.233 -.122 .396 .535 1.00
-.013 .000 .131 -.009 -.034 -.123 -.348 1.00
.048 .371 .484 .537 -.294 -.358 -.267 .302 1.00
-.327 .054 .130 .115 -.012 -.079 -.182 .136 .150 1.00
-.241 .012 .062 .072 -.092 -.058 -.168 .085 .124 .709 1.00
-.105 .003 -.010 -.015 -.021 -.123 -.055 .022 -.014 .347 .431 1.00
-.143 .040 .066 -.002 .109 -.035 .065 .039 -.060 .047 -.253 .086 1.00
-.099 -.081 -.195 -.251 -.289 .004 -.122 .065 .117 .008 .058 .019 -.013 1.00
.071 .080 .012 .177 -.053 .061 .477 -.434 .133 -.182 -.191 -.008 .108 .153 1.00
.338 .411 .435 .353 -.424 -.224 -.372 .125 .152 .004 .072 .000 -.103 -.210 -.282 1.00
.005 .026 -.154 -.293 -.019 -.023 -.078 .040 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00









APPENDIX D
SIMPLIFIED CORRELATION MATRIX FOR EXPLANATORY AND OUTCOME
VARIABLES 1990.














1. Female-perp IPH
2. Econ. dep. index
3. M/F income
4. M/F education
5. M/F % employment
6. % Fem divorced
7. % Unmarried HH
8. Shelter rate
9. Legal service rate
10. Male batt. rate
11. Referral rate
12. Res. mobility
13. % Hispanic (log)
14. Office rate
15. South

1. Male-perp IPH
2. Econ. dep. index
3. M/F income
4. M/F education
5. M/F % employment
6. % Male divorced
7. % Unmarried HH
8. Shelter rate
9. Legal Service rate
10. Male batt. rate
11. Referral rate
12. Res. mobility
13. % Hispanic (log)
14. Office rate
15. South


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1.00
.359 1.00
-.221 -.241 1.00
-.190 -.393 .429 1.00
-.003 .209 .220 .208 1.00
.005 -.116 -.076 .050 -.126 1.00
.003 .000 -.236 -.084 -.105 -.122 1.00
-.281 -.012 .118 -.039 -.085 .036 -.129 1.00
-.223 .074 .078 -.017 -.068 .079 -.086 .678 1.00
-.078 .026 -.077 -.178 -.071 .127 -.118 .293 .316 1.00
-.115 -.012 .034 -.046 -.072 .025 .017 .164 -.009 .133 1.00
-.244 -.473 .046 .111 -.226 .213 .225 .028 .016 .009 .079 1.00
-.058 -.063 -.159 .032 .152 .109 .198 -.126 -.105 -.045 .077 .279 1.00
.347 .442 -.403 -.345 -.146 -.132 -.004 -.014 -.044 .039 -.049 -.449 -.167 1.00
.183 .156 .014 -.083 -.041 -.189 -.234 -.008 .010 .074 -.067 -.025 -.262 .105 1.00

1.00
.254 1.00
-.250 -.298 1.00
-.177 -.403 .429 1.00
.006 .210 .220 .208 1.00
.050 .016 -.094 -.009 -.152 1.00
.004 .004 -.236 -.084 -.105 -.107 1.00
-.280 -.008 .118 -.039 -.085 .054 -.129 1.00
-.245 .083 .078 -.017 -.068 .079 -.086 .678 1.00
-.039 .040 -.077 -.178 -.071 .137 -.118 .293 .316 1.00
-.121 -.005 .034 -.046 -.072 .005 .017 .164 -.009 .133 1.00
-.227 -.454 .046 .111 -.226 .138 .225 .028 .016 .009 .079 1.00
.018 -.007 -.159 .032 .152 .014 .198 .198 -.105 -.045 .077 .279 1.00
.323 .452 -.403 -.345 -.146 -.048 -.004 -.004 -.044 .039 -.049 -.449 -.167 1.00
.122 .061 .014 -.083 -.041 -.188 -.234 -.34 .010 .074 -.067 -.025 -.262 .105 1.00









APPENDIX E
SIMPLIFIED CORRELATION MATRIX FOR EXPLANATORY AND OUTCOME
VARIABLES 2000.












1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. Female-Perp IPH 1.00
2. Econ. Dep. Index .229 1.00
3. M/F Income -.245 -.237 1.00
4. M/F Education -.148 -.380 .476 1.00
5. M/F % Employment -.124 -.184 .396 .535 1.00
6. %Fem Divorced .032 -.114 -.009 -.038 -.245 1.00
7.% Unmarried HH .067 .449 -.294 -.358 -.267 .209 1.00
8. Shelter Rate -.224 .107 -.012 -.079 -.182 .010 .150 1.00
9. Legal Service Rate -.197 .049 -.092 -.058 -.168 -.008 .124 .709 1.00
10. Male Batt. Rate -.100 .003 -.021 -.123 -.055 -.021 -.014 .347 .431 1.00
11. Referral Rate -.137 .036 .109 -.035 .065 -.034 -.060 .047 -.253 .086 1.00
12. Res. Mobility -.084 -.208 -.289 .004 -.122 .162 .117 .008 .058 .019 -.013 1.00
13.% Hispanic (log) -.058 .052 -.053 .061 .477 -.345 .133 -.182 -.191 -.008 .108 .153 1.00
14. Office Rate .311 .462 -.424 -.224 -.372 -.011 .152 .004 .072 .000 -.103 -.210 -.282 1.00
15. South .096 -.059 -.019 -.023 -.078 .070 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00
c
1. Male-Perp IPH 1.00
2. Econ. Dep. Index .180 1.00
3. M/F Income -.226 -.260 1.00
4. M/F Education -.105 -.390 .476 1.00
5. M/F % Employment -.007 -.161 .396 .535 1.00
6. % Male Divorced -.013 .030 -.034 -.123 -.348 1.00
7. % Unmarried HH .048 .500 -.294 -.358 -.267 .302 1.00
8. Shelter Rate -.327 .104 -.012 -.079 -.182 .136 .150 1.00
9. Legal Service Rate -.241 .051 -.092 -.058 -.168 .085 .124 .709 1.00
10. Male Batt. Rate -.105 -.008 -.021 -.123 -.055 .022 -.014 .347 .431 1.00
11. Referral Rate -.143 .032 .109 -.035 .065 .039 -.060 .047 -.253 .086 1.00
12. Res. Mobility -.099 -.189 -.289 .004 -.122 .065 .117 .008 .058 .019 -.013 1.00
13.% Hispanic (log) .071 .110 -.053 .061 .477 -.434 .133 -.182 -.191 -.008 .108 .153 1.00
14. Office Rate .338 .426 -.424 -.224 -.372 .125 .152 .004 .072 .000 -.103 -.210 -.282 1.00
15. South .005 -.154 -.019 -.023 -.078 .040 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00











APPENDIX F
VARIANCE INFLATION FACTORS FOR SIMPLIFIED MODELS

Fem-Perp90 Male-Perp90 Fem-PerpOO


Female economic deprivation index

Male economic deprivation index

Ratio of male to female education

Ratio of male to female income

Ratio of male to female employment

Percent female divorced

Percent male divorced

Percent unmarried households

Male batterers programs per 100,000 males

Shelters per 100,000 females

Legal service programs per 100,000 females

Referral services per 100,000 females

Percent hispanic (log)

Residential mobility

Officer rate per 1,000

South


1.900 2.028


1.549

1.615

1.438

1.226



1.343

1.226

2.056

2.056

1.108

1.348

1.908

1.780

1.237


1.987

1.546

1.614

1.457



1.197

1.314

1.227

2.054

2.061

1.111

1.361

1.890

1.789

1.242


2.068

2.039

2.369

1.450



2.799

1.397

2.391

2.864

1.290

2.078

1.634

2.076

1.959


Male-PerpOO


2.006

2.084

2.034

2.381



1.623

2.908

1.397

2.385

2.855

1.308

2.176

1.616

2.026

1.891










APPENDIX G
CORRELATION MATRICES FOR DOMESTIC VIOLENCE SERVICE VARIABLES.

1990 1 2 3 4 5 6 7
1. Shelters 1.00
2. Hotlines .906 1.00
3. Counseling .906 .995 1.00
4. Children counseling .877 .920 .922 1.00
5. Legal services .646 .723 .736 .661 1.00
6. Batterers counseling .392 .419 .420 .422 .283 1.00
7. Number of referrals .328 .413 .418 .301 .101 -.241 1.00


2000 1 2 3 4 5 6 7
1. Shelters 1.00
2. Hotlines .904 1.00
3. Counseling .900 .981 1.00
4. Children counseling .923 .946 .956 1.00
5. Legal services .851 .921 .959 .933 1.00
6. Batterers counseling .630 .702 .734 .715 .733 1.00
7. Number of referrals -.163 -.144 -.153 -.191 -.246 -.095 1.00









REFERENCES


Agresti, Allen. 1996. An Introduction to Categorical Data Analysis. New York: John Wiley and
Sons.

Allard, Mary Ann, Randy Albelda, Mary Ellen Colten, and Carol Cosenza. 1997. In Harms
Way? Domestic Violence, AFDC Receipt, and Welfare Reform in Massachusetts.
Boston, MA: University of Massachusetts.

Allison, Paul D. 1994. Using panel data to estimate the effects of events. Sociological Methods
and Research 23: 179-199.

Anderson, Kristin L. 1997. Gender, status, and domestic violence: An integration of feminist and
family violence approaches. Journal ofMarriage and the Family 59:655-669.

Archer, John. 2000. Sex differences in aggression between heterosexual partners: A Meta-
analytic review. Psychological Bulletin 126:651-680.

Avakeme, Edem F. 1999. Females' labor force participation and intimate femicide: An empirical
assessment of the backlash hypothesis. Violence and Victims 14:277-291.

Bailey, William C. 1984. Poverty, inequality, and city homicide rates: Some not so unexpected
results. Criminology 22:531-550.

Bailey, William C., and Ruth D. Peterson. 1995. Gender inequality and violence against women:
The case of murder. In Crime and Inequality, eds. John Hagan and Ruth D. Peterson.
Stanford, CA: Stanford University Press.

Baron, Larry, and Murray A. Straus. 1987. Legitimate violence, violent Attitudes, and rape: A
test of the cultural spillover theory. Annals of the New York Academy of Sciences
528:79-110.

Berk, Richard A., Sarah F. Berk, Donileen R., Loseke, and David Rauma. 1983. Mutual combat
and other family violence. In The dark side offamilies: Current family violence
research, eds. David Finkelhor, Richard J. Gelles, Gerald T. Hotaling, & Murray A.
Staus. Beverly Hills, CA: Sage.

Barnard, George W, Hernan Vera, Maria I. Vera, and Gustave Newman. 1982. Till death do us
part: A study of spouse murder. Bulletin of the American Academy of Psychiatry and
the Law 10:271-280.

Blau, Judith R., and Peter M. Blau. 1982. The cost of inequality: Metropolitan structure and
violent crime. American Sociological Review 47:114-129.

Block, Carolyn R., and Anigone Christakos. 1995. Intimate partner homicide in Chicago over 29
Years. Crime & Delinquency 41:496-526.









Block, Carolyn R. 2000. The Chicago Women's Health Risk Study. Risk of Serious Injury or
Death in Intimate Violence: A Collaborative Research Project. Final Report to the
National Institute of Justice. Available at
http://www.icjia.state.il.us/public/pdf/cwhrs/cwhrs.pdf.

Boritch, Helen, and John Hagan. 1990. A century of crime in Toronto: Gender, class, and
patterns of social control, 1859 to 1955. Criminology 28:567-600.

Box, Steven, and Chris Hale. 1983. Liberation and female criminality in England and Wales.
British Journal of Criminology 23:35-49.

Box, Steven, and Chris Hale. 1984. Liberation/Emancipation, economic marginalization, or less
chivalry: The relevance of three theoretical arguments to female crime patterns in
England and Wales, 1951-1980. Criminology 22:473-497.

Braithwaite, John. 1979. Inequality, Crime, and Public Policy. London: Routledge & Kegan
Paul.

Browne, Angela. 1985. Assault and homicide at home: When battered women kill. In Advances
in Applied Social Psychology, vol. 3, eds. Michael J. Saks and Leonard Saxe.
Hillsdale, NJ: Lawrence Erlbaum.

Browne Angela. 1986. Assault and homicide at home: When battered women kill. In Advances
in Applied Social Psychology, eds. Michael J. Saks and Leonard Saxe. Hillsdale, NJ:
Erlbaum.

Browne, Angela. 1987. When Battered Women Kill. New York: Free Press.

Browne, Angela, and Robert L. Flewelling. 1986. Women as victims or perpetrators of homicide.
Presented at the Annual Meeting of the American Society of Criminology, Atlanta,
October 29-November 1.

Browne, Angela, and Kirk R. Williams. 1989. Exploring the effect of resource availability and
the likelihood of female-perpetrated homicides. Law and Society Review 23:75-94.

Browne, Angela, and Kirk R. Williams. 1993. Gender, intimacy, and lethal violence. Gender &
Society 7:78-98.

Browne, Angela, Kirk R. Williams, and Donald G. Dutton. 1999. Homicide between intimate
partners: A 20-year review. In Homicide: A Sourcebook of Social Research, eds. M.
Dwayne Smith and Margaret A. Zahn. Thousand Oaks, CA: Sage.

Brush, Lisa D. 1990. Violent acts and injurious outcomes in married couples: Methodological
issues in the National Survey of Families and Households. Gender & Society 4:56-67.

Byrne, James M., and Robert J. Sampson. 1986. The Social Ecology of Crime. New York:
Springer-Verlag.









Bureau of Justice Statistics. 2006. Homicide trends in the United States. Washington, DC.
Available at www.ojp.usdoj.gov/bj s/homicide/homtrnd.htm

Bureau of Labor Statistics. 2008a. Employment status of the civilian noninstitutional population
16 years and over by sex, 1973 to date. Washington, DC. Available at
www.bls.gov/CES

Bureau of Labor Statistics. 2008b. Employment status of the civilian noninstitutional population,
1942 to date. Washington, DC. Availabile at www.bls.gov/CES

Bursik, Robert J. Jr. 1988. Social disorganization theories of crime and delinquency: Problems
and prospects. Criminology 26:519-551.

Cameron, Adrian C., and Pravin K. Trivedi. 1998. Regression Analysis of Count Data.
Cambridge University Press, Cambridge.

Campbell, Jacquelyn C. 1992. 'If I can't have you, no one can:' Power and control in homicide
of female partners. In Femicide: The Politics of Women Killing, eds. Jill Radford and
Diana E. H. Russell. New York: Twayne Publishers.

Cazenave, Noel A., and Margaret A. Zahn. 1992. Women, murder and male dominance: Police
reports of domestic violence in Chicago and Philadelphia. In Intimate Violence:
Interdisciplinary Perspectives, eds. E. C. Viano. Washington, DC: Hemisphere.

Chilton, Roland, and Dee Weber. 2000. Uniform Crime Reporting Program [United States]:
Arrests by Age, Sex, and Race for Police Agencies in Metropolitan Statistical Areas,
1960-1997 [Computer file]. 2nd ICPSR version. Amherst, MA: University of
Massachusetts [producer]. Ann Arbor, MI: Inter-university Consortium for Political
and Social Research [distributor].

Chimbos, Peter D. 1978. Marital Violence: A Study ofInterpersonal Homicide. San Francisco: R
& E Associates.

Cloward, Richard A., and Lloyd Ohlin. 1960. Delinquency and Opportunity: A Theory of
Delinquent Gangs. New York: Free Press.

Cohen, Albert K. 1955. Delinquent Boys: The Culture of the Gang. New York: Free Press.

Crowell, Nancy A., and Ann W. Burgress. 1996. Understanding Violence Against Women.
Washington, DC: Research Council National Academy Press.

Daly, Martin, and Margo Wilson. 1988. Homicide. New York: Aldine.

Daniel, A. E., and P.W. Harris. 1982. Female homicide offenders referred for pre-trial
psychiatric examination: A descriptive study. Bulletin of the American Academy of
Psychiatry and Law 10: 261-269.









DeWees Mari A., and Karen F. Parker. 2003. Women, region, and types of homicide. Homicide
Studies 7:368-393.

Dobash, Russell P., R. Emerson Dobash, Margo Wilson, and Martin Daly. 1992. The myth of
sexual symmetry in marital violence. SocialProblems 39:401-421.

Dugan, Laura, Daniel S. Nagin, and Richard Rosenfeld. 1999. Explaining the decline in intimate
partner homicide: The effects of changing domesticity, women's status, and domestic
violence resources. Homicide Studies 3:187-214.

Dugan, Laura, Daniel S. Nagin, and Richard Rosenfeld. 2003. Exposure reduction or retaliation?
The effects of domestic violence resources on intimate-partner homicide. Law &
Society Review 37:169-198.

Durose, Matthew R., Carolyn Wolf Harlow, Patrick Langan, Mark Motivans, Ramona R.
Rantana, and Erica L. Smith. 2005. Family Violence Statistics. Washington D.C.:
Department of Justice, Bureau of Justice Statistics.

Ellis, Lee, and Charles Beattie. 1983. The feminist explanation for rape: An empirical test.
Journal of Sex Research 19:74-93.

Fagan, Jeffrey, and Angela Browne. 1994. Violence between spouses and intimates: Physical
aggression between women and men in intimate relationships. In Understanding and
Preventing Violence. Volume 3: Social influences, eds. Albert J. Reiss, Jr., & Jeffrey
A. Roth. Washington, DC: National Academy Press.

Felson, Richard B. 2006. Is violence against women about women or about violence? Contexts
5:21-25.

Flewelling, Robert L., and Kirk R. Williams. 1999. Categorizing homicides: The use of
disaggregated data in homicide research. In Homicide: A Sourcebook of Social
Research, eds. M. Dwayne Smith and Margaret A. Zahn. Thousand Oaks, CA: Sage.

Fox, James Allan. 2005. Uniform Crime Reports [United States]: Supplementary Homicide
Reports, 1976-2002 [Computer file]. Compiled by Northeastern University, College of
Criminal Justice. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political
and Social Research [producer and distributor], 2005.

Fox, James Allan. 2004. Missing data problems in the SHR. Homicide Studies 8:214-254.

Fox, James Alan, and Marianne W. Zawitz. 2006. Homicide Trends in the United States.
Washington, DC: Bureau of Justice Statistics. Available at www.ojp.usdoj.gov/bjs/

Gallup-Black, Adria. 2005. Twenty years of rural and urban trends in family and intimate partner
homicide. Homicide Studies 9:149-173.









Gardner, William, Edward P. Mulvey, & Esther C. Shaw. 1995. Regression analyses of counts
and rates: Poisson, overdispersed poisson, and negative binomial models
Psychological Bulletin 118:392-404.

Gartner, Rosemary, Kathryn Baker, and Fred C. Pampel. 1990. Gender stratification and the
gender gap in homicide victimization. Social Problems 37:593-612.

Gauthier, DeeAnn K., and William B. Bankston. 1997. Gender equality and the sex ratio of
intimate killing. Criminology 35:577-600.

Greene, William H. 2000. Econometric Analysis. Macmillin. Goetting, Ann. 1988. Patterns of
homicide among women. Journal ofInterpersonal Violence 3:3-19.

Goetting, Ann. 1995. Homicide in Families and Other Special Populations. New York: Springer.

Greenberg, Stephanie W.,William M. Rohe, and Jay R. Williams. 1982. Safe and Secure
Neighborhoods: Physical Characteristics and Informal Territorial Control in High
andLow Crime Neighborhoods. Washington, DC: National Institute of Justice.

Greenburg, David F., and Valerie West. 2001. State prison populations and their growth, 1971-
1991. Criminology 39: 615-653.

Greenfield, Laurence A., Michael R. Rand, Diane Craven, Patsy A. Klaus, Craig A. Perkins,
Cherly Ringel, Greg Warchol, Cathy Maston, and Fox, James Allan. 1998. Violence by
Intimates: Analysis ofData on Crimes by Current or Former Spouses, Boyfriends, and
Girlfriends (NCJ-167237). Washington, DC: U.S. Department of Justice.

Harris, Anthony R., Stephen H. Thomas, Gene A. Fisher, and David J. Hirsch. 2002. Murder
and medicine: The lethality of criminal assault 1960-1999. Homicide Studies 6:128-
166.

Hausman, Jerry, Bronwyn H. Hall, and Zvi Griliches. 1984. Econometric models for count data
with an application to the patents-R & D relationship. Econometrica 52:909-938.

Heimer, Karen. 2000. The nature of crime: Continuity and change. Changes in the gender gap in
crime and women's economic marginalization. Criminal Justice 1:427-483.

Heimer, Karen, Stacy M. Wittrock, and Halime Unal. 2005. Gender, crime and the economic
marginalization of women. In Gender and Crime: Patterns of Victimization and
Offending, eds. Karen Heimer and Candace Kruttschitt. New York: New York
University Press.

Hunnicutt, Gwen, and Lisa M. Broidy. 2004. Liberation and economic marginalization: A
reformulation and test of (formerly?) competing models. Journal of Research in Crime
and Delinquency 4:130-155.









Johnson, David R. 1995. Alternative methods for the quantitative analysis of panel data in family
research: Pooled time-series models. Journal ofMarriage and the Family 57:1065-
1077.

Junger-Tas, Josine and Ineke Haen Marshall. 1999. The self-report methodology in crime
research. In Crime and Justice: A Review of the Research, ed. Michael Tonry.
Chicago: University Press.

Jurik, Nancy C., and Russ Winn .1990. Gender and homicide: a comparison of men and women
who kill. Violence and Victims 5:227-242.

Kaukinen, Catherine. 2004. Status compatibility, physical violence, and emotional abuse in
intimate relationships. Journal ofMarriage and Family 66:452-471.

Land, Kenneth C., Patricia L. McCall, and Lawrence E. Cohen. 1990. Structural covariates of
homicide rates: Are there any invariances across time and social space? American
Journal of Sociology 95:922-963.

Langan, Patrick A., and Christopher A. Innes. 1986. Preventing Domestic Violence Against
Women. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Maddala, G. S. 1983. Limited-dependent and Qualitative Variables in Econometrics. New York:
Cambridge University Press.

Makepeace, James M. 1989. Dating, living together, and courtship violence. In Violence in
Dating Relationships: Emerging Social Issues, eds. Maureen. A. Pirog-Good and Jan
E. Stets. New York: Praeger.

Makepeace, James M. 1997. Courtship violence as process: A developmental theory. In Violence
Between Intimate Partners: Patterns, Causes, and Effect, ed. Albert P. Cardarelli.
Boston: Allyn & Bacon.

McCall, Patricia L., and Karen F. Parker, K. F. 2005. A dynamic model of racial competition,
racial inequality, and interracial violence. Sociological Inquiry 75:273-293.

Merton, Robert K. 1968. Social Theory and Social Structure. New York: Free Press.

Messner, Scott. F., and Richard Rosenfeld. 1999. Social structure and homicide: Theory and
research. In Homicide Studies: A Sourcebook of Social Research, eds. M. Dwayne
Smith and Margaret Zahn. Thousand Oaks, CA: Sage.

Messner, Steven F., Glenn D. Deane, Luc Anselin, and Benjamin Pearson-Nelson. 2005.
Locating the vanguard in rising and falling homicide rates across U.S. cities.
Criminology 43:661-696.

National Coalition Against Domestic Violence. 1991. 1991 National Directory of Domestic
Violence Programs: A Guide to Community .\liheir, Safe Home and Service
Programs. Washington, DC: National Coalition Against Domestic Violence.









National Coalition Against Domestic Violence. 1999. 1999-2000 National Directory of Domestic
Violence Programs: A Guide to Community .\l/here, Safe Home and Service
Programs. Denver, CO: National Coalition Against Domestic Violence.

Nisbett, Richard E., and Dov Cohen. 1996. Culture ofHonor: The Psychology of Violence in the
Sn.,lh Boulder, CO: Westview.

Osgood, D. Wayne. 2000. Poisson-based regression analysis of aggregate crime rates. Journal of
Quantitative Criminology 16:21-43.

Osgood, D. Wayne, & Chambers, JeffF. 2000. Social Disorganization outside the metropolis:
An analysis of rural youth violence. Criminology 38:81-116.

Parker, Karen F. 2004. Industrial shift, polarized labor markets, and urban violence: Modeling
the dynamics between the economic transformation and disaggregated homicide.
Criminology 42:619-644.

Parker, Karen. F., Patricia L. McCall, and Kenneth C. Land. 1999. Determining social-structural
predictors of homicide. In Homicide: A Sourcebook of Social Research, eds. M.
Dwayne Smith and Margaret A. Zahn. Thousand Oaks, CA: Sage.

Paulsen, Derek J., and Victoria E. Brewer. 2000. The spousal SROK revisited: A comparison of
Chicago and Houston intimate partner homicide ratios. Gender Issues 18:88-100.

Peterson, Ericka S. L. 1999. Murder as self-help: Women and intimate partner homicide.
Homicide Studies 3:30-46.

Puzone, Carol, A., Linda E. Saltzman, Marcie-Jo M. Kresnow, Matie P. Thompson, and James
A. Mercy. 2000. National trends in intimate partner homicide. Violence Against
Women 6:409-426.

Reckdenwald, Amy, and Karen F. Parker. 2008. The influence of gender inequality and
marginalization and types of female offending. Homicide Studies 12:208-226.

Reidel Michael, and Joel Best. 1998. Patterns in intimate partner homicide. Homicide Studies
2:305-320.

Rodriquez, S. Fernando, and Veronica Henderson. 1995. Intimate homicide: victim-offender
relationship in female-perpetrated homicide. Deviant Behavior: An Interdisciplinary
Journal 16:45-57.

Rosenfeld, Richard. 1997. Changing relationships between men and women: A note on the
decline in intimate partner homicide. Homicide Studies 1:72-83.

Rosenfeld, Richard. 2000. Patterns in adult homicide: 1980-1995. In The Crime Drop in
America, eds. Alfred Blumstein and Joel Wallman. Cambridge, UK: Cambridge
University Press.









Russell, Diane E. 1975. The Politics ofRape: The Victim's Perspective. New York: Stein and
Day.

Saunders, Daniel, and Angela Browne. 2000. Intimate partner homicide. In Case Studies in
Family Violence, 2nd, eds. Robert T. Ammerman and Michael Hersen. New York:
Plenum Press.

Schechter, Susan. 1982. Women andMale Violence. Boston: South End Press.

Schwartz, Martin D. 1987. Gender and injury in spousal assault. SociologicalForum 20:61-75.

Shackelford, Todd K. 2001. Partner-killing by women in cohabiting relationships and marital
relationships. Homicide Studies 5:253-266.

Shackelford, Todd K., and Jenny Mouzos. 2005. Partner killing by men in cohabiting and marital
relationships. Journal ofInterpersonal Violence 20:1310-1324.

Sherman, Lawrence, and Richard Berk. 1984. The specific deterrent effects of arrest for
domestic assault. American Sociological Review 49: 261-272.

Silverman, Robert, A., and S. K. Mukherjee. 1987. Intimate homicide: An Analysis of violence
social relationships. Behavioral Sciences and the Law 5:37-47.

Smith, M. Dwayne, and Victoria R. Brewer. 1992. A sex-specific analysis of correlates of
homicide victimization in United States cities. Violence and Victims 4:279-86.

Smith, Paige Hall, Kathryn E. Moracco, and John D. Butts. 1998. Partner homicide in context.
Homicide Studies 2:400-421.

Smith, M. Dwayne, and Margaret A. Zahn. 1998. Homicide: A Sourcebook of Social Research.
Thousand Oaks, CA: Sage.

STATA. 2003. Cross-sectional time-series reference manual. STATA Press.

Steffensmeier, Darrell, and Dana Haynie. 2000a. Gender, structural disadvantage, and urban
crime: Do macrosocial variables also explain female offending rates? Criminology
38:403-438.

Steffensmeier, Darrell, and Dana Haynie. 2000b. The structural sources of urban female violence
in the United States. Homicide Studies 4:107-134.

Steffensmeier, Darrell, and Cathy Streifel. 1992. Time-series analysis of the female percentage
of arrests for property crimes, 1960-1985: A test of alternative explanations. Justice
Quarterly 9:77-103.









Stets, Jan E., and Murray A. Straus .1990. Gender differences in reporting marital violence and
its medical and psychological consequences. In Physical violence in American
families: Risk factors and adaptations to violence in 8,145families. Murray A. Straus
and Richard J. Gelles. New Brunswick, N.J.: Transaction.

Stout, Karen D., and Patricia Browne. 1995. Legal and social differences between men and
women who kill intimate partners. Affilia 10:194-205.

Stark, Evan, Ann Flitcraft, and William Frazier. 1979. Medicine and patriarchal violence: the
social construction of a "private" event. International Journal of Health Services
9:461-493.

Straus, Murray A. 1993. Physical assaults by wives: A major social problem. In Current
Controversies on Family Violence, eds. Richard J. Gelles and Donileen R. Loseke.
Thousand Oaks, CA: Sage.

Straus, Murray A., Richard J. Gelles, and Suzanne K. Steinmetz. 1980. Behind closed doors:
Violence in the American family. Doubleday, Garden City, NY.

Totman, Jane. 1978. The Murderess: A Psychological Study of Criminal Homicide. San
Francisco, CA: R & E Research Associates.

U.S. Bureau of the Census. 1994. Census of Population and Housing, 1990. Summary Tape File
la and 3a. Washington D.C.: U.S. Department of Commerce, Bureau of the Census,
1991.

U.S. Bureau of the Census. 2003. Census of Population and Housing, 2000. Summary Tape File
1 and 3. Washington, D.C.: U.S. Department of Commerce, Bureau of the Census,
2002.

U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting Program
Data [United States]: Police Employee (LEOKA) Data, 2000 [Computer file].
Compiled by the U.S. Dept. of Justice, Federal Bureau of Investigation. ICPSR ed.
Ann Arbor, MI: Inter-university Consortium for Political and Social Research
[producer and distributor], 2002.

Vieraitis, Lynne M., Marian R. Williams. 2002. Assessing the impact of gender inequality on
female homicide victimization across U.S. cities: A racially disaggregated analysis.
Violence Against Women 8:35-63.

Villarreal, Andreas. 2004. The Social Ecology of Rural Violence: Land Scarcity,
theOrganization of Agricultural Production, and the Presence of the State. American
Journal of Sociology 110:313-348.

Walby, Sylvia. 1986. Patriarchy at Work. Minneapolis, MN: University of Minnesota Press.

Whaley, Rachel Bridges, and Steven F. Messner. 2002. Gender equality and gendered
Homicides. Homicide Studies 6:188-210.









Williams, Kirk R., and Robert L. Flewelling. 1987. Family, acquaintance, and stranger homicide:
Alternative procedures for rate calculations. Criminology 25:543-560.

Williams, Kirk R., and Robert L. Flewelling 1988. The Social Production of Criminal Homicide:
A Comparative Study of Disaggregated Rates in American Cities. American
Sociological Review 53:421-431.

Williams, Joyce E., and Karen A. Holmes. 1981. The SecondAssault: Rape andPublic Attitudes.
Westport, CT: Greenwood.

Wilson Martin. 1989. Spousal conflict and violence in evolutionary perspective. In Sociobiology
and the Social Sciences, eds. Robert W. Bell and Nancey J. Bell. Texas Tech
University Press.

Wilson Margo, and Martin Daly. 1992a. Till death do us part. In Femicide: The Politics of
Woman Killing, eds. Jill Radford and Diana E. H. Russell. Boston MA: Twayne.

Wilson, Margo, and Martin Daly. 1992b. Who kills whom in spouse killings? On the exceptional
sex ratio of spousal homicides in the United States. Criminology 30:189-216.

Wilson, Margo, Martin Daly, and Christine Wright. 1993. Uxoricide in Canada: Demographic
risk patterns. Canadian Journal of Criminology 35:263-291.

Wilson, Margo, Holly Johnson, and Martin Daly. 1995. Lethal and nonlethal violence against
wives. Canadian Journal of Criminology 37:331-361.

Wolfgang, Marvin E. 1958. Patterns of Criminal Homicide. Philadelphia: University of
Pennsylvania.

Yang, Bijou, and David Lester. 1988. The participation of females in the labor force and rates of
personal violence (suicide and homicide). Suicide andLife-Threatening Behavior
18:270-78.









BIOGRAPHICAL SKETCH

Amy Reckdenwald grew up in Ithaca, New York. After graduating from high school she

moved to Pittsburgh, Pennsylvania where she obtained her bachelor's degree in psychology with

a double major in statistics from Carnegie Mellon University. After completing her

undergraduate degree in 2001 she moved to Tampa, Florida and later received her master's

degree in criminology from the University of South Florida in 2004. In 2008 she completed her

Ph.D. in criminology, law and society at the University of Florida in Gainesville, Florida. After

graduating from the University of Florida, Dr. Amy Reckdenwald decided to remain in Florida

and accepted a position as an Assistant Professor at Florida Atlantic University, in Boca Raton,

Florida.





PAGE 1

1 EXAMINING CHANGES IN MALE AND FE MALE INTIMATE PARTNER HOMICIDE OVER TIME, 1990-2000 By AMY RECKDENWALD A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Amy Reckdenwald

PAGE 3

3 ACKNOWLEDGMENTS First and forem ost, I would like to thank all of my family and friends for all their love and encouragement through my many year s of college. I honestly would never have finished without their endless support and I know they are as happ y as I am to see my graduate school career finally come to an end. I would also like to thank my entire dissertation committee, Dr. Karen Parker, Dr. Lonn Lanza Kaduce, Dr. Alex Piquero, and Dr. Barbara Zsembik, for all their support and suggestions in the writing of this diss ertation. I especially want to thank Dr. Karen Parker for all her help and guidance through my 4 years at UF as well as all the assistance she gave in the completion of this dissertation. She is a mentor as well as a good friend. I would also like to thank Dr. Richard Hollinger for sitting in on my defense last mi nute and giving me many good suggestions to improve this dissertation a nd Dr. Larry Winner for all his help with the statistical analysis.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3LIST OF TABLES................................................................................................................. ..........7LIST OF FIGURES.........................................................................................................................8ABSTRACT.....................................................................................................................................9CHAPTER1 INTRODUCTION..................................................................................................................11Decline in Intimate Partner Homicide.................................................................................... 11Importance of Examining Intimate Partner Homicide............................................................ 12Overview of Theoretical Perspectives.................................................................................... 14Importance of Structural Changes and Domestic Violence Awareness................................. 17Research Questions............................................................................................................. ....20Significance of Study..............................................................................................................20Summary.................................................................................................................................222 LITERATURE REVIEW.......................................................................................................28Theoretical Perspectives....................................................................................................... ..30Exposure Reduction Hypothesis...................................................................................... 30Factors that Reduce Exposure......................................................................................... 32Exposure Reduction Predictions...................................................................................... 37Backlash or Retaliation Hypothesis................................................................................. 38Backlash Predictions.......................................................................................................41Economic Deprivation and Marginalization................................................................... 41Economic Deprivation and Marginalization Predictions................................................ 45Summary..........................................................................................................................45Hypotheses..............................................................................................................................45Exposure Reduction......................................................................................................... 46Backlash..........................................................................................................................46Economic Deprivation and Marginalization................................................................... 47Change Models................................................................................................................ 473 DATA AND METHODOLOGY........................................................................................... 49Unit of Analysis............................................................................................................... .......49Sources of Data.......................................................................................................................50Homicide Data.................................................................................................................. ......50Domestic Violence Service Data............................................................................................ 51Measures.................................................................................................................................52

PAGE 5

5 Dependent Variable......................................................................................................... 52Independent Variables.....................................................................................................52Exposure reduction measures................................................................................... 53Backlash measures................................................................................................... 55Economic deprivation and marginalization measures.............................................. 56Controls.................................................................................................................... 58Missing Data...........................................................................................................................59Multicollinearity.....................................................................................................................61Methodology...........................................................................................................................63Descriptive Statistics....................................................................................................... 63Analytical Plan................................................................................................................ ........65Cross-Sectional Analyses................................................................................................65Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation........................ 664 RESULTS...............................................................................................................................74Cross-Sectional Analysis........................................................................................................74Female-Perpetrated Intimate Partner Homicide 1990..................................................... 74Male-Perpetrated Intimate Partner Homicide 1990......................................................... 76Female-Perpetrated Intimate Partner Homicide 2000..................................................... 77Male-Perpetrated Intimate Partner Homicide 2000......................................................... 79Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation........................ 815 DISCUSSION AND CONCLUSIONS.................................................................................. 90Discussion...............................................................................................................................90Exposure Reduction......................................................................................................... 91Backlash..........................................................................................................................93Economic Deprivation and Marginalization................................................................... 95Summary..........................................................................................................................95Changes Over Time.........................................................................................................97Limitations.................................................................................................................... ..........98Future Research......................................................................................................................99Conclusion............................................................................................................................100APPENDIX A Glossary................................................................................................................................103B Correlation Matrix for Explanat ory and Outcome Variable 1990........................................ 109C Correlation Matrix for Explanat ory and Outcome Variables 2000...................................... 111D Simplified Correlation Matrix for Explanatory and Outcome Variables 1990....................113E Simplified Correlation Matrix for Expl anatory and Outcome Variables 2000....................115F Variance Inflation Factors for Simplified Models................................................................ 117

PAGE 6

6 G Correlation Matrices for Domes tic Violence Service Variables. ......................................... 118REFERENCES............................................................................................................................119BIOGRAPHICAL SKETCH.......................................................................................................129

PAGE 7

7 LIST OF TABLES Table page 3-1Means, standard deviations (in parentheses) for all variables, 1990 and 2000 (N= 178). 693-2T-test scores for change from 1990 to 2000 (N=178)....................................................... 703-3Variance inflation factors for all variables included in the models (N=178).................... 713-4Principal components analysis af ter varimax rotation 1990 (N=178)............................... 724-1Zero-inflated negative binomial regression equations with coefficients (and ZScores) for gender-specific in timate partner homicide 1990............................................. 844-2Zero-inflated negative binomial regression equations with coefficients (and ZScores) for gender-specific in timate partner homicide 2000............................................. 854-3Summary of cross-sectiona l results, 1990 and 2000..........................................................864-4Fixed-effects regression coefficients fo r the relationship between changes in predictor variables and cha nges in logged female intimate partner homicide counts for 143a US cities, 1990-2000............................................................................................ 874-5Fixed-effects regression coefficients fo r the relationship between changes in predictor variables and cha nges in logged male intimate partner homicide counts for 174a US cities, 1990-2000.................................................................................................884-6Summary of change model results, 1990 to 2000.............................................................. 89

PAGE 8

8 LIST OF FIGURES Figure page 1-1Total homicide rate per 100,000 reside nts for large cities, 1989-2001............................. 241-2Gender-specific intimate partner homic ide rates for larg e cities, 1989-2001....................251-3 Proportion of all homicides invol ving intimate partners, 1989-2001............................... 261-4Summary of percent change from 1990 to 2000 in total homicide rate, femaleperpetrated intimate partner homicide rate and male-perpetrated intimate partner homicide rate.................................................................................................................. ....272-1Detailed theoretical and conceptual model on exposure reduction, backlash, economic deprivation and marg inalization and maleand fe male-perpetrated intimate partner homicide............................................................................................................... .48

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EXAMINING CHANGES IN MALE AND FE MALE INTIMATE PARTNER HOMICIDE OVER TIME, 1990-2000 By Amy Reckdenwald August 2008 Chair: Lonn Lanza-Kaduce Cochair: Karen F. Parker Major: Criminology, Law and Society Research on intimate partner homicide has increas ed in the recent years. This is partially due to the dramatic decline witn essed over the last couple of decad es in these types of homicides as well as the growth that has occurred in public awarene ss and policy responses toward domestic violence. Recent intimate partner hom icide research has predominately focused around two perspectives to explain th e relationship between intimate partner homicide and domestic violence resources the exposure reduction hyp othesis and the backlash or retaliation hypothesis, with results that support both (Dugan, Nagin, and Rosenfeld, 1999; 2003). The exposure reduction hypothesis proposes that do mestic violence resources that reduce the exposure or contact between intimate partners shou ld decrease the probability of intimate partner homicide, while the backlash hypothesis suggests the opposite. That is, domestic violence interventions may have unintended consequences and increase the risk of intimate partner homicide if they threaten male domin ance and control over their partners. The contradictory results from these studies have fr ustrated advocates and have made them question their efforts to make females safer. Societal remedies to lowe r intimate partner homicide seem to be addressing male-perpetrated intimate partner violence and homi cide insufficiently. It is important to gain a

PAGE 10

10 better understanding of what fact ors are truly influencing gende r-specific intimate partner homicide, during a time that is marked by signi ficant transformations in domestic violence legislation and response toward domestic violence with the enactment of the Violence Against Womens Act of 1994. This research helps to address the contradi ctory findings while controlling for a number of structur al factors that have shown to be important in the homicide literature. The current study examines both the ar guments of exposure reduction and backlash, in addition to economic deprivation an d marginalization to explain th e observed patterns in maleand female-perpetrated intimate partner homicide over time, something that has not been done in the literature to date. Measures of key concepts will be co llected for 2 decennial points (1990 and 2000). Supplemental homicide files for this time pe riod and census data will be utilized. Poisson regression models will be used to investigate wh ich theoretical perspective is associated with maleand female-perpetrated intimate part ner homicide in 1990 and 2000. Pooled crosssectional fixed effect time series regression will be used to determine whether changes from 1990 to 2000 in key structural indicators influe nce trends in maleand female-perpetrated intimate partner homicide during this same time period. Overall, resu lts suggest that key theoretical indicators do in fact influence the trends in female-perpetrated intimate partner homicide, but not male-perpetrated intimate partner homicide.

PAGE 11

11 CHAPTER 1 INTRODUCTION Only in the area of partner hom icides do wo mens perpetration rates approach that of men. (Browne, Williams, and Dutton 1999: 150) Decline in Intimate Partner Homicide Research on intim ate partner homicide1 has increased in recent years. Increased attention toward intimate partner homicide has resulted from the dramatic decline in perpetration that has been witnessed in these types of homicides. This decline coincided with the remarkable decline that was observed in total homicides rates that still puzzles researchers. Beginning in the early 1990s and extending into the 21st century, Amer icans witnessed an unprecedented drop in violent crime rates, specifically a drop in hom icide rates. This drop was witnessed across the U.S., but was seen in large cities2 in particular. From 1990 to 2000, ho micide rates in large cities dropped roughly 42%3. See Figure 1-1 for an illustration of the decline in total U.S. homicide rates in large cities4. Interestingly, the decline in adult homicide rates ha s been recorded for all victim-offender relationship categor ies, but the sharpest decline in homicides have been observed in the family category for both males and fe males (Rosenfeld, 2000), with most of those homicides generally occurring between spouses, ex-spouses, boyfriends, and girlfriends (Bureau of Justice Statistics, 200 6; Durose et al., 2005). Statistics reveal that U.S. intimate partner homicide c ounts have been declining for over two decades (Bureau of Justice Statistics, 2006 ; Greenfield et al., 1998). See Figure 1-2 for 1 Intimate partners include spouses, ex-spouses, or boyfriends and girlfriends. 2 Large cities have a population of 100,000 residents or more. 3 See Figure 1-4 for summary of percent change from 1990 to 2000. 4 Homicide rates per 100,000 population were calculated by dividing the homicide counts by the population and then multiplying that number by 100,000 to obtain a rate. Denominators were obtained from the 1990 and 2000 U.S. Census Bureau. Annual population estimates were calcula ted for between census years by subtracting the 1990 population from the 2000 population and then dividing that difference over time by 10 (i.e., number of years). The annual population estimate was then subtracted from the 1990 population to obtain a population estimate for 1989 and added to 1990 and each year after to obtain a population estimate for each of the following years.

PAGE 12

12 gender-specific total intimate partner ho micide rates for large cities, 1989-20015. Surprisingly, maleand female-perpetrated intimate partner homicide rates appear to follow a similar pattern, though rates of male-perpetrated intimate partner homicide are higher than female-perpetrated rates. That is, both male and female intimate part ner homicide rates fluctuat e but are fairly stable from 1989 until 1992. From 1992 to 1994 a dramatic spike in the rates of intimate partner homicide is apparent. After 1994, intimate partner homicide rates level off. Female-perpetrated intimate partner homicide rates remain fairly stable from 1995 to 2001, while male-perpetrated intimate partner homicide rates decline slightl y. Though, the rates of male and female intimate partner homicide show a similar pa ttern over time, a greater decline is observed in the rate of female-perpetrated intimate partner homicide. Speci fically, the rate of male-perpetrated intimate partner homicide decreased 33% from 1989 to 2001, whereas the rate of female-perpetrated intimate partner homicide decreased 45% during this same time. The reason for the more substantial decrease in the rate of female-perpe trated intimate partner homicide compared with male-perpetrated intimate partner homicide is unknown and remains an interest to researchers. Importance of Examining In timate Partner Homicide There are many reasons that make intimate partner homicide an important topic for research. First and foremost, as stated earlier, in timate partner homicide ha s decreased in recent years with no concrete explanation. Furthermor e, females are more likely to be killed by an intimate partner than any other offender. Among female homicide victims, roughly 1 in 3 is killed by an intimate partner, whereas among male homicide victims, only 5% are killed by an intimate partner. Moreover, although intimate partner homicide rates have been decreasing, the proportion of female homicide victims killed by an intimate partner has gradually increased from 1989 to 2001 (27.7% and 31.4%, respectively), whereas the proportion of male homicide victims 5 See footnote 4 for the rate calculation.

PAGE 13

13 killed by an intimate has decrea sed during this same time peri od (5.3% and 2.8%, respectively). This is evident in Figure 1-3. Specifically, fr om 1989 to 2001 there has been a 13% increase in the proportion of females killed by an intimate partner compared to a 47% decrease in the proportion of males killed by an intimate partner. In addition, the intimate partner is the only victim-offender relationship category where female-p erpetrated homicide ra tes approach that of men (Browne, Williams, and Dutton, 1999). This is apparent when examining the difference (i.e., the gap) between maleand female-perpe trated intimate partner homicide rates. Typically, males commit roughly 90% of all homicides occurring in the United States, making males 10 times more likely than females to commit murder (Bureau of Justice Statistics, 2006). Males are more likely than females to be perpetrators for every victim-offender relationship category (i.e., intimat e partners, family members, acquaintances, and strangers). However, when females kill they are more likely to kill someone that is close to them, such as an intimate partner or a family member, making th e intimate partner the only victim-offender relationship where female-perpetration approach es that of men (Browne, Williams, and Dutton, 1999). The gap in the difference between malea nd female-perpetrated intimate partner homicide rates for large cities has narrowed in recent years. This is apparent when examining Figure 1-2. The gap between the ra tes of maleand female-perpetrated intimate partner homicide was fairly stable from 1989 until 1992. In 1992, the rate of intimate partner homicide increased for both males and females, but far more drastically for males killing their female-partners. This difference in the rise in the rates of intimate partner perpet ration resulted in a widening of the gap between the rates of malea nd female-perpetrated intimate partner homicide. As the rates of maleand female-perpetrated intimate partner ho micide plummeted and then leveled off after

PAGE 14

14 1994, the gap in perpetration rates began to consistently narrow. The rate of male-perpetrated intimate partner homicide gradually decreased while the rate of female-perpetrated intimate partner homicide remained fairly stable, resu lting in a gradual narrowing of the gap in perpetration rates that continued through 2001. Figure 1-4 summarizes the decline in the rate of total U.S. homicides in large cities, as well as the decline in the rates of maleand female-perpetrated intimate partner homicide from 1990 to 2000. The total rate of homicide decreased 42% during this time period. Both maleand female-perpetrated intimate partner homicides decr eased as well. The rate of female-perpetrated intimate partner homicide had the largest de crease of 52% compared to 34% for maleperpetrated intimate partner homicide. No confir med explanation is available for the greater intimate partner homicide rates among males compared to females or the gender-specific differences that have been observed over time. Overview of Theoretical Perspectives Three theoretical perspectives are of interest in this study the exposure reduction hypothesis, the backlash hypothesis, and economic deprivation and m arginalization. Previous research has supported each of th ese perspectives; theref ore all three perspec tives are believed to influence intimate partner homicide uniquely. The exposure reduction hypothesis proposes that the reduction in the time spent (i.e., exposure) between couples in violent relationshi ps results in less intimate partner homicide, because couples have less contact with one another and therefore have less opportunities to kill their intimate partner. A number of factors have been consid ered to cause exposure reduction effects on intimate partner homicide, such as decl ining marriage rates, increasing divorce rates, greater economic and educational status of female s, and the availability of domestic violence resources and services (Dugan et al., 1999; 2003; Rosenfeld, 1997).

PAGE 15

15 The backlash hypothesis proposes that the incr eased status of females threatens males control and dominance over females and will resu lt in more intimate partner homicide, because males will use more violence to dominate and cont rol their female partners. Factors that have been thought to produce backlash ef fects overlap with some of the factors that are considered to produce exposure reduction effects. Some of the factors considered include: greater economic, employment, and educational status of females and the availability of domes tic violence services when they fail to adequately reduce contact betw een violent partners (D ugan et al., 1999; 2003). Research supports these two main theoretical perspectives to explain intimate partner homicide (Dugan et al., 1999; 2003). However, the contradictory results from these studies have frustrated advocates. For instance, Dugan et al. (1999) reported an exposur e reduction effect of domestic violence resources on male intimate part ner homicide victimization, but not on female intimate partner homicide victimization, suggesti ng that the movement to prevent domestic violence only really made men safer and not women. In addition, Dugan et al. (2003) reported that some domestic violence services actually crea te a backlash effect and seem to put some women in more danger of intimate partner homic ide victimization (i.e., prosecutors willingness to prosecute). Both of these perspectives need to be examined more closely to determine which factors are truly influencing gender-s pecific intimate partner homicide. Due to the contradictory findings from the above two perspectives it is important to also consider other factors. This study draws from the economic deprivati on and marginalization literature to gain a better unders tanding of intimate partner homicide. Economic deprivation and marginalization have been shown to explain hom icide in general and th erefore seems reasonable to include in the present rese arch. Ideas behind economic depriv ation and marginalization come from both the strain and feminist literatures. Stra in theorists relate the opportunity structure to

PAGE 16

16 violence. Merton, ([1949] 1968: 185248) argues that deviance is a result of an unequal distribution of opportunities among classes in society. According to Merton, the distribution of opportunities in the social structur e determines which classes are most likely to be involved in crime and deviance. Males or females that lack certain opportunities becaus e of deprivation (i.e., poverty and unemployment) may be frustrated or strained and this strain may influence intimate partner homicide perpetration. On the other hand, feminists scholars tend to focus on inequalities between men and women. Feminist literature ex amining inequalities tend to concentrate on the economic marginalization hypothesis, which fo cuses on the economic disadvantage of women relative to men (Heimer, 2000). The economic marginalization hypothesis argues that the economic disadvantage of women (i.e., female unemployment and poverty) is a significant predictor of female crime (Box and Hale 1983, 1984; Heimer, 2000; Heimer, Wittrock, and Unal, 2005; Hunnicutt and Broidy, 2004; Steffensmeier and Streifel 1992). Extant homicide research has incorporated the ideas behind ec onomic deprivation and ma rginalization (Bailey, 1984; Blau and Blau, 1982; Land, McCall, and Cohen, 1990; Parker, McCall, and Land, 1999; Steffensmeier and Haynie, 2000a; 2000b; Whaley and Messner, 2002); however limited research has included indicators of economic deprivation and marginalization to explain intimate partner homicide (for an exception see Reckdenwald and Parker, 2008). Economic deprivation and marginalization in terms of poverty or unempl oyment may influence male and female intimate partner homicide perpetration. All three theoretical perspectives have rece ived support in the liter ature; though at times some of the findings are mixed and even cont radictory. This study hopes to determine which theoretical factors significantl y contribute to maleand female-perpetrated intimate partner

PAGE 17

17 homicide and whether or not similar theoretica l factors influence male and female intimate partner homicide. Importance of Structural Changes and Domestic Violence Aw areness Declines in both maleand female-perpetrat ed intimate partner homicide rates were not the only marked changes witnessed during the 1990 and 2000 time period. Changes were also seen in domesticity, the status of women, leve ls of economic deprivation and marginalization, and the availability of domestic violence resource s. Any of these structural changes may have had an impact on the observed pattern s in intimate partner homicide. From 1990 to 2000, there were significant changes in marital and non-marital domesticity. Marriage rates plummeted and divorce rates increased, while the percent of unmarried partner households increased from 3.5% in 1990 to 5.2% in 2000 (US Census Bureau, 2003). Declining marriage rates a nd increasing divorce rates influe nce intimate partner homicide in terms of opportunity. With fewer people entering marriage and more getting divorced there is less opportunity for violent married intimate partners to kill each other. However, while rates of marriage and divorce are changing so are the rates of unmarried partners living together. With more and more individuals choosing to live t ogether and postpone marriage, the opportunity increases for intimate partner homicide between non-married intimate partners. Furthermore, the inherent dynamics of this type of relationship is different from a marital relationship and may have an effect on th e rate of intimate partner homicide. Moreover, the status of females increas ed from 1990 to 2000. Females employment status, income, and educational attainment all imp roved. For instance, the percentage of females 16 and over that were employed in 1989 was 54.3% compared with 57.4% in 1999 (Bureau of Labor Statistics, 2008a). Improvements in the st atus of females would afford females the opportunity to be able to leave an abusive relations hip before the violence resulted in death to

PAGE 18

18 themselves or their partners; thus, resulting in a lowering of intimate partner homicide. However, these improvements in females status may ha ve unintended consequences on intimate partner homicide. Some males may feel threatened by an in crease in the status of females and may react in a violent and possibly lethal ma nner towards their female partne rs to gain back the perceived loss of control. In addition to changing domesticity and impr oving status of females, the 1990s were a time of economic prosperity. The percent of indi viduals that reported liv ing in poverty decreased from 13.1% in 1989 to 12.4% in 1999 (U.S. Bureau of the Census, 2003). In addition to this, the unemployment rate decreased as well. In 1989, the percentage of the population that was unemployed was 5.3%, compared with 3.7% in 1999 (Bureau of Labor Statistics, 2008b). Economic deprivation may influence the dyna mics within the household. Less stress and frustration over lack of employment and living in poverty may result in a decrease in violence towards intimate partners. Domestic violence service and resource availabi lity increased in the 1970s, but more so after the enactment of the 1994 VAWA. During this time, domestic violence policies and programs that offered services for females and even males expanded considerably. The availability of domestic violence resources and services has been considered a key factor in the explanation of the decline in intimate partner homic ide (Dugan et al., 1999; 2003). Before the 1990s, many jurisdictions did not allow police to arrest on a misdemeanor charge unless they had witnessed the act. The battered Wo mens Movement pressed for redefining the act of domestic violence from a domestic disturbance (i.e., someth ing that is seen as a private family matter) to defining it as unacceptable and a criminal offense that will not be tolerated (Schechter, 1982). By the mid 1970s resources began to become availa ble to abused women. However, it could be

PAGE 19

19 argued that domestic violence against women was not seen as a national social problem until the enactment of the 1994 VAWA. The VAWA si gnificantly increased funding for domestic violence services, and supported domestic viol ence specialization in local police departments and prosecutor offices (Dugan et al., 2003: 171) Some examples of the VAWAs programs include policies to encourage the prosecution of abusers, victim services, and domestic violence prevention programs. Although this act specifi cally acknowledges violence against women, violence, both lethal and non-lethal, shoul d be acknowledged against both men and women. Domestic violence services should influen ce intimate partner homicide by limiting the opportunity for contact between violent intimate partners and thus limit the opportunities for partners to kill each other. Despite the reports that the increased availa bility of domestic violence services and resources is a significant factor in the explanation of the decline in intimate partner homicide over the past couple of decades (Browne and Williams, 1989; Dugan et al., 1999; 2003), resources have historically been directed at females and more so married females (Browne and Williams, 1993; Browne, Williams, and Dutton, 1999). This gives females options besides killing their partners and may in fact be a main reason that male intimate partner homicide victimization has decreased signi ficantly. However, this explan ation does not hold for female intimate partner homicide victimization. That is, female intimate partner homicide victimization has not seen as drastic of a reduction as ma le intimate partner homicide victimization. Taken together, there have been many change s witnessed over the la st couple of decades in intimate partner homicide as well as in domesticity, the status of women, economic deprivation, and in the availability of domestic violence resources. It is important to determine if these changes can account for the significant de cline in intimate partner homicide. Examination

PAGE 20

20 reveals that rates of female-perpetrated in timate partner homicide from 1990 to 2000 had a greater decline than male-perpetrated intimate pa rtner homicide. It is important to determine what factors account for this de cline as well as possible reasons why the decline was not as drastic for male-perpetrated intimate partner homicide. Research Questions One of the m ain purposes of this study is to examine gender-specific intimate partner homicide and determine which structural correlate s contribute to the observed patterns seen in 1990 and 2000. In addition, this study will examin e the changes between covariates in 1990 and 2000 to determine if any of the changes can ac count for the changes in maleand femaleperpetrated intimate partner homicide over time. The following research ques tions will be addressed: 1. What is the impact of structural measures on maleand female-perpetrated intimate partner homicide in 1990 and 2000? 2. Will criminological theories concerning domestic violence help explain maleand female-perpetrated intimate partne r homicide in 1990 and in 2000? 3. Does the availability of domestic violence services influence intimate partner homicide in 1990 and 2000? 4. What is the impact of change over time (1990-2000) in key stru ctural variables on maleand female-perpetrated intimate partner homicide trends? Significance of Study This study expands on the intimate partne r homicide literature in several ways. First, this study draws from three theoretical persp ectives in an effort to explain intimate partner homicide over time. The domestic violence lit erature has acknowledged both the importance of reducing the exposure between violent partners as well as backlash or retaliation by male

PAGE 21

21 partners as a response or cons equence to interventions designe d to reduce violence between couples. However, this research has found resu lts that support both pe rspectives. This study draws from two domestic violence perspectives while also controlling for a third theory of economic deprivation and marginal ization in attempt to more clearly understa nd trends in intimate partner homicide. This study examines both maleand femaleperpetrated intimate pa rtner homicide and compares their trends over time. This study not only documents change but also explains changes in intimate partner homicide trends over time and gives a better understanding of what is correlated with this observed change. Examining changes in key indicators on intimate partner homicide over time has not been done in the li terature. The dynamic nature of key indicators will be examined to determine whether or not th e changes from 1990 to 2000 in key indicators correlates with the change in maleand female -perpetrated intimate partner homicide over time. For example, it is of interest to determine if the change in the percent divorced in cities from 1990 to 2000 is related to the changes in intimate partner homicide during this same time period. Focusing on changes over time will give a better understanding of the differential trends in maleand female-perpetrated intimate partner homicide. The time period being examined in this study depicts a time where there has been significant changes in homicide itself as well as domesticity (i.e., living arrangements), the status of females, and economic deprivation and margin alization. Moreover, during this time, domestic violence was being seen as a national social problem and funding for domestic violence services were increasing significantly with the enactme nt of the 1994 Violence Against Womens Act (VAWA). This study controls for the influence of various domestic violence services while

PAGE 22

22 examining other structural factors in an effort to better understand maleand female-perpetrated intimate partner homicide over time. Summary This is a m acro-level study that proposes to build on existing literature by further considering the effects of key indicators that represent the exposure reduction and backlash perspectives, while also includi ng a third theoretical perspectiv e of economic deprivation and marginalization. Data will be collect ed to capture these theoretical perspectives in an effort to further the research on intimate partner homicide. Guided by these three perspectives there are two main research questions that this study hopes to address. Firs t, this research will determine what structural factors correlate with malea nd female-perpetrated intimate partner homicide cross-sectionally (1990 and 2000) before examini ng changes overtime. Next this research will document changes between key structural factors to determine if these changes correlate with the changes in maleand female-perpetrated intimate partner homicide over time. Intimate partner homicide has gone through some remarkable changes over the past couple of decades. Both maleand femaleperpetra ted intimate partner homicide have witnessed drastic declines. However, gender-specific trends in intimate partner hom icide rates show that female-perpetrated intimate partner homicide has declined more drastically than maleperpetrated intimate partner homicide, thus lessening the gap between male and female intimate partner homicide perpetration rate s. Recent research has predomin ately used two main ideas to explain the over two decade dec line in intimate partner homicide the exposure reduction hypothesis and the backlash hypothesis. Research re sults favor both perspec tives. Research also points to the importance of the emergence of domestic violence services and programs with the Battered Womens Movement in the 1970s and the VAWA of 1994. This study is designed to examine these two theoretical perspectives as well as a third theoretical perspective of economic

PAGE 23

23 deprivation and marginalization, which has shown to be very important in the homicide literature (Bailey, 1984; Blau and Bla u, 1982; Land, McCall, and Cohen, 1990; Parker, McCall, and Land, 1999; Steffensmeier and Haynie, 2000a; 2000b; Whal ey and Messner, 2002), in an attempt to gain a better understanding of th e differential gender-specific trends in intimate partner homicide. Structural factors and th e availability of domestic violen ce services are of particular importance. This dissertation begins with Chapter 2 detailing the three theoretical perspectives that are of interest in this stu dy. In Chapter 3 the data and methodology are discussed. Chapter 4 provides the results of the multivariate analysis. Chapter 5 contains conclusions and directions for future research.

PAGE 24

24 0 5 10 15 20 25 1989199019911992199319941995199619971998199920002001YearRate per 100,000 residents Total Homicide Rate Figure 1-1. Total homicide rate per 100,000 residents for large cities, 1989-2001.

PAGE 25

25 0 1 2 3 4 5 6 7 8 9 10 1989199019911992199319941995199619971998199920002001YearRate per 100,000 persons 15 years and older Male-Perpetrated Intimate Partner Homicide Female-Perpetrated Intimate Partner Homicide Figure 1-2. Gender-specific intimate partne r homicide rates for large cities, 1989-2001.

PAGE 26

26 0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 30.00% 35.00% 1989199019911992199319941995199619971998199920002001Year% of Victims Killed by an Intimate Male Female Figure 1-3. Proportion of all homicides involving intimate partners, 1989-2001.

PAGE 27

27 1990 2000 % Change Total Homicide Rate 20.9827 12.1792 41.9 Female-Perpetrated IPH Rate 1.520151 0.730003 51.9 Male-Perpetrated IPH Rate 5.066648 3.325714 34.36 Figure 1-4. Summary of percent change from 1990 to 2000 in total homicide rate, femaleperpetrated intimate partner homicide rate and male-perpetrated intimate partner homicide rate.

PAGE 28

28 CHAPTER 2 LITERATURE REVIEW While research on intim ate partner homicide is growing, the majority of research on intimate partner homicide is directed at describing victimization and/or offending patterns (Block and Christakos, 1995; Browne and Williams, 1993; Gallup-Blac k, 2005; Puzone et al., 2000; Reidel and Best, 1998; Shackelfo rd and Mouzos, 2005) rather th an analyzing these patterns by controlling for the impact of key factors that ma y have an influence on intimate partner homicide over time (Block and Christakos, 1995; Browne and Williams, 1993; Puzone et al., 2000; Reidel and Best, 1998). This research ha s found that the trends in intima te partner homicide during this time period vary not only by the gender of the victim and offender, but also by the type of relationship between the victim and offender. For instance, Puzone et al. (2000) examined U.S. homicide vic timization from 1976-1995 and found that the decline in homicide victimization rates were more pronounced for husbands (75%) than wives (41%), boyfriends (68%) more than girlfriends (21%)6, and ex-husbands (88%) more than ex-wives (41%). Specifically, Puzone et al. (2000) found that the risk of being killed by a spouse in 1995 was 2.6 times greater fo r a wife than a husband, 2 times greater for a girlfriend than a boyfriend, and for the combined period of 1992-1995 was 3.8 times greater for an ex-wife than an ex-husband. Earlier work ex amining maleand female-perpetrated intimate partner homicide supports thes e findings (Browne and Williams, 1993). That is, Browne and Williams (1993) examined Supplemental Homicide Report data from 1976-1987 for all states and found that both married male-perpetrated (i.e., males killing their female spouses) and female-perpetrated (i.e., females killing their male spouses) intimate partner homicide declined, however the decline was more pronounced for the rate of wives killing husba nds than the rate of 6 However, this decline was not statistically significant.

PAGE 29

29 husbands killing wives. In a ddition, Browne and Williams (1993) showed that while marital intimate partner homicide has been decreasing ov er the years, male-perpetrated non-marital intimate partner homicide has actually been increasing. Browne and Williams (1993) concluded that the rate of unmarried females being killed by their intimate partner increased significantly during this period, while the rate that unmarried males were k illed by their intimate partner show no clear trend during the 1976-1987 period (87). Similarly, in Rosenfelds (1997) examination of intimate partner homicide vict imization in California between 1987 and 1996, rates of marital intimate partne r homicide decreased while rates of non-marital intimate partner homicide increased. Rosenfeld ( 1997) attributes this marked ch ange to an increase in the proportion of younger people who are not married. Along these lines, Riedel and Best (1998) also examined intimate partner homicide in Ca lifornia for the years 1987 to 1996 and found that intimate partner homicide is more common in co mmon-law relationships co mpared with spousal and boyfriend/girlfriend relationships They suggest that a reason for this may be because of the meaning that is attributed to the relationship (Makepeace, 1997) and the lack of stability (Makepeace, 1989). In summary, research descri bing recent trends in intimate partner homicide has found some interesting results. It a ppears that while total intimate partner homicide is decreasing, once one disaggregates by gender and relationship type different trends emerge. For instance, overall patterns show that males have experienced a greater decline in intimate partner homicide victimization than females. In addition, married intimate partne r homicide has decreased, while non-married intimate partner homicide has increased and this increase is especially pronounced for non-married female intimate partner victimization.

PAGE 30

30 Though research has not come to a definitive e xplanation for these trends, research that has controlled for the impact of key factors ha s tended to attribute the recent patterns to the social responses toward domestic violence; specificall y, the availability of domestic violence resources (Browne and Williams, 1989; Dugan et al., 1999; Dugan et al., 2003) and have found that the availability of legal (i.e., presence of statutes pertaining to domestic violence) and extralegal services (i.e., number of shelters and other pr ograms) to be related to the decline in the rates of female-perpetrated intimate partner homicide, but not associated with the rates of maleperpetrated intimate partner homicide (Browne a nd Williams, 1989). However, research has also shown that some domestic violence resources (i.e ., prosecutors wiliness to prosecute) that were intended to help women have the unintended co nsequence of making them more at risk for intimate partner homicide victimization (Dugan et al., 2003). Increasing di vorce rates, declining marriage rates, and an improved economic status of women have also shown to be important in explaining intimate partner homicide (Dugan et al., 1999; 2003; Rosenfeld, 1997). These factors will be examined more closely as I describe empi rical findings for the theoretical perspectives of interest in this study. Theoretical Perspectives Three theoretical perspectives are of in terest to this study. The following pages will explain each perspective the exposure reduc tion hypothesis, the backlash or retaliation hypothesis, and economic deprivation and margina lization. The section w ill end with detailed hypotheses for the current study. Exposure Reduction Hypothesis According to the exposu re reduction hypothesi s, limiting the exposure or contact of intimate partners to one another should decrease the probability of intimate partner homicide, because there is less exposure to a violent partne r and more opportunities to exit the relationship.

PAGE 31

31 Ideas behind the exposure reducti on perspective stem from three main findings from previous research. Domestic violence research s hows that intimate partner homicide is commonly the result of prolonged violence in the relationship (Browne, 1987; Campbell, 1992; Goetting, 1995; Smith, Moracco, and Butts, 1998; Totman, 1978) or a prior history of violence (Browne, 1987; Chimbos, 1978; Daniel and Harris, 1982; To tman, 1978). For instance, Campbell (1992) estimated that in Dayton, Ohio 64% of female intimate partner homicide victims were abused physically by their offender before they were ki lled. Also, they estimated that 79% of male intimate partner homicide victims had abused th eir offender previously during the relationship. In addition, self-report studies ha ve shown that abused women make many attempts to try to get outside help before actually committing the intimate partner homicide (Sherman and Berk, 1984). For instance, in reviewing police records in Detr oit and Kansas City, Sherman and Berk (1984) discovered that in 90% of the intimate partner homicide cases police had been called at least 1 time during the 2 years prio r to the homicide. Furthermore, they found that in 54% of the intimate partner homicide cases, police had been called 5 or more times. Other research has shown similar results (Goetting, 1995). Moreover, research has shown that in violent relationships women are more likely to be seriously injured (Berk et al., 1983; Browne, 1993; Brush, 1990; Crowell and Burgess, 1996; Fagan and Browne, 1994; Langan and Innes, 1986; Schwartz, 1987; Stark, Flitcraft, and Frazier, 1979; Stets and Straus, 1990; Stout and Brown, 1995; Straus 1993). In addition, victimprecipitation is more likely in female-perpetrated intimate partner homicide than maleperpetrated intimate partner homicide (Goe tting, 1995; Rosenfeld, 1997; Silverman and Mukherjee, 1987; Wolfgang, 1958). Victim-precipita tion means the victim initiated the incident

PAGE 32

32 that later resulted in their own death by their intimate partner. Rosenfelds (1997) results support this idea as do a number of other research studies (Campbell, 1992; Cazenave and Zahn, 1992; Goetting, 1995; Jurik and Winn, 1990; Silver man and Mukherjee, 1987; Wilson and Daly, 1992a). To summarize, intimate partner homicide is more likely to occur in a relationship where there is a history of violence. In these viol ent relationships, women are more likely to be seriously injured than men. Furthermore, vict im-precipitation is more likely in femaleperpetrated intimate partner homicide. These find ings suggest that wome n have very different motives for killing their intimate partners and wo men are more likely to kill intimate partners out of protection or self-defense (Bernard et al ., 1982; Block and Christ akos, 1995; Browne, 1985; 1987; Campbell, 1992; Daly and Wilson, 1988; Dobash et al., 1992; Dugan et al., 1999; Goetting, 1988; Peterson, 1999; Stout, 1991; Stout and Brown, 1995; see Archer, 2000 and Saunders and Browne, 2000 for a review). That is, wh en women kill intimate partners it is likely a result of attempting to preven t their partners from hurting their children or themselves, during a violent attack by their partner, or in an attempt to prevent a future attack (Browne, 1986; 1987; Dugan et al., 1999). According to Wilson and Daly (1992b) Unlike men, women kill male partners after years of suffering physical violence, after they have exhausted all available sources of assistance, when they feel trapped, and because they fear for their own lives (206). Factors that Reduce Exposure Availability of domest ic violence resources. A main factor that has been shown to influence female-intimate partner homicide by reducing the exposure of violent pa rtners is the availability of domestic violence resources. Browne and Williams (1989) was the first study to examine the influence of the availability of domestic violence resources (i.e., number of shelters; wife abus e programs other than shelters)

PAGE 33

33 and domestic violence legislati on (i.e., statutes providing civil in junction relief for victims of abuse; statutes providing temporary injunction relief during di vorce, separation, or custody proceeding; statutes defining the physical abuse of a family or household member as a criminal offense; statutes permitting warrantless arrest based on probable cause in domestic violence cases; statutes requiring data collection and the reporting of family violence by agencies that serve these families; and statutes providing f unds for family violence shelters) on femaleperpetrated intimate partner homicide. Following prior work on homicide and intimate partner homicide, specifically, (Browne and Flewelling, 1986) Browne and Williams (1989) examined the rate of intimate partner homicide in U.S. states for the year s 1976-1979 and 1980-1984 and found that indeed, domestic violence resources have the ability to provide abused females with alternatives besides killing their male partners Specifically, they found a negative relationship between both domestic violence resources and dome stic violence legislation and intimate partner homicide, however once controls were included in the analysis only the domestic violence resources remained significant. Other research has also supported this hypothesis and shown that social response to domestic violence may be important in explaining intimate partner homicide (Browne and Williams, 1989; Dugan et al., 1999; Dugan et al., 2003). According to Browne and Williams (1989) domestic violence resources provide the means for women (1) to seek protection for them selves and their children, such as emergency restraining orders against abuse and harassment or police removal of the abusive mate; (2) to employ more direct avenues of escape, such as sh elters where they can hide from the abuser for short periods; (3) to ut ilize third-party interventions, such as support groups, crisis counseling, and legal aid, which give advise and encouragem ent in identifying eff ective nonviolent means of responding to threat or danger; and, in some areas, (4) to recei ve the benefits from court-

PAGE 34

34 mandated treatment programs that work directly with the abuser on his problem with violence (76). Dugan et al. (1999) examined the ideas be hind exposure reduction as well, looking specifically at the availa bility of domestic violence resources, as well as declining domesticity (i.e., decrease in the marriage rate), divorce (i.e., increase in the divorce rate) and the improved economic status of women (i.e., fe males educational attainment re lative to males and females income relative to males). They hypothesized that these three factors would reduce the exposure between violent partners. Results are supportive of the exposure re duction perspec tive; however these factors contributed more to the decline in male intimate partner homicide victimization rather than female intimate partner homicide vi ctimization. That is, they found that domestic violence resources (i.e., legal a dvocacy) reduced homicides of ma rried males; however they did not find that domestic violence resources reduced homicides of married females. Dugan et al.s (2003) work was an extension of their previous (1999) work. They again examined the relationship between intimate part ner homicide and domestic violence resources, but they extended their research by including more places (i.e., 48 large cities with a population over 250,000), a longer period of ti me (i.e., 1976-1996), and utilize a richer source of domestic violence resource measures. As with their pr evious research they examined the exposure reduction effects7. They found that aggressive arrest polic ies have an exposure reduction effect such that more aggressive arrest policies in cities is associated with less unmarried intimate partner homicide. In addition, they also showed an exposure reducti on effect such that a decrease in Aid to Families with Dependent Children benef its is related to an increase in male victims killed by their girlfriends, sp ecifically African-American male s. They suggest that public assistance provided to poor women with children may cushion the financial impact of leaving a 7 Dugan et al. (2003) also examined backlash effects of domestic violence resources.

PAGE 35

35 violent partner (Allard et al., 1997), but reductio ns in AFDC benefits would limit opportunities for females to leave their abusive partners a nd live on their own; thus resulting in more opportunities for females to kill their abusive pa rtners. Other variables that were shown to produce an exposure reduction effect include: ma rriage rates, legal advocacy, warrentless arrest laws, and mandatory arrest laws. Declining domesticity. Other common factors that have been cons idered to produce exposure reduction effects are increasing divorce rates, and decreasing marriag e rates. Rising divorce rates would result in fewer married couples living together and woul d therefore reduce the exposure between violent couples. The same idea is behind falling marria ge rates, which would reduce the exposure of violent couples because fewer individuals would be getting married and living together. Rosenfeld (1997) examined intimate partner homici de trends in St. Louis and found that 30% of the decline in African American spousal homicides was attributable to falling marriage rates and rising divorce rates. Dugan et al. (1999) found that domesticity doe s in fact influence intimate partner homicide. They found that the decl ine in marriage rates is related to the decline in married male and female intimate partner victimization. That is they found that the higher the marriage rate in cities, the higher the rates that wives kill th eir husbands and husbands kill their wives. Conversely, a high divorce rate corresponde d to fewer wives killing their husbands8. Specifically, they found that for every reduction of 10,000 women entering marriage, 2.3 married womens lives are saved. The same decrease in marriage among men saves the lives of 3.8 married men andas 10,000 additional men are granted divorce, 3.8 married male homicides are avoided (203). Dugan et al. (2003) found similar results. Specifi cally, they found 8 However, declining marriage rates was related to an in crease in unmarried males killing their female partners.

PAGE 36

36 that lower marriage rates are related to fewer husbands being killed by their wives. Rosenfeld (1997) suggests that much of the decline in intimate partner homicide is a function of change in the rate of marriage with the ag e groups at highest risk of hom icide victimization and offending (73). However, decreasing marriage rates may mean that more individuals are cohabitating together without getting married. Cohabitation has been shown to be an important risk factor in intimate partner homicide (Shackelford and Mouz os, 2005). Research suggests that males and females in cohabitating relations hips are at a higher risk of intimate partner homicide victimization compared to males and females in married relationships (Daly and Wilson, 1988; Shackelford, 2001; Wilson, Daly, and Wright, 1993; Wilson, Johnson, and Daly, 1995). For instance, Wilson, Daly and Wright (1993) and Wilson, Johnson and Daly (1995) found that females that cohabitate with their partner are 9 times more likely to be killed by their intimate partner than are married females. In additi on, Shackelford (2001) found that men that are cohabitating with their female partners are 10 mo re likely to be victims of intimate partner homicide compared to men in married relati onships. Rodriquez and Henderson (1995) suggest that individuals in cohabitating relationships, having invested heav ily in such relationships, may be more likely to resort to violent retaliation (i.e., homicide) than those in dating or noncohabitating relationships because they do not per ceive themselves as having legal protection associated with marriage (48). Improved economic status of women. Improved economic status of women has al so been considered to produce exposure reduction effects. Improved economic status of women in terms of educational attainment, income, and employment increases the access of women to opportunities that may give them

PAGE 37

37 other options besides resorting to killing their male partners. Al so, increased status would reduce females dependence on males. D ugan et al. (1999) did indeed fi nd that females improved status was associated with intimate pa rtner homicide victimi zation, particularly male intimate partner homicide victimization. That is, the increase in females relative income is associated with a decline in married female-perpetrated homicide. Furthermore, an increase in females relative educational attainment is associated with a de cline in non-married ma le victimization. They suggest that more educated women are better able, and perhaps more willing, to exit violent relationships and thus avoid kil ling their partner (204-205). Exposure Reduction Predictions In summ ary, it appears that women in viol ent relationships may be looking for other options, but without other options than remaining in the relationshi p available may turn to lethal methods. Ideas behind the exposure reduction pers pective suggest that limiting the exposure or contact of intimate partners to one another shou ld decrease the probabili ty of intimate partner homicide, because there is less exposure to a violent partner and more opportunities to exit the relationship. Empirically, if exposure reduction efforts were ava ilable to intimate partners before either partner took the violence to a lethal level intimate partner homicide should decrease. Based on these ideas it seems likely that factors that reduce contact or exposure between intimate partners in violent relationships should have a significant impact on female-perpetrated intimate partner homicide, because certain factors may give women other options than having to resort to violence, lethal or not, to protect themselves. The availability of domestic violence resources and the possible effect on intimate partner homicide has gained increase a ttention. Results have been su pportive of the exposure reduction perspective; however this is only true of female-perpetrated intimate partner homicide. It appears that domestic violence services that are designe d to make females safer are actually making

PAGE 38

38 males safer instead. Divorce rates have also proved to be a significant factor influencing intimate partner homicide; as well as cohabitation of intimate partners, and improved status of females. It is believed that measures of exposure reduc tion will have a significant effect on femaleperpetrated intimate partner homicide by creating other options for female s besides killing their male partner. Backlash or Retaliation Hypothesis Although research has shown the importance of reducing the exposure between intim ate partners in violent relationships, it is well known that the highest risk for homicide is when the victim leaves the relationship and this is especially true for females being killed by their male partners (Block and Christakos, 1995; Block, 20 00). Research has suggested the possibility of retaliation by the abusive part ner from domestic violence interventions (Campbell, 1992; Goetting, 1995). Dugan et al. (2003) s findings are supportive of th is statement. Dugan et al. (2003) found a retaliation effect where domestic violence resources actually increased homicide between intimate partners because they failed to effectively reduce e xposure between intimate partners. In fact, the prosecuto rs willingness to prosecute viol ators of protection orders, though intended to reduce exposure between violent intim ate partners, actually caused a retaliation effect where homicide increased for married and unmarried white females and African-American unmarried males. They conclude d that being willing to prosecu te without providing adequate protection may be harmful (192). Contrary to exposure reducti on predictions, research has hypothesized that increased economic status of females would make it easier for females to exit an abusive relationship and may possibly threaten males control over their pa rtners. Dugan et al. (1999) found that net of other changes, more educated women are better able, and perhaps more willing, to exit violent relationships and thus avoid killing their partner (205), howev er improved status appears to

PAGE 39

39 increase the female intimate partner homicide victimization rate. Duga n et al. (1999) suggests that some men are threatened by their loss of power over their female partners as their status increases. The backlash perspective theorizes that incr eased male violence towards their intimate partners is due to a perceived loss of power or control because of womens improving economic conditions (Browne, 1987; Vierai tis and Williams, 2002). This hypothesis was first developed by Williams and Holmes (1981) and Russell (1975) as a warning of the potential consequences of womens increased economic conditions or gender equality during the womens movement. That is, womens improving economic conditions (i.e., increased income, employment, and educational attainment) would make women less dependent on males, which would threaten men and actually cause increased viol ence towards females by males. For instance, Avakame (1999) states initial reductions in gender inequality might cause increases in violence because it would frustrate males into intensifying their use of vi olence to reassert their diminishing patriarchal power and authority (927). Though measures of gender inequality ha ve been identified by feminist scholars as a cause of violence agai nst women, research has shown conflicting findings about the direction of the relationship. Some rese arch has been supportive of this view (Yang and Lester, 1988; Whaley and Messner, 2002), while ot hers have not been (Gartner et al., 1990). Whaley and Messner (2002) hypothesized and found that greater gender equality increased male killings of females, results directly consistent w ith the backlash perspectiv e. They concluded that greater gender equality is threatening to ma le dominance, and as such, it increases male violence against women (199). In addition, other research has shown that when male unemployment is high relative to females, wives killed by their husbands increases (Bailey and Peterson, 1995). However, other research shows contradictory results. Gartner et al. (1990)

PAGE 40

40 examined females and males risk of homici de victimization for the period 1950 to 1980 for 18 developed democracies (i.e., 13 western European nations, Canada, the U.S., New Zealand, Australia, and Japan) and found a relationshi p between higher female education and lower female homicide. They conclude that the fact that the relationship between higher female education and lower female homicide increases over time suggests that the status gained from education acts through incremental changes in po licies and norms proscrib ing violence against women (608). Attempts have been made to test the backlash perspectiv e on other forms of violence, such as rape (Avakame, 1999; Baron and Straus 1987; Ellis and Beatie, 1983), however mixed findings have resulted. For instance, Avakame (1999) examined rape data from the National Crime Victimization Survey (1992-1994) to determin e if females labor force participation had an increase on females rape victimization due to backlash, but did not find support. Instead, Avakame (1999) found that unemployed women we re at a greater likelihood of being raped. However, Ellis and Beattie (1983) showed that gender equality (i.e., in terms of male/female differences in earnings, educational attainment, and employment) is related to an increased likelihood of rape. Surprisingly, Baron and St raus (1987) found the opposite results more gender equality, lower likelihood of rape. Individual-level research has looked at status incompatib ilities between husbands and wives and the influence on physical and emotional abuse. Kaukinen (2004) examined the effect that men and womens relative economic contri butions to the family have on husband-wife physical and emotional abuse. Status compatib ility was broken down into three statuses

PAGE 41

41 traditional status, status parity, and status reversal.9 Kaukinen (2004) found that there was no effect of status compatibility on physical violence, but there was an effect for emotional violence. Specifically, women in status reversal relationships are at risk for emotional abuse by their partner. It seems that men in relationships where their female partners had a higher status exerted their control by emotiona lly abusing their partners. Backlash Predictions Despite the mixed results for the backlash perspective, research has suggested that power differences between partners increase the ri sk for abuse among women (Anderson, 1997). In addition, men who kill their intimate partner do so to dominate and control their female partner (Wilson, 1989; Wilson and Daly, 1992a, 1992b). It is th erefore important to include measures of backlash in the study of maleand female-perpetr ated intimate partner homicide to determine if these factors influence intimate partner homic ide in 1990 and 2000 and whether the changes in these factors contribute to the change s in intimate partner homicide over time. It is expected that measures of backlash will have a significan t impact on male-perpetrated intimate partner homicide, because males will use violence to exert control over their female partners in conditions where female status is high or in co nditions where they feel a perceived lack of control over their female partners. Economic Deprivation and Marginalization The im portance of economic deprivation as a factor influencing crime has been documented in both the strain and feminist literatu res and it is consistent with the other two perspectives of exposure reduction and backla sh. The strain literature posits opportunity structures to be related to crime rate s (Merton [1949], 1968). Merton ([1949], 1968: 185-248) 9 In traditional status the economic status between the woman and the partner favored the partner. In status parity the economic status between the woman and the partner was equa l. In status reversal the economic status between the woman and the partner favored the woman.

PAGE 42

42 argues that deviance is a result of an unequa l distribution of opport unities among classes in society. According to Merton, the distribution of oppor tunities in the social structure determines which classes are most likely to be involved in crime and deviance. Blocked or unavailable opportunities have been suggested to increa se offending (Cohen, 1955; Cloward and Ohlin, 1960; Merton, 1968). Caused from differentials in access to opportunities, economic inequalities in particular have been suggested to increase violence levels (Braithwait e, 1979; Blau and Blau, 1982). Specifically, Blau and Blau (1982) found that relative racial socioeconomic inequalities have a significant influence on criminal violence. That is, aggressive acts of violence seem to result not so much from lack of advantages as from being taken advantage of, not from absolute deprivation but from relative deprivation (126). Other research has examined the influence of economic inequalities on homicide (for review s see Messner and Rosenfeld, 1999). Research has shown that, when economic deprivation is measur ed as an index including many measures of deprivation it is an important predictor of homicide rates (Land, McCall, and Cohen, 1990; Parker, McCall, and Land, 1999). Feminist scholars are interested in a particular form of inequality, particularly genderedinequality and its influence on gendered-violence. Feminist literature signifies the patriarchal system of male domination and control over wome n as a cause of gender inequalities in society and ultimately the oppression of women in society (Walby, 1986). Feminist scholars argue that domestic violence is rooted in gender and pow er and represents mens active attempts to maintain dominance and control over women (Anderson, 1997: 655). When examining gender inequalities a majority of research tests the economic marginaliza tion hypothesis, which attributes female crime to the economic disadva ntage of women relative to men (Heimer, 2000).

PAGE 43

43 Specifically, the economic marginalization hypothesis suggests that as the economic disadvantage of women increases, so will female crime. Research has found support for this hypothesis in regards to female perpetrated o ffenses including intimate partner homicide (Box and Hale, 1983; 1984; Steffensmeier and Streif el, 1992; Dugan et al., 2003; Reckdenwald and Parker, 2008). To test the economic marginaliz ation hypothesis, research has measured the economic disadvantage of females in terms of fe male poverty and joblessness. For example, Box and Hale (1983; 1984) found that female unemployment was positively associated with female violent and property offenses. Impor tantly, research has shown that there is an overlap between factors that effect male and female offe nding (Boritch and Hagan, 1990; Steffensmeier and Haynie, 2000a; Steffensmeier and Haynie, 2000b). For instance, Steffensmeier and Haynie (2000a) examined the effects of structural disa dvantage (i.e., poverty, unemployment, income inequality, female-headed households, and percen t black) on male and female offending rates for multiple offenses (i.e., homicide, robbery, aggravated assault, burglary, larceny-theft). Steffensmeier and Haynie (2000) concluded that structural disa dvantage variables influenced female and male offending rates in similar ways. Surprisingly, the effects of the structural variables were stronger for male offending. The majority of studies have examined gender inequalities in relation to the killings of females (Bailey and Peterson, 1995; Gauthi er and Bankston, 1997; Avakeme, 1999), but examining the impact of gender inequality on male homicides is important as well (Whaley and Messner, 2002). For instance, Whaley and Messner (2002) explored the effects of both gender inequality and economic deprivation on gendere d-homicides (i.e., male offender-male victim, male offender-female victim, female offender-female victim, and female offender-male victim) in large cities for 1990 to 1994. Whaley and Messne r measured gender inequa lity as the ratio of

PAGE 44

44 male to female median income, the percent of males aged 16 and over who were employed relative to the percent of females aged 16 a nd over who were employed, the percent of males employed in the labor force, the percent of managers, executives and administrators who were male, and the ratio of males aged 25 or more w ith 4 or more years of college education to females aged 25 or more with 4 or more year s of college education. They measured economic deprivation as the percentage of black males, pe rcentage of black females, percentage of poor males, percentage of poor females, percentage of unemployed males, percentage of unemployed females, and the gini index of income inequality10. Results suggested that gender equality is positively related to rates of males killing othe r males and males killing females in southern cities. In addition, they found that economic de privation had a significant positive relationship with females killing males, females killing ot her females, males killing females, and males killing other males. Consistent with this research, other re search has found that poverty and economic inequality influence male and female homicide s (Gartner, Baker, and Pampel, 1990; Smith and Brewer, 1992). Also, research has shown that marri ed females are at a higher risk of intimate partner homicide victimization in cities where there are high levels of educational attainment and employment inequalities between men and wo men (Bailey and Peterson, 1995). Furthermore, research has found that womens ec onomic power is significantly re lated to their involvement in intimate partner homicide. Specifically, it has been shown that as womens economic resources increase their involvement in intimate partne r homicide decrease (Gau thier and Bankston, 1997). 10 The gini index represents an index of income inequa lity. The index can range from zero, representing perfect equality, to 1, representing perfect inequality.

PAGE 45

45 Economic Deprivation and Marginalization Predictions Although econom ic deprivation and marginalizat ion is talked about differently in the strain and feminist literatures, it is still appli cable to the current study and should have an impact on gender-specific intimate partner homicide. Economic deprivation is expected to increase both maleand female-perpetrated intimate partner homicide, either in terms of differential opportunity structures or marginalization. Summary In summ ary, three theoretical perspectives are examined to determine their ability in explaining maleand female-intimate partner homici de. Each theoretical perspective is believed to have a unique influence on maleand female-perpetrated intimate partner homicide, based on findings from prior literature. The exposure redu ction perspective and the backlash perspective have shown to be important in understanding in timate partner homicide. Economic deprivation and marginalization has shown to be an importa nt explanation of homicide offending, and is assumed to have an influence on intimate partner homicide as well. Many of the factors believed to represent the theoretical perspectives overlap in many ways. For instan ce, in prior research inequality has been used to represent economic deprivation and/or marg inalization (Whaley and Messner, 2002; Reckdenwald and Parker, 2008), as well as exposure redu ction (Dugan et al., 1999); however in this study inequalit y represents a measure of back lash. I have done my best to break up the variables based on ideas from other re search as well as my research hypotheses for this study. Hypotheses Based on the previous literatu re on intimate partner homicide and homicide in general, it is expected that indicators of exposure re duction, backlash, and ec onomic deprivation and marginalization will significantly impact intimate partner homicide for both males and females

PAGE 46

46 for the years 1990 and 2000. Figure 2-1 contains a de tailed theoretical and conceptual model of the theoretical perspectives on maleand female-perpetrated intimate partner homicide. As illustrated in Figure 2-1, it is believed that m easures of exposure redu ction will significantly decrease female-perpetrated intimate partner homicide, by reducing the exposure between violent partners and reducing the chance that females will have to kill their partners in selfdefense. In addition, it is hypothesized that measures of backla sh will significantly increase male-perpetrated intimate partner homicide, due to males perceived loss of control and dominance over their female partners. Furthermor e, it is believed that measures of economic deprivation and marginalization will increase bot h maleand female-perpetrated intimate partner homicide, whether it is from the l ack of opportunities or marginali zation. Also, it is expected that changes in key indicators from 1990-2000 will imp act changes in maleand female-perpetrated intimate homicide from 1990-2000. Examining these changes will give a cl earer picture of the differential declines witnessed for maleand fema le-perpetrated intimate partner homicide. As depicted in Figure 2-1, I pr opose the following hypotheses: Exposure ReductionH1: Measures of exposure redu ction will ha ve a significant negative impact on femaleperpetrated intimate partner homicide for bot h 1990 and 2000, based on research that suggests that some females are more likely to kill their intimate partner in self-defense and that exposure reduction efforts are mainly geared towards females. BacklashH2: Measur es of backlash will have a significa nt positive impact on male-perpetrated intimate partner homicide for both 1990 and 2000, based on the idea that an increase in the status of females causes some males to kill their female partners in an effort to exert control and dominance over them.

PAGE 47

47 Economic Deprivation and MarginalizationH3: Measures of econom ic deprivation and ma rginalization will have a significant positive impact on both maleand female-perpetrated intimate partner homicide for both 1990 and 2000, due to the linkage in the literat ure of the role that economic de privation and marginalization has on crime. Change ModelsH4: The change from 1990 to 2000 in the theo retical measures of exposure reduction will correlate with the change in female-perpe trated intimate partner homicide from 1990-2000. H5: The change from 1990 to 2000 in the theoretical measures of backlash will correlate with the change in male-perpetrated inti mate partner homicide from 1990-2000. H6: The change from 1990 to 2000 in the theoretical measures of deprivation and marginalization will correlate with the change in both femaleand male-perpetrated intimate partner homicides from 1990-2000. Data is collected for the years 1990 and 2000 to address these hypotheses. The purpose of this study is to address the c ontradictory findings of previous research on intimate partner homicide and reach a better understanding of the tr ends in intimate partner homicide that have been witnessed over this time pe riod. To do this, this study draws on the exposure reduction and backlash perspectives, as well as ideas behi nd economic deprivation and marginalization.

PAGE 48

48 Figure 2-1. Detailed theoretical and conceptual model on exposur e reduction, backlash, economic deprivation and marginalization and maleand female-perpetrated intimate partner homicide. Female-Perpetrated Intimate Partner Homicide Male-Perpetrated Intimate Partner Homicide + + + Exposure Reduction % Divorced % Unmarried Households Shelter Rate Legal Assistance Rate Batterers Counseling Rate Referral Rate Economic Deprivation % Living in Poverty % Unemployed % HHs on Public Assistance Backlash (higher levels of equality) Ratio F/M Median Income Ratio F/M Education Ratio F/M Employment

PAGE 49

49 CHAPTER 3 DATA AND METHODOLOGY This se ction details the methodological strategy used in this study. To begin, the data sources and the unit of analysis are discussed. Next, all measures are operationalized. After examining the data three methodological issues are apparent. First, there is a number of missing cases in the dataset in regards to offender info rmation and domestic violence resource information. Techniques for dealing with this will be presented. Second, ther e is evidence of collinearity between predictors included in this study. To address this concern, principal components (specifically principal axis factoring) analys is is employed. Third, Ordinary Least Squares regression estimations are inappropriate due to th e skewed nature of the dependent variables. Poisson-based regression techniques will be utilized for the cross-sectional models, because the dependent variables are based on di screte counts of rare events and have a skewed distribution. These issues will be outlined in the followi ng pages. Unit of Analysis For this study, the unit of an alysis is cities with a popul ation of 100,000 or m ore in 1990. This definition of urban cities is consistent with the Bureau of Justice Statistics definition of large cities (Bureau of Justice Statistics 2006). The resulting urba n sample contains 200 cities that meet that criterion. Of those cities 178 cities were used in the analysis. Chattanooga, TN, Citrus Heights, CA, East Los Angeles, CA, Kansas City, KS, Lowell, MA, Metairie, LA, Omaha, NB, Overland Park, KS, Paradise, NV, Springfield, IL, St erling Heights, MI, Worchester, NY, and all cities in the state of Florida were droppe d from the analysis due to not reporting at all during either the period 1989-1991 or 1999-2001. Chicago, IL was the final city dropped due to being an extreme outlier. The average city population in 1990 is 323,445 ( =

PAGE 50

50 629,308), ranging from 100,217 to 7,322,564. The averag e city population in 2000 is 355,578 ( = 686,124), ranging from 95,658 to 8,008,278. Sources of Data Multip le data sources were used for the current study. The dependent variables are based on data from the Supplemental Homicide Files for the years 1989-1991 and 1999-2001 (Fox, 2005). U.S. Bureau of Census summary files 111 and 312 for 1990 and 2000 are the sources of data for the independent variables. Uniform Crime Reports: Police Employee (LEOKA) Data are the source of data for police force size (cons istent with work by Steffensmeier and Haynie 2000a; 2000b). Also, the 1991 and 1999-2000 Domestic Violence Service Directory, collected by the National Coalition Against Domestic Vi olence, is the source for the data on the availability of domestic violence services. These years were chosen because they were the years published by the National Coalition Against Do mestic Violence and represented 1990 and 2000 closest. Homicide Data The hom icide data used in this study were obtained from the Supplemental Homicide Reports (SHR). The SHRs are part of the Unif orm Crime Reporting Program (UCR). In addition to monthly criminal offense information compile d for UCR purposes, law enforcement agencies submit supplemental data to the Federal Bureau of Investigation on homicide incidents. The SHRs contain detailed, incident-level data on nearly all murders and nonnegligent manslaughters that have occurred in the United States for a given year. SHRs contain information for each 11 Summary File 1 presents 100percent population and housing data for the total population, for 63 race categories, and for many other race and Hispanic or Latino categories. See Appendix A for a more detailed definition. 12 Summary File 3 presents data on population and housing long-form subjects, such as income and education. See Appendix A for a more detailed definition.

PAGE 51

51 homicide incident, including info rmation on victim and offenders gender as well as victim and offender relationship information. Domestic Violence Service Data The data on dom estic violence services were obtained from the 1991 and 1999-2000 Domestic Violence Service Directory, collected by the National Coalition Against Domestic Violence. Unfortunately, the National Coalition Ag ainst Domestic Violence does not collect data ever year. These years were chosen because they represented 1990 and 2000 the closest. The information about each domestic violence program for each city is provided by the domestic violence programs themselves. For 1999-2000, th e National Directory includes information on whether 16 service categories are provided by the domestic violence program in each city. These service categories include: sh elter, safe homes, 24-hour hotlines, TTY/TDD (ability to communicate by telephone with women w ho are hearing impaired or deaf), counseling/advocacy, non-resident support groups childrens counseling/programs, rape/sexual assault services, emergency tran sportation, legal services/advocac y, wheelchair accessible, age limit for male children, counseling for batterers, shelter capacity, maximum length of stay, and languages other than English spoken. For 1991, th e National Directory includes information on whether 14 service categories are provided by the domestic violence program in each city. These service categories include: sh elter, safe homes, 24-hour hot lines, accept collect calls, counseling/advocacy, non-resident support groups childrens counseling/programs, fees for service, transportation provide d, legal services/advocacy, coun seling for batterers, wheelchair accessible, deaf services, and languages spoken besides English.

PAGE 52

52 Measures Dependent Variable The Supplem ental Homicide Reports (SHR) ar e the data source for the two dependent variables maleand female-perpe trated intimate partner homicide counts of arrest for the years 1990 and 2000. To determine the counts of maleand female-perpetrated intimate partner homicide for a given city and a given year, the counts were aggregated to the city-level for each year by the offenders gender and intimate partne r relationship. Consistent with other work (Messner et al., 2005; Villarreal, 200 4) arrest counts for each city are based on three year sums (1989, 1990, 1991; and 1999, 2000, 2001) to control for any fluctuations in reporting and relative low frequency of offending. For the current study only murd ers and nonnegligent manslaughters with single victim and single offender were incl uded. This is consistent with other research (Flewelling and Williams, 1999; Williams and Flewelling, 1987). Also, intimate partners include spouses, common-law spouses, ex-spouses, and boyfriends/girlfriends. Homosexual relationships were excluded for the purpose of this study, b ecause the male-female dynamic is of particular importance and because of the small number of cases (Dugan et al., 2003). Independent Variables U.S. Bureau of Census data for the year s 1990 and 2000 were used as information for the majority of the independent variables. Police Em ployee (LEOKA) Data were the source of data for the officer rate. For simplicity, in the followi ng section the independent variables are broken down into 4 categories of measures: exposure reduction, backlash, economic deprivation and marginalization, and control variables. Gender-speci fic indicators are utilized when appropriate. It is acknowledged that many of these measures may overlap and could possibly fall within another category of measures. Fo r the purpose of this study the m easures are divided with the

PAGE 53

53 purpose of capturing the three theoretical perspe ctives of exposure reduction, backlash, and economic deprivation an d marginalization. Exposure reduction measures To assess whether domesticity and domestic vi olence service availability had an exposure reducing effect on intimate partne r homicide, eight measures of exposure reduction are included in this study. These measures in clude: percentage of divorced males, percentage of divorced females, percentage of unmarried partner house holds, the number of shelters offered per 100,000 females, the number of legal services offe red per 100,000 females, the number of referrals offered per 100,000 females, and the number of male batterers counseling services offered per 100,000 males. Percentage of divorced males. The first meas ure of exposure reduction is the percent of males that are divorced. This was calculated by dividing the number of ma les aged 15 and older who are divorced by the total number of males ag ed 15 and older. This calculation was then multiplied by 100 to obtain a percent. 100 age) of years 15 Males No.( divorced) are that age of years 15 Males No.( Percentage of divorced females. The percent of females that are divorced was calculated the same as the percent of males divorced. The number of females aged 15 and older who are divorced was divided by the total number of females aged 15 and older and then multiplied by 100 to obtain a percent. 100 age) of years 15 Females No.( divorced) are that age of years 15 Females No.( Percentage of unmarried partner househol ds. Another measure of exposure reduction is the percent of unmarried partner households. Th is measure captures cohabitation by unmarried

PAGE 54

54 partners. Unmarried partners incl ude male householders living with a female partner and female householders living with a male partner. This m easure was calculated by dividing the number of unmarried partner households by the total numbe r of households and then multiplying this by 100 to obtain a percent. 100 ) Households no. Total( ) Households Partner Unmarried No.( Shelter rate. To measure the rate of the numbe r of shelters available to females a measure of the number of shelters for each city and year per 100,000 females 15 years and older was computed. The number of shelters was divided by the total number of females aged 15 years and older and then multiplied by 100,000 to obtain a rate. This is similar to Dugan et al. (1999)s calculation for the availability of hotli nes, counseling, and legal services. older) and years 15 aged females ofNumber ( year) andcity each for shelters of(Number x 100,000 Legal service rate. The legal service rate is de signed to measure the availability of legal assistance, such as in obtaining restraining orders, court accompaniment, legal clinics or advocacy. To measure the rate for the number of le gal services available to females a measure of the number of legal services fo r each city and year per 100,000 females 15 years and older was computed. The number of legal services was di vided by the total number of females aged 15 years and older and then multiplied by 100,000 to obtain a rate. older) and years 15 aged females ofNumber ( year) andcity each for services legal of(Number x 100,000 Number of referrals rate. Many cities may not have offere d a particular service (i.e., shelters, hotlines, counseling, childrens counseling, and legal services) but were able to refer individuals to other programs that offered the service. To capture this, a measure for the rate of

PAGE 55

55 the number of referrals per 100,000 females aged 15 and older for a particular city offered was calculated. The number of referrals was divided by the total numb er of females aged 15 years and older and then multiplied by 100,000 to obtain a rate. older) and years 15 aged females ofNumber ( year) andcity each for referrals of(Number x 100,000 Male batterers counseling rate. To measure the rate for the number of male batterers counseling services available to males a measur e of the number of male batterers counseling services for each city and year per 100,000 ma les 15 years and older was computed. The number of batterers counseling services was divided by the total numbe r of males aged 15 years and older and then multiplied by 100,000 to obtain a rate. older) and years 15 aged males ofNumber ( year) andcity each for services counseling batterers ofumber ( N x 100,000 Backlash measures Three ratio measures of female status rela tive to men were incl uded in this study to capture the ideas behind backlash. These measures have been used in previous research (Whaley and Messner, 2002). They include: the ratio of male to female median income, the ratio of the percentage of male employment to the percentage of female em ployment, and the ratio of the percentage of male education to the percentage of female education. A value of 1.0 on the ratio measures would represent equality between males and females. Values greater than 1.0 represent greater disadvantage of females to males. Ratio of male to female median income. The first backlash measure is the ratio of male to female median income. This measure was calculated by dividing the male median income by the female median income.

PAGE 56

56 income) median Female( income median Male( Ratio of the percentage of male employment to the percentage of female employment. The second backlash measure is the ratio of the percentage of male employment to the percentage of female employment. This measur e was calculated by dividing the percent of males aged 16 or older who were employed in the civilian labor force by the percent of females aged 16 or older who were employed in the civilian labor force. force)labor civilian in employed are whomoreor 16 aged females of %( force)labor civilain in the employed are whomoreor 16 aged males of %( Ratio of the percentage of males with college education to the percentage of females with college education. The final backlash measure is the ratio of the percentage of males with a college education to the percentage of female s with a college education. This measure was calculated by dividing the percent of males aged 25 and older with 4 or more years of college education by the percent of females aged 25 and ol der with 4 or more years of college education. education) college of years moreor 4 older with and 25 aged females of %( education) college of years moreor 4 older with and 25 aged males of %( Economic deprivation and marginalization measures Two gender-specific measures were incl uded in the models to capture economic deprivation and marginalization. These measur es included: percentage of males living in poverty, percentage of females living in poverty, percentage of males unemployed, and the percentage of females unemployed. A third meas ure of the percentage of households on public assistance was also included to measure economic deprivation. Percentage of males living in poverty. The first measure is the percentage of males living in poverty. It was calculated by dividing the nu mber of males below the officially defined

PAGE 57

57 poverty threshold in either 1989 or 1999 by the total number of males for whom poverty status was determined in 1989 or 1999 and then multiplied by 100 to obtain a percent. 100 1989/1999) in ed deterermin wasstatus poverty for whom Females Total( )level poverty below incomes with 1989/1999 in determined wasstatus poverty for whom Females (No. Percentage of females living in poverty. The next measure is the percent of females living in poverty. It was calculated by dividing the numb er of females below th e officially defined poverty threshold in 1989 or 1999 by the total number of female s for whom poverty status was determined in 1989 or 1999 and then multiplied by 100 to obtain a percent. 100 1989/1999) in ed deterermin wasstatus poverty for whom Males Total( )level poverty below incomes with 1989/1999 in determined wasstatus poverty for whom Males (No. Percentage of males unemployed. Another measure represen ting economic deprivation and marginalization is the percent of male s unemployed. This measure was calculated by dividing the number of males 16 yrs and older who are unemployed by the total number of males aged 16 and older and then multiplying this by 100 to obtain a percent. 100 age) of years 16 Males No.( ) unemployed are that age of years 16 Males No.( Females unemployed. Another measure representi ng economic deprivation and marginalization is the percent of females unemp loyed. This measure wa s calculated by dividing the number of females 16 years and older who are unemployed by the total number of females aged 16 and older and then multiplying this by 100 to obtain a percent. 100 age) of years 16 Females No.( ) unemployed are that age of years 16 Females No.( Percentage of households on public assistance. The final measure representing economic deprivation and marginalization is the percent of households on public assistance income. This

PAGE 58

58 measure was calculated by dividing the number of households on public assistance income by the total number of households and then multiplying this by 100 to obtain a percent. 100 ) Households of No.( ) assistance publicon Households of No.( Controls Four control measures were included in this study. They include: residential mobility, officer rate, log of the percen t Hispanic, and the south. Residential mobility. Residentia l mobility (i.e., percent of the population who have lived in current household for 5 years or less) was included in this study to represent a standard social disorganization measure. It was measured by dividing the number of pe ople in the population 5 years and older that had lived in a different house 5 years prior by the total population 5 years and older and then multiplying this by 100 to obtai n a percent. This variable has been suggested to influence intimate partner homicide (Brown e and Williams, 1989; Williams and Flewelling, 1988), particularly female-perpetrated intimate pa rtner homicide as a result of male aggression. Browne and Williams (1989) found that disruptions in life in the form of population mobility were associated with greater female-perpetrated intimate partner homicide. They suggest that mobility may isolate females from support networks that would gi ve females other avenues when conflicts arise instead of resorting to lethal violence. 100 age) of years 5 Population( 1995/1985) in house different ain lived that age of years 5 Population ( Officer rate. The officer rate was calculate d by dividing the total number of officers for each city by the total population and then this wa s multiplied by 1,000 to obt ain a rate of officers per 1,000 persons in the population. This calculation is consistent with other work (DeWees and Parker, 2003; Steffensmeier and Haynie, 2000a).

PAGE 59

59 Log Percent Hispanic. Another control variable is the percent Hispanic population. This variable is included in the models to account for the fact that research has shown that Hispanics are less likely to be involved in intimate partner homicide (Pau lsen and Brewer, 2000). In their analysis of spousal sex ratios of killing (i.e., the number of fe male intimate partner homicide perpetrators for every 100 male intimate partner homicide perp etrators), Paulsen and Brewer (2000) found that the largest disp arities between male-perpetrated intimate partner homicide and female-perpetrated intimate partner homicide was for Hispanics. One reason could be that domestic violence and homicide is viewed differently in the Hispanic culture. To calculate the percent Hispanic the total Hispanic population wa s divided by the total population and multiplied by 100 to obtain a percent. The natu ral logarithm of the percent Hisp anic was then calculated due to the skewed nature of the independent variable. 100 )Population Total( )population Hispanic Total(Ln South. A dummy coded variable for th e south is used in the present analysis to control for any regional differences in intimate partne r homicide offending. Research has shown that homicide rates are higher in the south (Nisbett and Cohen, 1996) and th at partner homicide is higher in the south as well (Browne and W illiams, 1989; Nisbett and Cohen, 1996). This measure is based on Census Bureau definitions of southern regions (i.e., coded as 1 for cities located in the southern region and 0 for ci ties not located in the southern region). Missing Data Missing information in the homicide data was problematic. For the most part this is caused by the fact that partic ipation in the Supplemental Ho micide Reports program is 1,000 1990/2000) in population Total( officers) police ofnumber (Total

PAGE 60

60 completely voluntary. Therefore, some law enfor cement agencies fail to report their homicide incidents to the FBI. According to the Bureau of Justice Statistic s (2006) the SHRs are just over 90% complete. Though the coverage is high, ther e are still a number of homicides that go unaccounted for. This underreporting by law enfo rcement agencies was corrected with an adjustment factor based on the to tal number of homicide incidents reported to the Uniform Crime Reporting program (see Fox, 2004 for discussion on weight calculation; Williams and Flewelling, 1987). That is, SHR records were adjusted so that State and national total counts of murder and nonnegligent manslaughter matched UCR es timates. It is importan t to note that this weighting process assumes that the missing record s are not systematically different from those that have been reported to the FBI. In additi on, there is a problem with using offender data because there is a growing problem with unsol ved murders. Ignoring homicides with missing offender information understates homicide offend ing. A weighting strategy based on available information about the victims murdered in both solved and unsolved homicides is used in the present study to adjust for missing offender data (see Fox, 2004 for disc ussion). In the current analysis all cases are weighted to reflect U CR estimated U.S. homici de counts and imputed offender data. Missing data were also an issue with the domestic violence service data. There were instances where a program in a city did not report whether or not a particul ar service was offered. In these cases, the mean of that particular se rvice across cities was imputed. For the domestic violence service variables included in the mode l, a dummy variable for whether the mean was imputed for a city was included in the final m odels (i.e., 1 = mean imputation, 0 = no mean imputation for a city) to determine if the missing values were missing at random. Results showed

PAGE 61

61 that the missing values were missing at random and therefore the mean imputations were appropriate to use. Multicollinearity There are two ways to assess mulitcollinearity: (1) examining bivariate correlations between the variables and, (2) examining varian ce inflation factors. The bivariate correlation matrices for all variables in 1990 and 2000 are pr esented in Appendix B and Appendix C. After examination of the correlation matric es it is evident that there is collinearity and partiality among the regressors. While there is no clear standard, researchers generally ac cept a correlation greater than .500 to mean that there may be problems with mulitcollinearity. Examination shows that there are correlations between the independent va riables that are above .500. For instance, in 1990 and 2000 the percent of female unemployment and the percent of females living in poverty are highly correlated (r =.782 and r = .816 respectiv ely). The percent of females living in poverty and the percent of households on pu blic assistance are co rrelated (r = .786 and r = .751). The percent of females unemployed and the percent of households on public assistance income are also correlated (r = .793 and r = .751). Furtherm ore, the percent of male unemployment and the percent of males living in povert y are highly correlated (r=.786 and r = .788 respectively). The percent of males living in pove rty and the percent of househol ds on public assistance are correlated (r = .766 and r = .757). The percen t of males unemployed and the percent of households on public assistance income are also correlate d (r = .863 and r = .816). To be sure that multicollinearity is not a problem variance inflation factors were calculated. Generally it is accepted that VIFs gr eater than 4 represent multicollinearity (Fisher and Mason, 1981:109). However, some have sugges ted that a value of 10 or greater indicates multicollinearity (Ott and Longnecker, 2001:652 ; Pindyck and Rubinfeld, 1998). Variance inflation factors for all indepe ndent variables are presented in Table 3-3. For the female-

PAGE 62

62 perpetrated intimate partner homicide model in 1990, the percent of females living in poverty, the percent of females unemployed, and the perc ent of households on public assistance have a variance inflation factor greater than the value of 4 (5.465, 4.887, 5.192, respectively). None of the other independent variables have a variance inflation factor grea ter than 4. As in the femaleperpetrated intimate partner homicide model, the same three variables have variance inflation factors greater than four th e percent of males living in poverty, th e percent of males unemployed, and the percent of households on public assistance (4.697, 6.970, 4.871, respectively). Again, none of the other indepe ndent variables have a variance inflation factor greater than 4. For the female-perpetrated intimate partner homicide model in 2000, the percent of females living in poverty and the percent of households on public assistance have variance inflation factors that are greater than the valu e of 4 (4.827 and 4.496 respectively). In addition, the percent of females that are unemployed has a vari ance inflation factor that is very close to the value of 4 (3.834). None of the ot her independent variables have a variance inflation factor close to or greater than 4. For the male-perpetrated intimate partner homicide model in 2000, the percent of males living in pove rty, the percent of males une mployed, and the percent of households on public assistance all have variance inflation factors greater than 4 (4.040, 4.494, and 4.865 respectively). As with the female-perpetr ated intimate partner m odel, none of the other independent variables have a variance inflat ion factor close to or greater than 4. Due to relatively high variance inflation fact ors, the percent living in poverty, the percent unemployed, and the percent of households li ving on public assistance were included in a confirmatory factor analysis to determine if these variables loaded together. An index was developed using principle axis factoring with varimax rotation. The principal components

PAGE 63

63 analysis for 1990 is displayed in Table 3-4 and th e analysis for 2000 is presented in Table 3-5. The measures of percent unemployment, percent poverty, and the percent of household living in poverty loaded together in both the male-perpetrated and female -perpetrated models for both time periods. An index was create d that is referred to as the male economic deprivation index and the female economic deprivation index. The simplified correlation matrices for 1990 and 2000 are presented in Appendix C and Appendix D. In addition, variance inflation factors for the simplified models are disp layed in the Appendix F.13 The bivariate correlation matrices for the domestic violence service variables clearly show that many of the individual domestic violen ce services are highly collinear with each other (See Appendix G). Four variables were of particular interest because th ey represented service information that were collected for both time periods and contained the most information. These variables included whether or not shelters, legal services, male ba tterers counseling, and referral services were offered for a city.14 In the final analyses composite measures of these variables are utilized. Examining variance inflation factors revealed that collinearity was not an issue between these variables. Methodology Descriptive Statistics The means and standard deviations for all th e variables in 1990 and 2000 can be seen in Table 3-1. Examining the dependent variables shows that both maleand female-perpetrated 13 The variance inflation factors for the simplified male-per petrated and female-perpetrated models in 1990 and 2000 reveal that after principal components is considered and the index made, the variance inflation factors do not exceed 4.0. 14 Initially an index was created that measured the availability of domestic violence services, which included the number of programs per 100,000 females offered for each city and the number of different domestic violence services offe red (i.e., whether the domestic violence program provided shelters, a 24-hour hotline, counseling/advocacy, childrens counseling, and legal/services/advocacy) pe r 100,000 females. The index for the availability of domestic violence services did not obtain si gnificance in the models. However, once the services were measured separately individual services reached significance. Therefore, indivi dual services were included in the final analysis, instead of the combined index.

PAGE 64

64 intimate partner homicides decreased from 1990 to 2000. The mean of female-perpetrated intimate partner homicides (i.e., 3 year sum of counts) in 1990 is 6.553 ( = 10.864). In comparison, the mean of female-perpetrated intimate partner homici des in 2000 is 2.970 ( = 4.779). The mean of male-perpetrated intim ate partner homicides in 1990 is 11.569 ( = 18.378) and the mean in 2000 is 8.770 ( = 13.929). For the indicators of exposure reduction, both the percent of males divorced and females divorced decreased from 1990 to 2000, whereas th e percent of unmarried partner households increased. In addition, the rate of the number of shelters available to females and the rate of the number of legal services avai lable to females increased from 1990 to 2000 (.354 to 1.49 and .313 to 1.42, respectively). In contrast the rate of the number of male batterers counseling programs and the rate of the number of referral services offered decreased from 1990 to 2000 (.746 to .552 and .089 to .075, respectively). Examining the indicators of backlash shows that the disparities between male and female educational attainment and empl oyment decreased (1.33 to 1.14 and 1.24 to 1.19, respectively), whereas the disparity be tween male and female income increased from 1990 to 2000 (1.42 to 1.55). This suggests that over time, the gap betw een males and females educational attainment and employment are becoming more narrower, while at the same time the gap between males and females median income is widening. For the indicators of economic deprivation a nd marginalization, both the male and female economic deprivation indices decreased fr om 1990 to 2000. In 1990 the mean of the female economic deprivation index was 29.38 ( = 11.97) and in 2000 it was 15.88 ( = 6.22). In 1990 the mean of the male economic deprivation index was 26.97 ( = 11.28) and in 2000 was 14.88 ( = 5.88).

PAGE 65

65 Table 3-2 reports the paired sample t-test values to show whether the variable means are significantly different between 1990 and 2000. All the variable m eans are significantly different except for the mean of the num ber of referral serv ices per 100,000 females. In summary, statistics show that both maleand female-perpe trated intimate partner homicide decreased from 1990 to 2000. In addition, for both males and fe males, economic depriv ation and the percent divorced decreased from 1990 to 2000, as well as the ratio of male to females educational attainment, the ratio of male to female employm ent, and the rate of male batterers counseling programs. In contrast, the ratio of male to female median income, percent of unmarried households, the rate of the number of shelters, the rate of the nu mber of legal services, percent Hispanic, residential mobility, and the officer rate increased from 1990 to 2000. Analytical Plan Cross-Sectional Analyses Cross-sectional regression m odels are used to analyze the impact of the theoretical concepts at two points in time (1990 and 2000). Po isson regression models will be used for the current study, because the dependent variables ar e based on discrete counts of rare events and have a skewed distribution. In addition, the depe ndent variables include a number of cities with zero counts, all of which make OLS inappropriate to use (Osgood, 2000; Osgood and Chambers, 2000). The poison regression model is the simplest re gression model for count data. However, in the poison regression model the variance is forced to equal the mean, which is often not the case with count data. After examining the mean and vari ance of the current dist ribution it is clear that the mean and the variance are not equal (Cameron and Trivedi, 1998). The negative binomial regression model is more appropriat e to use because it allows for overdispersion of the data (i.e.,

PAGE 66

66 where the variance exceeds the mean) (Gardne r et al., 1995; Osgood, 2000). In the negative binomial distribution the mean and the variance are as follows: )/( )Var( )(2k [1/k is the dispersion parameter] Though negative binomial models take into a ccount overdispersion, they usually under predict the number of zero counts in the dependent variable. Since both maleand femaleperpetrated intimate partner homicides are rare ev ents and therefore result in a number of zero counts across cities, zero-inflated negative binomia l regression is more appropriate and will be utilized. Furthermore, the vuong test produced a z-value of 14.55 (p = .000) for the maleperpetrated intimate partner homicide model and a z-value of 15.05 (p = .000) for the femaleperpetrated intimate partner homicide model. A significant value means that the zero-inflated model is a better fit of the data. Excess zero c ounts may cause overdispe rsion and zero-inflated models take this into consideration (Cameron and Trivedi, 1 998). In both models the intimate partner homicide count was converted into an equivalent of a rate by including the logged gender-specific population size multiplied by a factor of 3 (i.e., to account for the 3 year pooled dependent variable counts) as an exposu re variable in the model (Agresti, 1996). Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation Using STATA, version 9.1, pooled cross-sectio nal time series analyses with fixed effect estimation are used to estimate the impact of ke y theoretical constructs on changes from 1990 to 2000 in maleand female-perpetrated intimate pa rtner homicide. An Ordinary Least Squares fixed effects regression15 was employed as follows 15 Negative binomial fixed effects and random effects regression models were estimated, due to the large number of zero counts and skewed distribution fo r the dependent variables. However, the model for the female-perpetrated intimate partner homicide data would not allow for the es timation of the Hausman test because the data failed to

PAGE 67

67yit = xit + i + t + it i = 1, N; t =1, T where xit is the vector of covariates for both cities (i) and time periods (t), with dummy variables controlling for city (i) and time trends ( t). A time variable for year 1990 was included in the analysis to control for period effects, with the year 2000 as the reference period. There are advantages to this type of analysis. For instan ce, pooled cross-sectional time series analysis increases the number of observations (i t) and also allows one to model time and space and generalize between them. Due to the skewed nature of the dependent variables both the female-perpetrated intimate partner homicide count over 3 years and the male-perpetrated intimate partner homicide count over 3 years was transformed using a logarithmi c transformation. An issue arose with the logarithmic transformation when the count of inti mate partner homicide equaled zero. The log of a zero equals To fix this problem, all zero counts we re also coded as zero for the logarithmic transformation of the intimate partner homicide count. To account for this a dummy variable was included in the models that represented zero coun ts for intimate partner homicide (See Hausman, Hall, and Griliches, 1984 for rationale). Also, the log of the population 15 and older multiplied by a factor of 3 was included in the m odels to control for the population size. Both fixed-effects and random-effects models are appropriate to use for pooled crosssectional analysis, however for the current study the fixed effect estimation was used for a meet the asymptotic assumptions of th e Hausman test. The assumptions of th e Hausman specification test are often hard to meet. The Hausman test assumes that one of th e estimators is efficient (i.e., has minimal asymptotic variance). This assumption is violated if observations are clustered or pweighted or if the model is misspecified (STATA 2008). For comparison purposes, both female-perpetrated and male perpetrated intimate partner homicide were modeled using OLS fixed effects regression.

PAGE 68

68 variety of reasons: (1) it expre sses variables only in te rms of change within a unit (Parker, 2004), which is appropriate because the current study seeks to determine the relationship between change in key theoretical construc ts and change in the dependent va riable for a given city; (2) it controls for all omitted and included time invari ant variables (Allison, 1994; Johnson, 1995); and (3) as described below, the Hausman specification test indicated that the fixed effects model was more appropriate to use than the random effects model. The results of the Hausman specification test (1978) revealed that fixed effects estimation was more appropriate than random effects estima tion for both the male-perpetrated and femaleperpetrated intimate partner homicide multivariate models. The Hausman specification test examines the null hypothesis of no difference be tween the parameter estimates of the more efficient random effects model and the fixed effects model, whereas the alternative hypothesis states that there is a significant difference be tween the random effects models parameters and the fixed effects models parameters. This difference indicates bias in the random effects estimation and thus the fixed e ffects estimation is preferred (Greene, 2000; Hausman et al., 1984; Hsaio, 1986; Maddala, 1983). The Hausman specification test pr oduces a Chi-square statistic, such that a significant p-value indicates that fixe d effects estimation is more appropriate and an insignificant p-value indicates that random effects estimation is more appropriate.

PAGE 69

69 Table 3-1. Means, standard deviat ions (in parentheses) for all vari ables, 1990 and 2000 (N= 178). 1990 2000 Female-perpetrated intimate partner homicide 6.553(10.864) 2.970(4.779) (counts summed over 3 years) Male-perpetrated intimate partner homicide 11.569(18.378) 8.770(13.929) (counts summed over 3 years) Percent female divorced 13.31(4.24) 12.25(2.08) Percent male divorced 9.92(3.12) 9.31(2.03) Percent unmarried households 3.91(1.54) 5.25(1.068) Male batterers counseling per 100,000 males .746(1.09) .552(.738) Shelters per 100,000 females .354(.276) 1.49(.909) Legal service programs per 100,000 females .313(.289) 1.42(1.167) Referral services per 100,000 females .089(.243) .075(.187) Ratio of male to female education 1.33(.140) 1.14(.098) Ratio of male to female income 1.42(.173) 1.55(.180) Ratio of male to female employed 1.24(.106) 1.19(.086) Female economic deprivation Index 29.38(11.97) 15.88(6.22) Male economic deprivation index 26.97(11.28) 14.88(5.88) Percent hispanic 13.40(15.67) 18.61(18.17) Residential mobility 50.22(7.88) 51.40(6.71) Officer rate per 1,000 1.98(.910) 2.22(1.04) South .303(.461) .303(.461)

PAGE 70

70 Table 3-2. T-test scores for change from 1990 to 2000 (N=178). Mean 1990 Mean 2000 Mean Diff t Female-perpetrated intimate partner homicide 6.55(10.86) 2.97(4.78) -3.58 -5.83** Male-perpetrated intimate partner homicide 11.57(18.38) 8.77(13.93) -2.80 -4. 32** Percent female divorced 13.31(4.24) 12.25(2.08) -1.06 -3.45** Percent male divorced 9.92(3.12) 9.31(2.03) -.61 -2.76** Percent unmarried households 3.91(1.54) 5.25(1.07) 1.34 12.17** Male batterers counseling per 100,000 males .75(1.09) .55(.74) -.20 -2.99** Shelters per 100,000 females 35(.28) 1.49(.91) 1.14 18.11** Legal service programs per 100,000 females .31(.29) 1.42(1.17) 1.11 13.29** Referral services per 100,000 females .09(.24) .08 (.19) -.01 -.695 Ratio of male to female education 1.33(.14) 1.14(.10) -.19 -28.94** Ratio of male to female income 1.42(.17) 1.55(.18) .13 12.09** Ratio of male to female employed 1.24(.11) 1.19(.09) -.05 -5.97** Female economic deprivation Index 29.38(11.97) 15.88(6.22) -13.5 -13.64** Male economic deprivation index 26.97(11.28) 14.88(5.88) -12.09 -2 4.85** Percent hispanic 13.40(15.67) 18.61(18.17) 5.21 14.03** Residential mobility 50.22(7.88) 51.40(6.71) 1.18 2.76** Officer rate 1.98(.910) 2.22(1.04) .24 4.74** + p < .10 p < .05 ** p < .01

PAGE 71

71 Table 3-3. Variance inflation factors for all variables included in the models (N=178). F-perp90 M-Perp90 F-perp00 M-perp00 Percent female divorced 1.228 1.464 Percent male divorced 1.204 1.666 Percent unmarried households 1.366 1.360 2.830 2.940 Male batterers programs per 100,000 males 1.245 1.241 1.405 1.399 Shelters per 100,000 females 2.198 2.138 2.392 2.387 Legal service programs per 100,000 females 2.101 2.129 2.874 2.861 Referral services per 100,000 females 1.179 1.163 1.300 1.323 Ratio of male to female median income 1.883 1.818 2.110 2.159 Ratio of male to female education 1.580 1.550 2.080 2.120 Ratio of male to female employment 1.443 1.468 2.446 2.514 Percent female unemployment 4.887 3.834 Percent male unemployment 6.970 4.494 Percent females in poverty 5.465 4.827 Percent males in poverty 4.697 4.040 Percent of households on public a ssistance 5.192 4.871 4.496 4.865 Percent hispanic 1.591 1.463 2.359 2.302 Residential mobility 1.976 2.219 1.696 1.718 Officer rate per 1,000 1.796 1.845 2.098 2.056 South 1.720 1.629 2.246 2.063

PAGE 72

72 Table 3-4. Principal components analys is after varimax rotation 1990 (N=178). VARIABLES Female Model Male Model Percent below Poverty .881 .836 Percent Unemployed .888 .941 Percent of Households on Public Assistance .893 .917 Eigenvalue 2.574 2.611 % Variance Explained 85.81% 87.037% Note: only factor loadings greater than 0.5000 are reported

PAGE 73

73 Table 3-5. Principal co mponents analysis after va rimax rotation 2000 (N=178). VARIABLES Female Model Male Model Percent below Poverty .903 .856 Percent Unemployed .903 .921 Percent of Households on Public Assistance .832 .886 Eigenvalue 2.546 2.574 % Variance Explained 85.851 85.809% Note: only factor loadings greater than 0.5000 are reported

PAGE 74

74 CHAPTER 4 RESULTS The following chapter presents the results of the cross-sect ional analyses and the pooled cross-sectional time series analyses. Cross-Sectional Analysis Tables 4-1 and 4-2 present the analysis from the zero-inflated negative binomial regression equations for maleand female-perpe trated intimate partner homicide counts in 1990 and 2000. Table 4-3 summarizes the findings. So me interesting findings emerge regarding exposure reduction, backlash, and economic de privation and marginal ization on maleand female-perpetrated intimate partner homicide. The following section will begin by explaining the findings from 1990 for both the fe male-perpetrated intimate partner model and the maleperpetrated intimate partner model and then will end with the cross-sectional results from the 2000 models. Female-Perpetrated Intima te Partner Homicide 1990 After examining the female-perpetrated in timate partner homicide model for 1990 it is clear that the first hypothesis is not supported. The only exposure reduction measure that showed slight significance was the percent of females divorced in cities (p < .1). It was found that as the percent of females divorced in c ities increases, the counts of fe male-perpetrated intimate partner homicide increase as well. Specifically, a one standard deviation incr ease in the percent of females divorced results in a 13% increase in female-perpetrated intimate partner homicides {[exp (.029 x 4.24) = 1.130]}. However, this is not in the predicted direc tion. According to the exposure reduction perspective, re ducing the exposure between violent couples (i.e., an increase in female % divorced) should reduce the likeli hood of female-perpetrated intimate partner homicide. Surprisingly, none of the other exposure reduction indi cators, including the measures

PAGE 75

75 for domestic violence service availability, are significantly related to female-perpetrated intimate partner homicide. Results are consistent with the second hypothe sis. None of the backlash indicators are significantly associated with female-perpetr ated intimate partner homicide in 1990 when controlling for a number of other stru ctural factors. It was hypothesized that as females status in terms of income, employment, and educational a ttainment increased male-perpetrated intimate partner homicide would increase as well. Backlash indicators were not predicted to be related to female-perpetrated intimate partner homicide. Consistent with the third hypothesis, econom ic deprivation and marginalization is significantly related to female-perpetrated intimate partner homicide in the predicted direction. That is, the higher the female economic deprivation index in large cities, th e higher the counts of female-perpetrated intimate partner homicide. Specifically, a one standard deviation increase in the economic deprivation index results in a 40% increase in female-perpetrated intimate partner homicides {[exp (.028 x 11.97 = 1.398]}. Just as predicted, it was hypot hesized that economic deprivation and marginalization would have a si gnificant impact on female-perpetrated intimate partner homicide, such that increases in econom ic deprivation and marginalization would be related to higher female-perpetrated intimate partner homicide in cities. In addition, despite findings from previous research none of the control variables are significant in the model. Residential mobility, the officer rate, log of the percent Hispanic, or whether or not the city was located in the south ar e not significantly related to female-perpetrated intimate partner homicide in 1990. However, all of the coefficien t signs are in the predicted direction.

PAGE 76

76Male-Perpetrated Intimate Partner Homicide 1990 After examining the male-perpetrated intimate partner homicide model for 1990 it is clear that contrary to the fi rst hypothesis, a number of the exposure reduction measures are related to male-perpetrated intimate partne r homicide. It was hypothe sized that exposure reduction indicators would impact female-perpetrated intimate partner homicide by reducing the contact between violent partners. However, th ree exposure reduction indicators are associated with male-perpetrated intimate partner homicid e. The number of legal services per 100,000 females is negatively associated with male-perpe trated intimate partner homicide. Specifically, a one standard deviation increase in the number of legal servi ces per 100,000 females results in a 15% decrease in male-perpetrated intimate part ner homicide counts {[exp (-.578 x .289 = .846]}. Moreover, the percent of males divorced is positively related to male-perpetrated intimate partner homicide. That is, a one standard deviat ion increase in the pe rcent of males divorced results in an 18% increase in male-perpetrated intimate partner homicide counts {[exp (.053 x 3.12 = 1.179]}. This contradicts the ideas behind exposure reduction. One would assume that as the exposure between partners decreased, particularly through divorce, in timate partner homicide would decrease as well. Furthermore, the number of shelters per 100, 000 females is positively related to maleperpetrated intimate partner homicide. That is, a one standard deviation incr ease in the number of shelters per 100,000 females results in a 26% in crease in male-perpetr ated intimate partner homicide counts {[exp (.828 x .276 = 1.257]}. No ne of the other exposure reduction indicators reached statistical significance. Also surprisingly, there is a lack of support for the second hypothesis. It was hypothesized that the backlash i ndicators would be significantly related to male-perpetrated intimate partner homicide, such that an increase in females status w ould be related to an

PAGE 77

77 increase in male-perpetrated intimate partner homicide as males attempt to gain control back in their relationships. However, none of the backlash indicators are signific antly related to maleperpetrated intimate partner homicide. Consistent with the third hypothesis, economic deprivation is significantly related to male-perpetrated intimate partner homicide counts in cities for 1990. The higher the male economic deprivation index in cities, the higher the counts of male-perpetrated intimate partner homicide. Specifically, a one standa rd deviation increase in the male economic deprivation index results in a 20% increase in male-perpetrated intimate partner homicide counts {[exp (.016 x 11.28 = 1.198]}. This is not surprising and is in the predicted direction. The only control variable that is significant in the model is whether or not the city was located in the South. Acco rding to the results, if the city is located in the South the likelihood of male-perpetrated intimate partner homicide incr eases. Other research has found similar results. For instance, DeWees and Parker (2003) found that female victimization is higher in the South. In summary, there are clear differences between which structural variables account for maleand female perpetrated intimate partner homicide in 1990. Exposure reduction measures influence both maleand female-perpetrated in timate partner homicides. However more of the measures influence male-perpetrated intimate part ner homicide and in particular, a couple of the domestic violence service indicators. None of th e backlash measures influence maleor femaleperpetrated intimate partner homicide in 1990. As predicted, economic deprivation and marginalization influence both maleand female-perpetrated intimate partner homicide in 1990. Female-Perpetrated Intima te Partner Homicide 2000 Examining the female-perpetrated intimat e partner homicide model for 2000 reveals some very different results than the 1990 m odel. In 2000, two of the exposure reduction indicators are significantly related to female-perpetrated intimate partner homicide. Inconsistent

PAGE 78

78 with the first hypothesis, the number of shelters per 100,000 females in cities is positively related to female-perpetrated intimate partner homicide. That is, a one standard deviation increase in the number of shelters per 100,000 fe males results in a 55% increase in female-perpetrated intimate partner homicide counts {[exp (.48 x .909 = 1. 547]}. Although a somewhat surprising finding, this was also found in the male-perpetrated intimate partner homicide model for 1990. In the predicted direction, the number of le gal services per 100,000 females in cities is negatively related to female-perpetrated intimat e partner homicide. That is, a one standard deviation increase in the number of legal services per 100,000 fe males results in a 28% decrease in female-perpetrated intimate partner homicid e counts {[exp (-.279 x 1.167 = .722]}. This was also seen in the male-perpetrated model for 1990. Contrary to the second hypothe sis, all three backlash indi cators are significantly related to female-perpetrated intimate partner homicide. Results suggest the higher the ratio of male to female median income (i.e., females status relativ e to males decreases) in cities, the lower the female-perpetrated intimate partner homicide. Sp ecifically, a one standard deviation increase in the ratio of male to female median income re sults in a 19% decrease in female-perpetrated intimate partner homicide counts {[exp (-1.156 x .180 = .812]}. Also, the ratio of male to female employment is negatively related to female-perpe trated intimate partner homicide. That is, the higher the ratio of male to female employment (i.e., gap widens) in cities, the lower femaleperpetrated intimate partner homicide. Specifically, a one standard deviation increase in the ratio of male to female employment results in a 23% decrease in female-perpetrated intimate partner homicide counts {[exp (-3.081 x .086 = .767]}. However, in the opposite dire ction, the ratio of males to females educational attainment is positively related to female-perpetrated intimate partner homicide. That is, a one standard deviati on increase in the ratio of males to females

PAGE 79

79 educational attainment (i.e., gap widens) results in a 22% increase in female-perpetrated intimate partner homicide counts {[e xp (2.042 x .098 = 1.222}. Not supporting the third hypothesis, the economic deprivation index is not significant in the 2000 model. This is surprising consideri ng that the economic deprivation index was significant at the .001 le vel in the 1990 model (p < .001). However, examining descriptive statistics does show a significant decrease in the female economic deprivation index from 1990 to 2000. The lack of significance in the economic deprivation index in 2000 may be due to the differences in the U.S. economy in the 1980s an d 1990s. The U.S. experienced a recession in the late 1980s until the early 1990s. This was followed by a substantial period of economic growth throughout the 1990s. The only control variable that is significant in the female-perpetrated model is the officer rate. That is, the higher the officer rate pe r 1,000 persons, the lower the count of femaleperpetrated intimate partner homicide. This sugges ts that more police officers would be able to react quicker to domestic violence calls and ther efore prevent many intimate partner homicides. Male-Perpetrated Intimate Partner Homicide 2000 The main difference between the male-perpetrated intimate partner homicide model in 2000 compared with 1990 is that the economic deprivation index is not significant. This is surprising, considering that economic depriva tion was significant at the .001 level in 1990 (p<.001). As stated earlier, the lack of significance in the economic deprivation index in 2000 may be due to the differences in the U.S. economy in the 1980s and 1990s. However, all other results are similar. Of the exposure reduction indicators, the per cent of males that are divorced in cities is positively related to male-perpetrated intimate pa rtner homicide. That is, the higher the percent of males divorced in cities, the higher the count of male-perpetrated intimate partner homicide.

PAGE 80

80 Specifically, a one standard deviation increase in the percent of males divorced results in a 12% increase in male-perpetrated intimate partner ho micide counts {[exp (.058 x 2.03 = 1.124]}. This follows the findings from the 1990 male-perpetrated model. Also, the number of shelters per 100,000 female s is positively related to male-perpetrated intimate partner homicide. That is, the higher the number of sh elters per 100,000 females in cities, the higher the count of male-perpetrated intimate partner homicide. Specifically, a one standard deviation increase in the number of shelters per 100,000 females results in a 20% increase in male-perpetrated intimate partne r homicide counts {[exp (.202 x .909 = 1.202]}. Though contradictory to the expos ure reduction perspective, it is not all that surprising considering that the same finding showed up in the 1990 male-perpetrated model. Moreover, the number of legal services per 100,000 females is negatively related to maleperpetrated intimate partner homicide. Specifically, the higher the number of legal services per 100,000 females in cities, the lower the count of male-perpetrated intimate partner homicide. That is, a one standard deviation increase in the number of legal services per 100,000 females results in a 25% decrease in male-perpetrated intimate partner homicide counts {[exp (-.244 x 1.167 = .752]}. This too, was a significant finding in 1990. Again surprisingly, none of the backlash i ndicators are related to male-perpetrated intimate partner homicide in 2000. It appears that m easures of females stat us relative to males are not predictive of male-perpetrated intimate partner homicide. Also, none of the control indicators are significant predictors of male-perpetrated intimate partner homicide. In summary, many of the structural indicat ors influence maleand female-perpetrated intimate partner homicide. Exposure reduction measures influence both maleand femaleperpetrated intimate partner homicides, with two domestic violence service variables

PAGE 81

81 significantly related to both; howev er both are not in the predicted direction. Contrary to what was predicted, backlash indicators influence fe male-perpetrated intimate partner homicide in 2000, but are not related to male-perpetrated inti mate partner homicide in 2000. Also opposite of the predictions, economic deprivation and margin alization did not influence either maleor female-perpetrated intimate partner homicide. Pooled Cross-Sectional Time Series Analysis: Fixed Effect Estimation Table 4-4 and 4-5 displays the results from the pooled cross-sectional Ordinary Least Squares fixed-effects regression models. Table 4-6 summarizes the findings. Interestingly, results suggest that the theoretical constructs explain changes in female-perpetrated intimate partner homicide, but not male-perpetrated intim ate partner homicide. Specifically, changes in female-perpetrated intimate partner homicide co unts are associated with variables used in exposure reduction explanations, but not with indicat ors used in backlash explanations. That is, cities that are characterized by increases in th e percent of females that are divorced are also characterized by decreasing counts of female -perpetrated intimate partner homicide. Surprisingly, the only domestic violence service variable that is significant in the femaleperpetrated intimate partner homicide model is the number of shelters per 100,000 females and the relationship between the cha nge in this variable and the change in female-perpetrated intimate partner homicide is in the opposite directi on of what was originally predicted. Cities that show increases in the number of shelters per 100,000 females al so show increases in femaleperpetrated intimate partner homicide counts. No significant relations hip is found between changes in the percent of unmarried partner households, the number of legal services available to females, the number of batterers counseling programs available to males, or the number of referral services offered to females and the fe male-perpetrated intimate partner homicide count.

PAGE 82

82 None of the changes in the indicators for backla sh are significantly related to the change in female-perpetrated intimate part ner homicide from 1990 to 2000. Furthermore, changes in the female economic deprivation index are significantly associated with changes in female-perpetrated intimate partner homicide counts. This suggests that intra-city changes in female economic depriv ation were strongly associated with within-city increases in female-perpetrated intimate partner homicide counts in the model. Cities that are characterized by increases in the percent of fema les in poverty, the percent of females that are unemployed, and the percent of households on public assistance from 1990 to 2000 are also characterized by increases in female-perpe trated intimate partner homicide counts. Overall, there is mixed support for the changes in the theoretical perspectives on the changes in female-perpetrated intimate partner ho micide counts from 1990 to 2000. A change in economic deprivation within cities is significantly associated with changes in female-perpetrated intimate partner homicide counts. Also, changes in two indicators of exposure reduction were related to changes in female-per petrated intimate partner homic ide counts; however one was in the opposite direction of what was predicted. Interestingly, support is not found for the ch anges in the variables identified by the theoretical perspectives on the changes in ma le-perpetrated intimate partner homicide counts from 1990 to 2000. Maleand female-perpetrated intimate partner homicides are unique events with different motives that requ ire a unique set of explanations. From this analysis, it appears that changes in factors that in fluence female-perpetrated intimat e partner homicide from 1990 to 2000 have no influence on the perpetration of ma le intimate partner hom icide from 1990 to 2000. Research examining these trends has acknowledg ed the greater support am ong certain factors in explaining female-perpetrated intimate partner ho micide. For instance, Dugan et al. (1999) find

PAGE 83

83 that indicators of exposure reduction primarily influence male intimate partner homicide victimization.

PAGE 84

84 Table 4-1. Zero-inflated negativ e binomial regression equations w ith coefficients (and Z-Scores) for gender-specific intimate partner homicide 1990. Female Model (N=178) Male Model (N=178) Exposure Reduction Percent divorced (gender-specific) .029 (1.84)+ .053 (3.26)** Percent unmarried households -.024 (-.62) -.024 (-.82) Male batterers programs ra te -.017 (-.27) .028 (.59) Shelter rate -.015 (-.04) .828 (2.72)** Legal service rate -.001 (-.00) -.578 (-1.99)* Referral services rate -.060 (-.20) -.202 (-.80) Backlash Ratio of male to female education .396 (-.75) -.221 (-.51) Ratio of male to female income .969 (1.39) -.041 (-.10) Ratio of male to female employment -.514 (-.83) -.256 (-.46) Economic Deprivation Economic deprivation index (gender-sp ecific) .028 (3.92)** .016 (2.80)** Controls Percent Hispanic (log) -.057 (-1.23) -.041 (-1.06) Residential mobility .010 (.88) -.002 (-.23) Officer rate -.069 (-.75) -.050 (-.68) South .255 (2.02) .332 (3.13)** Constant -12.44** -10.359** Maximum likelihood R-square .831 .821 Log likelihood -350.368** -462.997** + p < .10 p < .05 ** p < .01

PAGE 85

85 Table 4-2. Zero-inflated negativ e binomial regression equations w ith coefficients (and Z-Scores) for gender-specific intimate partner homicide 2000. Female Model (N=178) Male Model (N=178) Exposure Reduction Percent divorced (gender-specific) .052 (1.18) .058 (1.93)+ Percent unmarried households .105 (.79) .112 (1.35) Male batterers program rate .077 (.61) .076 (.94) Shelter rate .48 (3.95)** .202 (2.69)** Legal service programs rate -.279 (-2.41)* -.244 (-3.32)** Referral services rate .023 (.04) -.565 (-1.55) Backlash Ratio of male to female education 2.042 (1.72)+ .537 (.70) Ratio of male to female income -1.156 (-1.66)+ .019 (.05) Ratio of male to female employment -3.081 (-1.97)* -.040 (-.04) Economic Deprivation Economic deprivation index (gender-specific) .013 (.72) .009 (.71) Controls Percent Hispanic (log) .001 (.01) -.045 (-.71) Residential mobility -.020 (-.98) -.010 (-.84) Officer rate -.193 (-1.86)+ .057 (.95) South .218 (.91) .085 (.59) Constant -8.61** -11.99** Maximum likelihood R-square .783 .8084 Log likelihood -245.044** -424.747** + p < .10 p < .05 ** p < .01

PAGE 86

86 Table 4-3. Summary of cross-sectional results, 1990 and 2000. 1990 2000 Female Model Male Model Female Model Male Model Exposure Reduction % divorced (+) % divorced (+) # shelters (+) # legal services (-) # shelters (+) # legal services (-) % divorced (+) # shelters (+) # legal services (-) Backlash NONE NONE M/F income (-) M/F education (+) M/F employ (-) NONE Economic Deprivation EcoDep Index (+) EcoDep Index (+) NONE NONE Control NONE South (+) Officer Rate (-) NONE

PAGE 87

87 Table 4-4. Fixed-effects regressi on coefficients for the relationshi p between changes in predictor variables and changes in logged female intimate partner hom icide counts for 143a US cities, 1990-2000 b t Exposure Reduction Female percent divorced -.036+ -1.74 Percent unmarried households .065 1.26 Male batterers program rate .055 .60 Shelter rate .261* 2.15 Legal service programs rate -.035 -.40 Referral services rate -.268 -.90 Backlash Ratio of male to female education .095 .13 Ratio of male to female income -.184 -.29 Ratio of male to female employment .299 .32 Economic Deprivation Female economic deprivation index .035* 2.37 Controls Percent Hispanic (log) -.260+ -1.70 Residential mobility -.023 -1.03 Officer rate -.111 -.99 1990b -.030 -.10 Female dummy log zeroc -.754** -5.51 Female population 15 and over (log)d -.534** -2.81 Constant 8.90** 2.65 Within R2 .537 Hausman test .000** Notes: (a) 70 observations (35 groups) were dropped due to all zero outcomes (i.e. zero count for intimate partner homicide for both 1990 and 2000); (b) coefficients for pe riod effects are relative to the reference category, 2000; (c) represents a dummy variable for imputation of zero for log of a zero count; (d) female population is multiplied by a factor of three to account for the dependent variable multiplied by a factor of three. + p < .10 p < .05 ** p < .01

PAGE 88

88 Table 4-5. Fixed-effects regressi on coefficients for the relationshi p between changes in predictor variables and changes in logged male intimate partner homicide counts for 174a US cities, 19902000 b t Exposure Reduction Male percent divorced -.009 -.28 Percent unmarried households .034 .66 Male batterers program rate .113 1.40 Shelter rate .035 .33 Legal service programs rate .097 1.16 Referral services rate -.025 -.09 Backlash Ratio of male to female education .389 .78 Ratio of male to female income .360 .69 Ratio of male to female employment -.834 -.99 Economic Deprivation Male economic deprivation index .008 .66 Controls Percent Hispanic (log) -.013 -.09 Residential mobility -.010 -.77 Officer Rate .016 .17 1990b .338 1.16 Male dummy log zeroc -1.01** -6.79 Male population 15 and over (log)d .296 .68 Constant -1.99 -.34 Within R2 .303 Hausman Test .008** Notes: (a) 8 observations (4 groups) were dropped due to all zero outcomes (i.e. zero count for intimate partner homicide for both 1990 and 2000); (b) coefficients for pe riod effects are relative to the reference category, 2000; (c) represents a dummy variable for imputation of zero for log of a zero count; (d) male population is multiplied by a factor of 3 to account for the dependent variable multiplied by a factor of 3. + p < .10 p < .05 ** p < .01

PAGE 89

89 Table 4-6. Summary of change model results, 1990 to 2000. Female Model Male Model Exposure Reduction % divorced (-) # shelters (+) NONE Backlash NONE NONE Economic Deprivation EcoDep Index (+) NONE Control % Hispanic (-) NONE

PAGE 90

90 CHAPTER 5 DISCUSSION AND CONCLUSIONS The goal of t his study was to examine the trends in both maleand female-perpetrated intimate partner homicides from 1990 to 2000 to obtain a better understanding of the differential declines in maleand female-perpe trated intimate partne r homicide. Clearly, the decline in intimate partner homicide over time is a very complex issu e. This research suggests that one explanation will not suffice. This research was based on the theories of e xposure reduction, backlash, and economic deprivation and marginalization. Results suggest a lack of supp ort for these theories. Cross-sectional analyses are presented to evaluate the ability of the exposur e reduction perspective, the backlash perspective, and ideas behind economi c deprivation and marginalization in explaining maleand female-perpetrat ed intimate partner homicide counts in large cities at two time periods (1990 and 2000). Pooled cross-sectional time se ries analyses are presented to evaluate the effectiveness of the exposure reduction perspective, the backlash perspectiv e, and economic deprivation and marginalization in explaining the ch anges in maleand female -perpetrated intimate partner homicide in large cities over time. Discussion Dramatic changes were occurring in domesticity the status of female s in terms of income, employment and educational atta inment, economic deprivation, and domestic violence resources from 1990 to 2000 in large cities. For instance, during this time period the percentage of the population divorced decreased for both males and females, while the percentage of unmarried households increased. Furthermore, females have made gains in educa tional attainment and employment relative to ma les (i.e., status of females increased ). However, this is not true for females median income. Females median income relative to males actuall y decreased during this time period (i.e., status of females decreased). Also, economic deprivation in terms of poverty,

PAGE 91

91 unemployment, and public assistan ce income decreased for both ma les and females. Moreover, the availability of shelters and lega l services for females increased, whereas the availability of male batterers counseling programs and referral services decreased. Any of thes e changes may have an impact on the gender-specific trends in intimate partner homicide that ha ve been witnessed over the last couple of decades. Exposure Reduction Some of the structural predictors are relate d to intimate partner homicide perpetration. According to the exposure reduction perspective, it was hypothesized that measures that reduced the contact between violent partners would be asso ciated with female inti mate partner homicide perpetration in both 1990 and 2000. For 1990, cr oss-sectional analyses show that the exposure reduction measure of female percent divorced is in f act associated with female-perpetrated intimate partner homicide; however the relationship is in the opposite direction of what was predicted. Instead of the percent divorced being associated with a decrease in fema le-perpetrated intimate partner homicide, as the exposure reduction perspe ctive would predict, the percent of females divorced is associated with an increase in female-perpetrated in timate partner homicide. Other research has found similar results (Dugan et al ., 1999; 2003). One possibl e explanation may have to do with the dynamics that co me with the ending of a viol ent relationship. Research has suggested that the most dangerous time in a violen t relationship is when th e relationship is coming to an end (Campbell, 1992; Goetting, 1995). So me females may be reacting to male partners violence over the end of the relationship and may resort to lethal means to protect themselves. None of the other exposure reductio n measures were related to fema le-perpetrated intimate partner homicide in 1990. In 2000, two of the domestic violence serv ice variables reach si gnificance for femaleperpetrated intimate partner homicide, however in opposite directions. The number of shelters per

PAGE 92

92 100,000 females is associated with an increase in female-perpetrated intimate partner homicide. One reason for this finding may be that although the shelter rate is accounting for the number of shelters available per 100,000 fema les in cities, it may not be capturing shelter usage by females in abusive relationships. It is possible that the availability of shelters across cities is increasing but females are not taking adva ntage of them to reduce the exposure between themselves and their abusive partner. On the opposite end of the sp ectrum, females may in fact be using shelters to reduce the exposure between themselves and th eir violent partners, but once shelter stays end may find themselves in contact with their violen t partners and may have to resort to lethal violence to protect themselves, if adequate protection is not available. Furthe rmore, research has shown that the most dangerous time in an abusiv e relationship is when a partner is trying to end the relationship (Goetting, 1995). Some females ma y be reacting to male partners violence over the threat of the end of the rela tionship and may resort to lethal means to protect themselves. The second domestic violence service variab le associated with female-perpetrated intimate partner homicide is the number of legal services per 100,000 female s. This variable is associated with a decrease in female-perpetrated intimate partner homicide. It seems that as females are provided with more opportunities fo r legal assistance, such as in obtaining restraining orders, court accompaniment, legal cl inics or advocacy the opportunities for females to kill their male partners decreases. Interestingly, three of the six exposure re duction measures are a ssociated with maleperpetrated intimate partner homicides in 1990 a nd 2000. Similar to female -perpetrated intimate partner homicide in 2000, the percent of males divo rced is positively relate d to male-perpetrated intimate partner homicide. Based on suggestions fr om prior research (Goetting, 1995), it seems

PAGE 93

93 plausible that the end of a relationship ma y cause some males to use more violence towards their female partners, resulting in more male-perpetrated intimate partner homicide. The shelter rate and the legal se rvice rate were also significant ly related to male-perpetrated intimate partner homicide, however in opposite directions. Instead of reducing intimate partner homicide due to a decrease in opportunity, as the exposure reduction perspective would predict, the number of shelters per 100,000 females increases male intimate partner homicide perpetration. Although surprising, this finding is plausible in light of the rese arch that has found that maleperpetrated homicide is likely to occur when a partner is trying to leave an abusive relationship (Goetting, 1995). Some males may f eel threatened that their partne rs are leaving and staying at a shelter and may react in a lethal manner. More over, adequate protecti on may not be provided when females are ending th eir stay at a shelter. In addition, the number of legal services ava ilable to females is re lated to a decrease in male intimate partner homicide perpetration. It seems reasonable that, as females are provided with more opportunities for legal assistance, the opportunities fo r males to kill their female partners would decrease. Dugan et al. (2003) found similar results. That is they found that legal advocacy, measured as an index of the number of agencies with a separate budget for legal advocacy and the number of agencies that have lawy ers on staff, was related to fewer killings of white married females by their partners. Backlash According to the backlash perspective, it was hypothesized that measur es that may threaten male control and dominance would be associated with male intima te partner homicide perpetration in 1990 and 2000. Interestingly, support was not found for this hypothesis. Although surprising, backlash explanations may be less relevant in explaining male-perpetrated intimate partner homicide, because gender roles have changed over the years. Today, it is more acceptable for

PAGE 94

94 women to have jobs outside of the home and pursue higher education. Due to this transformation in gender roles, females economic and social stat us has increased. Furthermore, Felson (2006) surveyed 10,000 men and women about their spouse s behavior and concluded that males are no more controlling than females. Contrary to what was expected backlash correlates are associ ated with female-perpetrated intimate partner homicide in 2000. It appear s that, as the gap betw een male a nd female employment and income widens (i.e., females status relativ e to males decreases), femaleperpetrated intimate partner homicide decreases. Ho wever, as the gap between males and females educational attainment widens, female-perpetr ated intimate partner homicide increases. Females may react to disparities in income, employment and educational attainment just as it is predicted that males do, just in terms of frustration and not in terms of maintain ing control and dominance in the relationship. Along these lines, females appear to be taking their frustrations about the widening gap in educational attain ment out on their male partners, but not their frustrations about the widening gap in income or employment. Ande rson (1997) suggests that it is possible that females with lower status relative to males are more likely to engage in vi olence because they are less likely to be able to leave a violent relationship. That being said, females with a lower educational attainment status relative to male s may be less likely to be able to remove themselves from a violent situation, possibly be cause education affords females opportunities to learn about services available to reduce contact to prevent vi olence. Anderson (1997) found that women with slightly less education than their male partners have almost 3 times greater odds of perpetrating domestic violence th an women partnered with men processing the same amount of education (664).

PAGE 95

95Economic Deprivation and Marginalization According to the ideas behind economic deprivation and marginalization, it was hypothesized that measures of pove rty, unemployment and the percen t of the population on public assistance would be associated with both maleand female-perpetrated intimate partner homicides. Support for this hypothesis was found in 1990. Poverty, unemployment, and having to rely on public assistance income may significantly impact the dynamics within the home, either in terms of frustration or strain. Males and females may r eact to this strain by lashing out in a lethal manner at their intimate partners. Summary The results of this analysis offer support fo r economic deprivation an d marginalization in explaining both maleand female-perpetrated in timate partner homicide in 1990. Results also suggest an association between some of the variables representing the exposure reduction perspective and maleand female-perpetrated inti mate partner homicide in both 1990 and 2000. In general, the three exposure reduction indicators that are consistently signifi cant in 1990 and 2000 are gender-specific percent divorced and two indi cators measuring the av ailability of domestic violence services the shelter rate and the legal service rate. Importan tly, these measures are related to both maleand female-p erpetrated intimate partner homicides in a similar fashion. A one standard deviation increase in any of these measur es results in a significan t percentage change in intimate partner homicide counts. The availability of legal services in c ities supports the exposure reduction hypotheses. That is, an increase in the number of legal services per 100,000 females is related to an increase in male-p erpetrated intimate pa rtner homicide in 19 90 and 2000 and femaleperpetrated intimate partner homicide in 2000. However, not all of the significant exposure reduction measures influence intimate partner homicide in the predicted directi on. For instance, an increase in th e percent divorced increases both

PAGE 96

96 male and female intimate partner homicide. Furthermore, results suggest that an increase in the number of shelters per 100,000 females is related to an incr ease in male-perpetrated intimate partner homicide in 1990 and 2000 and female-perpetrated intimate partner homicide in 2000. This is a main domestic violence servic e measure that has historically been used to measure domestic violence services and was expect ed to have an exposure reduction effect on intimate partner homicide perpetration. Instead, it appears that th is measure is associated with more males and females reacting out in a lethal manner towards their intimate part ner, whether it is for protection or to maintain control and domina nce. Dugan et al. (2003) also f ound a backlash effect for one of their measures of domestic viol ence services. Specifi cally, they found that prosecutors willingness to prosecute violators of protection orders was related to an incr ease in homicides of married and unmarried white females and unmarried African-A merican males. They concluded that, although prosecutors willingness to prosecute is advantageous, if this willi ngness comes with out also being able to provide adequate protection for victims, homicide may result. As st ated earlier, if these services arent being utilized by females, they will be less likely to ha ve an exposure reduction effect on intimate partner homicide Also, given the fact that the likelihood of intimate partner homicide is extremely high when the relationship is ending, domestic viol ence resources need to provide adequate protection when th e relationship is ending or all th e good intentions of increasing shelter availability will not prevent intimate partner violence and homicide. Indicators measuring the backlash perspectiv e do not predict male-p erpetrated intimate partner homicide in 1990 or 2000 as was hypothesized. Interestingly, backlash indicators do reach significance in the 2000 female-per petrated intimate partner homicide model. However, the relationship between these variables and female-perpe trated intimate partner homicide is complex.

PAGE 97

97 A one standard deviation increase in any of thes e three ratio measures re sults in a significant percentage change in the female-perpetrated inti mate partner homicide in 2000. An increase in the ratio of median income and the ratio of male to female empl oyment has a negative effect on female-perpetrated intimate homici de (i.e., decreases), while an in crease in the ratio of male to female education has a positive ef fect (i.e., increases) on female-perpetrated intimate partner homicide. Changes Over Time In regards to changes over time, changes in structural variables fr om 1990 to 2000 were related to intimate partner homicide perpetration ov er time, however this is only true for femaleperpetrated intimate partner homicide. Interestingl y, although many of the structural predictors influence male-perpetrated intima te partner homicide cross-sectionally, these perspectives did not explain male-perpetrated intimate partner homici de over time at all. Results suggest that c ities that are characterized by changes in the percent of females divorced from 1990 to 2000 are characterized by changes in female-perpetrated intimate partner homicide counts. Also, cities that show changes in the numb er of shelters per 100,000 fema les from 1990 to 2000 also show changes in female-perpetrated intimate partner homicide counts. Furthe rmore, cities that are characterized by changes in the pe rcent of females in poverty, the percent of females that are unemployed, and the percent of households on public assistance from 1990 to 2000 are also characterized by changes in female-perpetrated intimate partner homicide counts. Lastly, cities that are characterized by cha nges in the logged percent Hisp anic population from 1990 to 2000 are characterized by changes in female-perpe trated intimate partner homicide counts. Overall, there is mixed support for the changes in the theoretical perspectives on the changes in female-perpetrated intimate pa rtner homicide counts from 1990 to 2000 and absolutely no support on the changes in male-p erpetrated intimate partner homicide counts.

PAGE 98

98 Changes in economic deprivation within cities are significantly associated with changes in female-perpetrated intimate partner homicide counts. Also, changes in two indicators of exposure reduction were related to changes in female-perpetrated intimate partner homicide counts; however the shelter rate was in the opposite direction of wh at was predicted and increased shelter availab ility appears to be associated with more harm than good if shelters are not being taken advantage of or if she lters are not providing adequate protection. Limitations There are some limitations to this research. First and foremost, the data utilized for the dependent variables was based on Supplemental Homici de Files. The deficiencies of the database have been well-documented (Junger-Tas and Ma rshall, 1999). These include underreporting, errors in the assignment of relationships, failure to identify relationship in a substantial portion of cases, and other missing information SHFs include information on homicides that were submitted by law enforcement agencies to the FBI. Submission of homicide information is completely voluntary and therefore not 100% accurate. In ad dition, many homicides are coded as having an unknown victim-offender relationshi p. A weighting proce dure was utilized to acco unt for this lack of reporting and lack of relationship information. Moreover, it is acknowledged that actual usag e of the domestic vi olence services was unable to be captured in cities. Ju st because shelters, legal servi ces, batterers counseling programs, and referrals are offered does not mean they are being utilized to the fullest by individuals in violent relationships. Also, the size of the domes tic violence service programs (i.e., service capacity) was unable to be captured. If domestic violence services are not being utilized or if programs have limited capacity, domestic violence services would have a li mited effect on intimate partner homicide.

PAGE 99

99 Also, it was difficult to divide the structural factors of inte rest into the three theoretical categories. It is possible that ma ny of the structural indicators could have fallen into another theoretical category. For instance, although for this study females status relative to males was considered to represent the backlash hypothesis, other research has considered females status relative to males as measuring exposure reduction, because an increase in the status of females affords them more opportunities to leave abusive relationships. In additi on, research has used females status relative to males to represent economic deprivation and marginalization. For the current study, all structural indicators were di vided into theoretical categories based on the developed hypotheses. Furthermore, data was limited to two time points (1990 and 2000) Although the 1990s witnessed drastic changes in inti mate partner homicide perpetration, changes in intimate partner homicide have been witnessed sin ce 1976. It would be interesting to expand the data to include earlier data. Also, data was limite d to large cities with a populati on over 100,000. It would be useful to examine data in a ll sizes of cities over time. Future Research Future research on intimate partner homicide could take many directions. First, it would be useful to examine intimate partner homicide as well as intimate partner aggravated assault, in addition to the access or availability of medical centers. Aggrav ated assault and homicide are similar events, with the exception of the outcome (i.e., one ends in death, whereas the other does not). Harris et al. (2002) argues that research would benefit by focusing on serious assaults, with only a small percentage that actually end in death. They show that the decl ine in the homicide rate parallels the increase in medical technology and support services. It is seems reasonable that the availability of medical care (i.e., trauma cent ers) may influence whether an intimate partner aggravated assault remained an aggravated assaul t or resulted in death and was categorized as an

PAGE 100

100 intimate partner homicide. Many intimate partner homicides may ha ve been aggravated assaults that became homicides when ade quate medical care was not available. Greater availability of medical resources may prevent many intimate part ner homicides and thus is an important factor and should be considered when examining intimate partner homicide over time. Also, it would be important to examine the cities that were excluded from the current analysis due to lack of reporting. The majority of the cities exclude d were located in the state of Florida. Research has shown that homicide rates are higher in Sout hern states (DeWees and Parker, 2003; Smith and Brewer, 1992). If intimate partner ho micide data could be gathered, it would be interesting to examine intimate pa rtner homicide in large cities in Florida to see if results were similar with the results fr om the current study. Furthermore, it would be important to include ot her factors in the analysis especially the age structure of the cities. Browne et al. (1999) examined age-specific ra tes of intimate partner homicide from 1980 to 1 995 and found that different trends emerge for different age groups and for males and females. They found that a decline in female-perpetr ated intimate partner homicide occurred in all age groups, except the age group 13 to 17 years old (rates ha ve remained low). The greatest decreases occurred in the 25 to 34 an d the 35 to 44 year ol d age groups. For maleperpetrated intimate partner homicide, increases were found for the 13 to 17 and the 18 to 24 year old age groups, while decreases were seen in the 25 to 34 and the 35 to 44 year old age groups. Since differential trends in intima te partner homicide have been witnessed for various age groups and younger age groups are more likely to participate in inti mate partner homicide, it would be important to control for ag e in future analyses. Conclusion This research offers many cont ributions to the literature on in timate partner homicide. Most importantly, this research examin es the dynamic nature of a nu mber of different structural

PAGE 101

101 variables on both maleand female-perpetrated intimate partner homici de over time. This is something that has not been done in the literature to date. The tim e period covered includes a time in history where dramatic change s were occurring in domesticity, the status of fe males, economic deprivation and marginalization, and the availab ility of domestic violence resources with the enactment of the Violence of Against Women Act in 1994. A main finding of this resear ch speaks to the ineffectiveness of domestic violence services. The number of shelters per 100,000 females is significantly related to intimate partner homicide in 1990 and 2000 and over time. Despit e all the efforts to increase she lter availability to females in violent relationships, it appears that the increase in availability is actually associated with an increase in intimate part ner homicide. For instance, in 1990 a nd 2000 the increase in the shelter rate was related to an increase in male-perpetrated intimate partner homicide. Furthermore, not only was this finding seen in 200 0 for female-perpetrated intimate partner homicide, this finding was witnessed over time as well. Efforts to prevent domestic viol ence and homicide need to make sure that they also provide adequate protection during times that are characterized by increased violence. This research adds to the knowledge of th e decline in female-perpetrated intimate homicide, but still leaves explanations behind the decline in male-perpetr ated unclear. Research needs to continue to examine ge nder-specific intimate partner ho micide, but particularly maleperpetrated intimate partner homicide. Surprisi ngly, none of the three theoretical perspectives were associated with male-perpetrated intimate partner homicide over ti me, whereas some of the predictors were associated with female-perpetrat ed intimate partner homicide over time. It is extremely important to gain a better understanding of predictors of male-perpetrated intimate partner homicide, given the fact that male-perpetrated intimate partner homicide has not decreased

PAGE 102

102 as sharply as female-perpetrated intimate part ner homicide over time de spite the increase in domestic violence resources geared towards preventi ng violence against females. It is essential to determine what really has influe nced male-perpetrated intimate partner homicide over time. This may be the only way to for us to see a larger decline in these types of homicides.

PAGE 103

103 APPENDIX A GLOSS ARY Place Places, for the reporting of decennial cen sus data, include census designated places, consolidated cities, and incorporated places. South Region South Atlantic Division: Delaware, Maryland, District of Columbia, Vi rginia, West Virginia, North Carolina, South Carolina, Georgia, Florida East South Central Division: Kentucky, Tennessee, Alabama, Mississippi West South Central Division: Arkansas, Louisiana, Oklahoma, Texas Employed All civilians 16 years old a nd over who were either (1) at work those who did any work at all during the reference week as pa id employees, worked in their own business or profession, worked on their own farm, or wo rked 15 hours or more as unpaid workers on a family farm or in a family business; or (2) we re with a job but not at work those who did not work during the reference w eek, but who had jobs or busin esses from which they were temporarily absent because of illnes s, bad weather, industrial dispute, vacation, or other personal reasons. Excluded from the employed are people whose only activity consisted of work around their own house (painting, repa iring, or own home housework) or unpaid volunteer work for religious, charitable, and simila r organizations. Also excluded are all institutionalized people and people on active duty in the United States Armed Forces. Unemployed All civilians 16 years old and over were cl assified as unemployed if they were neither at work nor with a job but not at work during the reference week, were looking for

PAGE 104

104 work during the last 4 weeks, and were availabl e to start a job. Also included as unemployed were civilians 16 years old and ove r who: did not work at all dur ing the reference week, were on temporary layoff from a job, had been informed th at they would be recalled to work within the next 6 months or had been given a date to return to work, and were available to return to work during the reference week, except for temporary illness. Civilian labor force Consists of people classified as employed or unemployed in accordance with the criteria described above. Labor force All people classified in the civili an labor force (i.e., employed and unemployed people), plus members of the U.S. Armed Forces (people on active duty with the United States Army, Air Force, Navy, Marine Corps, or Coast Guard). Not in labor force All people 16 years old and over who are not classified as members of the labor force. This category consists mainly of st udents, individuals taking care of home or family, retired workers, seasonal workers enumerated in an off-season who were not looking for work, institutionalized people (all institutionalized people are placed in this cat egory regardless of any work activities they may have done in the refe rence week), and people doing only incidental unpaid family work (fewer than 15 hours during the reference week). Household A household includes all of the people who occupy a housing unit. (People not living in households are classifi ed as living in group quarters. ) A housing unit is a house, an apartment, a mobile home, a group of rooms, or a single room occupied (or if vacant, intended for occupancy) as separate living quarters Householder The data on relationship to householder were derived from the question, How is this person related to Person 1, which was aske d of Persons 2 and higher in housing units. One person in each household is designated as the householder (Person 1). In most cases, the

PAGE 105

105 householder is the person, or one of the people, in whose name the home is owned, being bought, or rented. If there is no such person in the household, any adult household member 15 years old and over could be designated as the householder (i.e., Person 1) Households are classified by type according to the sex of the householder a nd the presence of rela tives. Two types of householders are distinguished: family house holders and nonfamily householders. A family householder is a householder living with one or more individuals re lated to him or her by birth, marriage, or adoption. The householder and all of the people in the house hold related to him or her are family members. A nonfamily househol der is a householder liv ing alone or with nonrelatives only. Unmarried partner An unmarried partner is a person who is not related to the householder, who shares living quarters, and who has a clos e personal relationship with the householder. Unmarried-Partner Household An unmarried-partner household is a household that includes a householder and an unmarried pa rtner. An unmarried partner can be of the same or of the opposite sex of the householder. An unma rried partner in an unmarried-partner household is an adult who is unrelated to the householder, but shares living quarters and has a close personal relationship with the householder. An unmarried-par tner household may also be a family household or a nonfamily household, depending on the presence or absence of another person in the household who is related to the householder. Th ere may be only one unmarriedpartner per household, and an unmarried part ner may not be included in a married-couple household as the householder cannot have bot h a spouse and an unmarried partner. Income of individuals Income for individuals is obtaine d by summing the eight types of income for each person 15 years old and over. Th e characteristics of individuals are based on the time of enumeration (April 1, 2000), even t hough the amounts are for calendar year 1999.

PAGE 106

106Median income The median divides the income distribut ion into two equal parts: one-half of the cases falling below the median income and one-half above the median. For households and families, the median income is based on the di stribution of the total number of households and families including those with no income. The me dian income for individuals is based on individuals 15 years old and over with income. Median income for households, families, and individuals is computed on the ba sis of a standard distribution (s ee the Standard Distributions section under Derived Measures ). Median income is rounded to the nearest whole dollar. Median income figures are calculated using linear interpolation if the width of the interval containing the estimate is $2,500 or less. If the wi dth of the interval containing the estimate is greater than $2,500, Pareto in terpolation is used. (For more information on medians and interpolation, see Derived Measures.) Divorced -This category includes people who are lega lly divorced and who have not remarried. How Poverty Status is Determined The poverty status of families and unrelated individuals in 1999 was determined using 48 thresholds (income cutoffs) arranged in a two dimensional matrix. The matrix consists of family size (from 1 person to 9 or more people) crossclassified by presence and number of family members under 18 years old (from no children presen t to 8 or more childre n present). Unrelated individuals and 2-person families were further differentiated by the age of the reference person (RP) (under 65 years old and 65 years old and over). To determine a persons poverty status, one compares the persons total family income with the poverty threshold appropriate for that persons family size and composition (see table below). If the total income of that persons family is less than the threshold appropriate for that family, then the person is considered poor, togeth er with every member of his or her family. If a person is not

PAGE 107

107 living with anyone related by birth, marriage, or adoption, then the persons own income is compared with his or her poverty threshold. Individuals for whom povert y status is determined Poverty status was determined for all people except institutionalized people, people in military group quarters, people in college dormitories, and unrelated individuals under 15 years old. These gr oups also were excluded from the numerator and denominator when calculating poverty rates. They are considered neither poor nor nonpoor. (Census Glossary) Residence 5 years ago The data on residence 5 years earlie r were derived from answers to long-form questionnaire Item 15, which was aske d of a sample of the population 5 years old and over. This question asked for the state (or foreig n country), U.S. county, city or town, and ZIP Code of residence on April 1, 1995, for those people who reported that on th at date they lived in a different house than their current residence. Residence 5 years earlier is used in conjunction with location of current residence to determin e the extent of residential mobility of the population and the resulting redistribution of the population across the various states, metropolitan areas, and regions of the country. Summary File 1 This Census 2000 file presents 100percent population and housing data for the total population, for 63 race categories, a nd for many other race and Hispanic or Latino categories. The data include age, sex, households, household relations hip, housing units, and tenure (whether the residence is ow ned or rented). Also included are selected characteristics for a limited number of race and Hispanic or Latino categories. The data are available for the U.S., census regions, census divisions, states and st atistically equivalent entities, counties and statistically equivalent entities, county subdi visions, places, census trac ts, block groups, census blocks, metropolitan areas, urban areas, Americ an Indian and Alaska Native areas, tribal

PAGE 108

108 subdivisions, Hawaiian home lands, Congressional district s, and ZIP Code ta bulation areas. Data are available down to the block level for many ta bulations, but only to th e census tract level for others. Available on CD-ROM, DVD, and American FactFinder. Summary File 3 This Census 2000 file presents da ta on population and housing long-form subjects, such as income and education. It incl udes population totals for an cestry groups. It also includes selected characteristics for a limited number of race and Hispanic or Latino categories. The data are available for the U.S., census regi ons, census divisions, st ates and statistically equivalent entities, counties and statistically equivalent entities, county subdivisions, places, census tracts, block groups, metropolitan areas, urban areas, American Indian and Alaska Native areas, tribal subdivisions, Hawa iian home lands, Congressional districts, and ZIP Code tabulation areas. Available on CD-ROM DVD, and American FactFinder. UCR Uniform Crime Reports

PAGE 109

109 APPENDIX B CORRELATION MATRIX FOR EXPLA NATORY AND OUTCO ME VARIABLE 1990.

PAGE 110

110 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1. Female-perp IPH 1.00 2. % Fem poverty .359 1.00 3. % Fem unempl .300 .782 1.00 4. % HH pub assis. .244 .786 .793 1.00 5. M/F income -.221 -.282 -.093 -.265 1.00 6. M/F education -.190 -.393 -.328 -.364 .429 1.00 7. M/F % employ -.003 .181 .249 .212 .220 .208 1.00 8. % fem div .005 -.106 -.124 -.041 -.076 .050 -.126 1.00 9. % unmarried .003 -.020 .053 .124 -.236 -.084 -.105 -.122 1.00 10. Shelter rate -.281 .024 -.107 -.026 .118 -.039 -.085 .036 -.129 1.00 11. Legal serv rate -.223 .093 .011 .026 .078 -.017 -.068 .079 -.086 .678 1.00 12. Male batt. rate -.078 .052 -.048 .016 -.077 -.178 -.071 .127 -.118 .293 .316 1.00 13. Referral rate -.115 -.048 .084 .034 .034 -.046 -.072 .025 .017 .164 -.009 .133 1.00 14. Res. mobility -.244 -.460 -.431 -.442 .046 .111 -.226 .213 .225 .028 .016 .009 .079 1.00 15. Hispanic(log) -.058 -.125 .110 .072 -.159 .032 .152 .109 .198 -.126 -.105 -.045 .077 .279 1.00 16. Officer rate .347 .467 .305 .388 -.403 -.345 -.146 -.132 -.004 -.014 -.044 .039 -.049 -.449 -.167 1.00 17. South .183 .189 .045 -.195 .014 -.083 -.041 -.189 -.234 -.008 .010 .074 -.067 -.025 -.262 .105 1.00 1. Male-perp IPH 1.00 2. % Male poverty .260 1.00 3. % Male unempl .199 .786 1.00 4. % HH pub assis. .195 .766 .863 1.00 5. M/F income -.250 -.322 -.193 -.265 1.00 6. M/F education -.177 -.405 -.335 -.364 .429 1.00 7. M/F % employ .006 .192 .218 .212 .220 .208 1.00 8. % Male div .050 .001 .024 .029 -.094 -.009 -.152 1.00 9. % Unmarried .004 -.021 .063 .124 -.236 -.084 -.105 -.107 1.00 10. Shelter rate -.280 .011 -.053 -.026 .118 -.039 -.085 .054 -.129 1.00 11. Legal Serv rate -.245 .091 .050 .026 .078 -.017 -.068 .079 -.086 .678 1.00 12. Male Batt. rate -.039 .062 -.018 .016 -.077 -.178 -.071 .137 -.118 .293 .316 1.00 13. Referral rate -.121 -.033 .063 .034 .034 -.046 -.072 .005 .017 .164 -.009 .113 1.00 14. Res. mobility -.227 -.389 -.537 -.442 .046 .111 -.226 .138 .225 .028 .016 .009 .079 1.00 15. Hispanic(log) .018 -.041 .075 .072 -.159 .032 .152 .014 .198 -.126 -.105 -.045 .077 .279 1.00 16. Officer rate .323 .426 .444 .388 -.403 -.345 -.146 -.048 -.004 -.014 -.044 .039 -.049 -.449 -.167 1.00 17. South .122 .151 -.160 -.195 .014 -.083 -.041 -.188 -.234 -.008 .010 .074 -.067 -.025 -.262 .105 1.00

PAGE 111

111 APPENDIX C CORRELATION MATRIX FOR EXPLANAT ORY AND OUTCOME VARIABLES 2000.

PAGE 112

112 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1. Female-perp IPH 1.00 2. % Fem poverty .257 1.00 3. % Fem unempl .219 .816 1.00 4. % HH pub assis. .147 .751 .751 1.00 5. M/F income -.245 -.256 -.200 -.178 1.00 6. M/F education -.148 -.365 -.365 -.326 .476 1.00 7. M/F % employ -.124 -.201 -.182 -.122 .396 .535 1.00 8. % fem div .032 -.061 -.146 -.145 -.009 -.038 -.245 1.00 9. % unmarried .067 .339 .382 .537 -.294 -.358 -.267 .209 1.00 10. Shelter rate -.224 .089 .095 .115 -.012 -.079 -.182 .010 .150 1.00 11. Legal serv rate -.197 .030 .033 .072 -.092 -.058 -.168 -.008 .124 .709 1.00 12. Male batt. rate -.100 .018 -.004 -.015 -.021 -.123 -.055 -.021 -.014 .347 .431 1.00 13. Referral rate -.137 .047 .065 -.002 .109 -.035 .065 -.034 -.060 .047 -.253 .086 1.00 14. Res. mobility -.084 -.164 -.155 -.251 -.289 .004 -.122 .162 .117 .008 .058 .019 -.013 1.00 15. Hispanic(log) -.058 -.061 .085 .177 -.053 .061 .477 -.345 .133 -.182 -.191 -.008 .108 .153 1.00 16. Officer rate .311 .482 .425 .353 -.424 -.224 -.372 -.011 .152 .004 .072 .000 -.103 -.210 -.282 1.00 17. South .096 .099 .001 -.293 -.019 -.023 -.078 .070 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00 1. Male-perp IPH 1.00 2. % Male poverty .210 1.00 3. % Male unempl .126 .788 1.00 4. % HH pub assis. .151 .757 .816 1.00 5. M/F income -.226 -.323 -.218 -.178 1.00 6. M/F education -.105 -.370 -.405 -.326 .476 1.00 7. M/F % employ -.007 -.124 -.233 -.122 .396 .535 1.00 8. % Male div -.013 .000 .131 -.009 -.034 -.123 -.348 1.00 9. % Unmarried .048 .371 .484 .537 -.294 -.358 -.267 .302 1.00 10. Shelter rate -.327 .054 .130 .115 -.012 -.079 -.182 .136 .150 1.00 11. Legal serv rate -.241 .012 .062 .072 -.092 -.058 -.168 .085 .124 .709 1.00 12. Male batt. rate -.105 .003 -.010 -.015 -.021 -.123 -.055 .022 -.014 .347 .431 1.00 13. Referral rate -.143 .040 .066 -.002 .109 -.035 .065 .039 -.060 .047 -.253 .086 1.00 14. Res. mobility -.099 -.081 -.195 -.251 -.289 .004 -.122 .065 .117 .008 .058 .019 -.013 1.00 15. Hispanic(log) .071 .080 .012 .177 -.053 .061 .477 -.434 .133 -.182 -.191 -.008 .108 .153 1.00 16. Officer rate .338 .411 .435 .353 -.424 -.224 -.372 .125 .152 .004 .072 .000 -.103 -.210 -.282 1.00 17. South .005 .026 -.154 -.293 -.019 -.023 -.078 .040 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00

PAGE 113

113 APPENDIX D SIMPLIFIED CORRELATION MATRIX FOR EXPLANATORY AND OUT COME VARIABLES 1990.

PAGE 114

114 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Female-perp IPH 1.00 2. Econ. dep. index .359 1.00 3. M/F income -.221 -.241 1.00 4. M/F education -.190 -.393 .429 1.00 5. M/F % employment -.003 .209 .220 .208 1.00 6. % Fem divorced .005 -.116 -.076 .050 -.126 1.00 7. % Unmarried HH .003 .000 -.236 -.084 -.105 -.122 1.00 8. Shelter rate -.281 -.012 .118 -.039 -.085 .036 -.129 1.00 9. Legal service rate -.223 .074 .078 -.017 -.068 .079 -.086 .678 1.00 10. Male batt. rate -.078 .026 -.077 -. 178 -.071 .127 -.118 .293 .316 1.00 11. Referral rate -.115 -.012 .034 -.046 -.072 .025 .017 .164 -.009 .133 1.00 12. Res. mobility -.244 -.473 .046 .111 -.226 .213 .225 .028 .016 .009 .079 1.00 13. % Hispanic (log) -.058 -.063 -.159 .032 .152 .109 .198 -.126 -.105 -.045 .077 .279 1.00 14. Office rate .347 .442 -.403 -.345 -.146 -.132 -.004 -.014 -.044 .039 -.049 -.449 -.167 1.00 15. South .183 .156 .014 -.083 -.041 -.189 -.234 -.008 .010 .074 -.067 -.025 -.262 .105 1.00 1. Male-perp IPH 1.00 2. Econ. dep. index .254 1.00 3. M/F income -.250 -.298 1.00 4. M/F education -.177 -.403 .429 1.00 5. M/F % employment .006 .210 .220 .208 1.00 6. % Male divorced .050 .016 -.094 -.009 -.152 1.00 7. % Unmarried HH .004 .004 -.236 -.084 -.105 -.107 1.00 8. Shelter rate -.280 -.008 .118 -.039 -.085 .054 -.129 1.00 9. Legal Service rate -.245 .083 .078 -.017 -.068 .079 -.086 .678 1.00 10. Male batt. rate -.039 .040 -.077 -. 178 -.071 .137 -.118 .293 .316 1.00 11. Referral rate -.121 -.005 .034 -.046 -.072 .005 .017 .164 -.009 .133 1.00 12. Res. mobility -.227 -.454 .046 .111 -.226 .138 .225 .028 .016 .009 .079 1.00 13. % Hispanic (log) .018 -.007 -.159 .032 .152 .014 .198 .198 -.105 -.045 .077 .279 1.00 14. Office rate .323 .452 -.403 -.345 -.146 -.048 -.004 -.004 -.044 .039 -.049 -.449 -.167 1.00 15. South .122 .061 .014 -.083 -.041 -.188 -.234 -.34 .010 .074 -.067 -.025 -.262 .105 1.00

PAGE 115

115 APPENDIX E SIMPLIFIED CORRELATION MATRIX FOR EXPLANATORY AND OUT COME VARIABLES 2000.

PAGE 116

116 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Female-Perp IPH 1.00 2. Econ. Dep. Index .229 1.00 3. M/F Income -.245 -.237 1.00 4. M/F Education -.148 -.380 .476 1.00 5. M/F % Employment -.124 -.184 .396 .535 1.00 6. % Fem Divorced .032 -.114 -.009 -.038 -.245 1.00 7. % Unmarried HH .067 .449 -.294 -.358 -.267 .209 1.00 8. Shelter Rate -.224 .107 -.012 -.079 -.182 .010 .150 1.00 9. Legal Service Rate -.197 .049 -.092 -.058 -.168 -.008 .124 .709 1.00 10. Male Batt. Rate -.100 .003 -.021 -.123 -.055 -.021 -.014 .347 .431 1.00 11. Referral Rate -.137 .036 .109 -.035 .065 -.034 -.060 .047 -.253 .086 1.00 12. Res. Mobility -.084 -.208 -.289 .004 -.122 .162 .117 .008 .058 .019 -.013 1.00 13. % Hispanic (log) -.058 .052 -.053 .061 .477 -.345 .133 -.182 -.191 -.008 .108 .153 1.00 14. Office Rate .311 .462 -.424 -.224 -.372 -.011 .152 .004 .072 .000 -.103 -.210 -.282 1.00 15. South .096 -.059 -.019 -.023 -.078 .070 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00 1. Male-Perp IPH 1.00 2. Econ. Dep. Index .180 1.00 3. M/F Income -.226 -.260 1.00 4. M/F Education -.105 -.390 .476 1.00 5. M/F % Employment -.007 -.161 .396 .535 1.00 6. % Male Divorced -.013 .030 -.034 -.123 -.348 1.00 7. % Unmarried HH .048 .500 -.294 -.358 -.267 .302 1.00 8. Shelter Rate -.327 .104 -.012 -.079 -.182 .136 .150 1.00 9. Legal Service Rate -.241 .051 -.092 -.058 -.168 .085 .124 .709 1.00 10. Male Batt. Rate -.105 -.008 -.021 -.123 -.055 .022 -.014 .347 .431 1.00 11. Referral Rate -.143 .032 .109 -.035 065 .039 -.060 .047 -.253 .086 1.00 12. Res. Mobility -.099 -.189 -.289 .004 -.122 .065 .117 .008 .058 .019 -.013 1.00 13. % Hispanic (log) .071 .110 -.053 .061 .477 -.434 .133 -.182 -.191 -.008 .108 .153 1.00 14. Office Rate .338 .426 -.424 -.224 -.372 .125 .152 .004 .072 .000 -.103 -.210 -.282 1.00 15. South .005 -.154 -.019 -.023 -.078 .040 -.523 -.115 -.121 .112 .053 .118 -.213 .107 1.00

PAGE 117

117 APPENDIX F VARIANC E INFLATION FACTORS FOR SIMPLIFIED MODELS Fem-Perp90 Male-Perp90 Fem-Perp00 Male-Perp00 Female economic deprivation index 1.900 2.028 Male economic deprivation index 1.987 2.006 Ratio of male to female education 1.549 1.546 2.068 2.084 Ratio of male to female income 1.615 1.614 2.039 2.034 Ratio of male to female employment 1.438 1.457 2.369 2.381 Percent female divorced 1.226 1.450 Percent male divorced 1.197 1.623 Percent unmarried households 1.343 1.314 2.799 2.908 Male batterers programs per 100,000 males 1.226 1.227 1.397 1.397 Shelters per 100,000 females 2.056 2.054 2.391 2.385 Legal service programs per 100,000 females 2.056 2.061 2.864 2.855 Referral services per 100,000 females 1.108 1.111 1.290 1.308 Percent hispanic (log) 1.348 1.361 2.078 2.176 Residential mobility 1.908 1.890 1.634 1.616 Officer rate per 1,000 1.780 1.789 2.076 2.026 South 1.237 1.242 1.959 1.891

PAGE 118

118 APPENDIX G CORRELATION MATRICES FOR DOMESTIC VIOLE NCE SERVICE VARIABLES. 1990 1 2 3 4 5 6 7 1. Shelters 1.00 2. Hotlines .906 1.00 3. Counseling .906 .995 1.00 4. Children counseling .877 .920 .922 1.00 5. Legal services .646 .723 .736 .661 1.00 6. Batterers counseling .392 .419 .420 .422 .283 1.00 7. Number of referrals .328 .413 .418 .301 .101 -.241 1.00 2000 1 2 3 4 5 6 7 1. Shelters 1.00 2. Hotlines .904 1.00 3. Counseling .900 .981 1.00 4. Children counseling .923 .946 .956 1.00 5. Legal services .851 .921 .959 .933 1.00 6. Batterers counseling .630 .702 .734 .715 .733 1.00 7. Number of referrals -.163 -.144 -.153 -.191 -.246 -.095 1.00

PAGE 119

119 REFERENCES Agresti, Allen. 1996. An Introduction to Categorical Data Analysis. New York: John Wiley and Sons. Allard, Mary Ann, Randy Albelda, Mary Ellen Colten, and Carol Cosenza. 1997. In Harms Way? Domestic Violence, AFDC Receipt, and Welfare Reform in Massachusetts. Boston, MA: University of Massachusetts. Allison, Paul D. 1994. Using panel data to estimate the effects of events. Sociological Methods and Research 23: 179-199. Anderson, Kristin L. 1997. Gender, status, and dome stic violence: An integration of feminist and family violence approaches. Journal of Marriage and the Family 59:655-669. Archer, John. 2000. Sex differences in aggres sion between heterosexual partners: A Metaanalytic review. Psychological Bulletin 126:651-680. Avakeme, Edem F. 1999. Females labor force part icipation and intimate femicide: An empirical assessment of the backlash hypothesis. Violence and Victims 14:277-291. Bailey, William C. 1984. Poverty, inequality, and city homicide rates: Some not so unexpected results. Criminology 22:531-550. Bailey, William C., and Ruth D. Peterson. 1995. Ge nder inequality and violence against women: The case of murder. In Crime and Inequality, eds. John Hagan and Ruth D. Peterson. Stanford, CA: Stanford University Press. Baron, Larry, and Murray A. Straus. 1987. Legi timate violence, violent Attitudes, and rape: A test of the cultural spillover theory. Annals of the New York Academy of Sciences 528:79-110. Berk, Richard A., Sarah F. Berk, Donileen R., Loseke, and David Rauma. 1983. Mutual combat and other family violence. In The dark side of families: Current family violence research, eds. David Finkelhor, Richard J. Gell es, Gerald T. Hotaling, & Murray A. Staus. Beverly Hills, CA: Sage. Barnard, George W, Hernan Vera, Maria I. Ve ra, and Gustave Newman. 1982. Till death do us part: A study of spouse murder. Bulletin of the American Academy of Psychiatry and the Law 10:271-280. Blau, Judith R., and Peter M. Blau. 1982. The cost of inequality: Me tropolitan structure and violent crime. American Sociological Review 47:114-129. Block, Carolyn R., and Anigone Christakos. 1995 Intimate partner homicide in Chicago over 29 Years. Crime & Delinquency 41:496-526.

PAGE 120

120 Block, Carolyn R. 2000. The Chicago Womens Health Risk St udy. Risk of Serious Injury or Death in Intimate Violence: A Collaborative Research Project. Final Report to the National Institute of Justice. Available at http://www.icjia.state.il.us/public/pdf/cwhrs/cwhrs.pdf. Boritch, Helen, and John Hagan. 1990. A century of crim e in Toronto: Gender, class, and patterns of social control, 1859 to 1955. Criminology 28:567. Box, Steven, and Chris Hale. 1983. Liberation a nd female criminality in England and Wales. British Journal of Criminology 23:35-49. Box, Steven, and Chris Hale. 1984. Liberation/Em ancipation, economic marginalization, or less chivalry: The relevance of three theoretical arguments to female crime patterns in England and Wales, 1951-1980. Criminology 22:473-497. Braithwaite, John. 1979. Inequality, Crime, and Public Policy. London: Routledge & Kegan Paul. Browne, Angela. 1985. Assault and homicide at home: When battered women kill. In Advances in Applied Social Psychology, vol. 3, eds. Michael J. Saks and Leonard Saxe. Hillsdale, NJ: Lawrence Erlbaum. Browne Angela. 1986. Assault and homicide at home: When battered women kill. In Advances in Applied Social Psychology, eds. Michael J. Saks and Leonard Saxe. Hillsdale, NJ: Erlbaum. Browne, Angela. 1987. When Battered Women Kill. New York: Free Press. Browne, Angela, and Robert L. Flewelling. 1986. Women as victims or perpetrators of homicide. Presented at the Annual Meet ing of the American Societ y of Criminology, Atlanta, October 29-November 1. Browne, Angela, and Kirk R. W illiams. 1989. Exploring the effect of resource availability and the likelihood of female-p erpetrated homicides. Law and Society Review 23:75-94. Browne, Angela, and Kirk R. Williams. 1993. Gender, intimacy, and lethal violence. Gender & Society 7:78-98. Browne, Angela, Kirk R. Williams, and Donald G. Dutton. 1999. Homicide between intimate partners: A 20-year review. In Homicide: A Sourcebook of Social Research, eds. M. Dwayne Smith and Margaret A. Zahn. Thousand Oaks, CA: Sage. Brush, Lisa D. 1990. Violent acts and injurious outcomes in married couples: Methodological issues in the National Survey of Families and Households. Gender & Society 4:56-67. Byrne, James M., and Robert J. Sampson. 1986. The Social Ecology of Crime. New York: Springer-Verlag.

PAGE 121

121 Bureau of Justice Statistics. 2006. Homicide trends in the United States. Washington, DC. Available at www.ojp.usdoj. gov/bjs/homicide/homtrnd.htm Bureau of Labor Statistics. 2008a. Employment status of the civilian noninstitutional population 16 years and over by sex, 1973 to date. Washington, DC. Available at www.bls.gov/CES Bureau of Labor Statistics. 2008b. Employment status of the civilian noninstitutional population, 1942 to date. Washington, DC. Availabile at www.bls.gov/CES Bursik, Robert J. Jr. 1988. Social disorganiza tion theories of crime and delinquency: Problems and prospects. Criminology 26:519-551. Cameron, Adrian C., and Pravin K. Trivedi. 1998. Regression Analysis of Count Data. Cambridge University Press, Cambridge. Campbell, Jacquelyn C. 1992. If I cant have yo u, no one can: Power an d control in homicide of female partners. In Femicide: The Politics of Women Killing, eds. Jill Radford and Diana E. H. Russell. New York: Twayne Publishers. Cazenave, Noel A., and Margaret A. Zahn. 1992. Women, murder and male dominance: Police reports of domestic violence in Chicago and Philadelphia. In Intimate Violence: Interdisciplinary Perspectives, eds. E. C. Viano. Washington, DC: Hemisphere. Chilton, Roland, and Dee Weber. 2000. Uniform Crime Reporting Program [United States]: Arrests by Age, Sex, and Race for Police Ag encies in Metropolitan Statistical Areas, 1960-1997 [Computer file]. 2nd ICPSR versi on. Amherst, MA: University of Massachusetts [producer]. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor]. Chimbos, Peter D. 1978. Marital Violence: A Study of Interpersonal Homicide. San Francisco: R & E Associates. Cloward, Richard A., and Lloyd Ohlin. 1960. Delinquency and Opportunity: A Theory of Delinquent Gangs. New York: Free Press. Cohen, Albert K. 1955. Delinquent Boys: The Culture of the Gang. New York: Free Press. Crowell, Nancy A., and Ann W. Burgress. 1996. Understanding Violence Against Women. Washington, DC: Research C ouncil National Academy Press. Daly, Martin, and Margo Wilson. 1988. Homicide. New York: Aldine. Daniel, A. E., and P.W. Harris. 1982. Female homicid e offenders referred for pre-trial psychiatric examination: A descriptive study. Bulletin of the American Academy of Psychiatry and Law 10: 261-269.

PAGE 122

122 DeWees Mari A., and Karen F. Parker. 2003. Women, region, and types of homicide. Homicide Studies 7:368-393. Dobash, Russell P., R. Emerson Dobash, Margo Wilson, and Martin Daly. 1992. The myth of sexual symmetry in marital violence. Social Problems 39:401-421. Dugan, Laura, Daniel S. Nagin, and Richard Rosenfeld. 1999. Expl aining the decline in intimate partner homicide: The effects of changing domesticity, womens status, and domestic violence resources. Homicide Studies 3:187-214. Dugan, Laura, Daniel S. Nagin, and Richard Rosenfeld. 2003. Exposure reduction or retaliation? The effects of domestic violence re sources on intimate-partner homicide. Law & Society Review 37:169. Durose, Matthew R., Carolyn Wolf Harlow, Pa trick Langan, Mark Motivans, Ramona R. Rantana, and Erica L. Smith. 2005. Family Violence Statistics. Washington D.C.: Department of Justice, Bureau of Justice Statistics. Ellis, Lee, and Charles Beattie. 1983. The feminist explanation for rape: An empirical test. Journal of Sex Research 19:74-93. Fagan, Jeffrey, and Angela Browne. 1994. Violence between spouses and intimates: Physical aggression between women and men in intimate relationships. In Understanding and Preventing Violence. Volume 3: Social influences, eds. Albert J. Reiss, Jr., & Jeffrey A. Roth. Washington, DC: National Academy Press. Felson, Richard B. 2006. Is violence agains t women about women or about violence? Contexts 5:21-25. Flewelling, Robert L., and Kirk R. Williams 1999. Categorizing homicides: The use of disaggregated data in homicide research. In Homicide: A Sourcebook of Social Research, eds. M. Dwayne Smith and Margaret A. Zahn. Thousand Oaks, CA: Sage. Fox, James Allan. 2005. Uniform Crime Reports [United States]: Supplementary Homicide Reports, 1976-2002 [Computer file]. Compiled by Northeastern University, College of Criminal Justice. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor], 2005. Fox, James Allan. 2004. Missing data problems in the SHR. Homicide Studies 8:214-254. Fox, James Alan, and Marianne W. Zawitz. 200 6. Homicide Trends in the United States. Washington, DC: Bureau of Justice Statis tics. Available at www.ojp.usdoj.gov/bjs/ Gallup-Black, Adria. 2005. Twenty years of rural and urban trends in fam ily and intimate partner homicide. Homicide Studies 9:149-173.

PAGE 123

123 Gardner, William, Edward P. Mulvey, & Esther C. Shaw. 1995. Regression analyses of counts and rates: Poisson, overdispersed po isson, and negative binomial models Psychological Bulletin 118:392-404. Gartner, Rosemary, Kathryn Baker, and Fred C. Pampel. 1990. Gender stratification and the gender gap in homicide victimization. Social Problems 37:593-612. Gauthier, DeeAnn K., and William B. Bankston. 1997. Gender equality and the sex ratio of intimate killing. Criminology 35:577-600. Greene, William H. 2000. Econometric Analysis. Macmillin. Goetting, Ann. 1988. Patterns of homicide among women. Journal of Interpersonal Violence 3:3-19. Goetting, Ann. 1995. Homicide in Families and Other Special Populations. New York: Springer. Greenberg, Stephanie W.,William M. Rohe, and Jay R. Williams. 1982. Safe and Secure Neighborhoods: Physical Characteristics and Informal Territorial Control in High and Low Crime Neighborhoods. Washington, DC: National Institute of Justice. Greenburg, David F., and Valeri e West. 2001. State prison populations and their growth, 19711991. Criminology 39: 615-653. Greenfield, Laurence A., Michael R. Rand, Diane Craven, Patsy A. Klaus, Craig A. Perkins, Cherly Ringel, Greg Warchol, Cathy Ma ston, and Fox, James Allan. 1998. Violence by Intimates: Analysis of Data on Crimes by Current or Former Spouses, Boyfriends, and Girlfriends (NCJ-167237). Washington, DC: U.S. Department of Justice. Harris, Anthony R., Stephen H. Thomas, Gene A. Fisher, and David J. Hirsch. 2002. Murder and medicine: The lethality of criminal assault 1960-1999. Homicide Studies 6:128166. Hausman, Jerry, Bronwyn H. Hall, and Zvi Grilic hes. 1984. Econometric models for count data with an application to the patents-R & D relationship. Econometrica 52:909-938. Heimer, Karen. 2000. The nature of crime: Conti nuity and change. Changes in the gender gap in crime and womens economic marginalization. Criminal Justice 1:427-483. Heimer, Karen, Stacy M. Wittrock, and Halim e Unal. 2005. Gender, crime and the economic marginalization of women. In Gender and Crime: Patterns of Victimization and Offending, eds. Karen Heimer and Candace Kruttschitt. New York: New York University Press. Hunnicutt, Gwen, and Lisa M. Broidy. 2004. Liberation and economic marginalization: A reformulation and test of (formerly?) competing models. Journal of Research in Crime and Delinquency 4:130-155.

PAGE 124

124 Johnson, David R. 1995. Alternative methods for the quantitative analysis of panel data in family research: Pooled time-series models. Journal of Marriage and the Family 57:10651077. Junger-Tas, Josine and Ineke Haen Marshall. 1999. The self-report methodology in crime research. In Crime and Justice: A Re view of the Research, ed. Michael Tonry. Chicago: University Press. Jurik, Nancy C., and Russ Winn .1990. Gender a nd homicide: a comparison of men and women who kill. Violence and Victims 5:227-242. Kaukinen, Catherine. 2004. Status compatibil ity, physical violence, and emotional abuse in intimate relationships. Journal of Marriage and Family 66:452. Land, Kenneth C., Patricia L. McCall, and Lawr ence E. Cohen. 1990. Structural covariates of homicide rates: Are ther e any invariances across ti me and social space? American Journal of Sociology 95:922-963. Langan, Patrick A., and Christopher A. Innes. 1986. Preventing Domestic Violence Against Women. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics. Maddala, G. S. 1983. Limited-dependent and Qual itative Variables in Econometrics. New York: Cambridge University Press. Makepeace, James M. 1989. Dating, livi ng together, and courtship violence. In Violence in Dating Relationships: Emerging Social Issues, eds. Maureen. A. Pirog-Good and Jan E. Stets. New York: Praeger. Makepeace, James M. 1997. Courtship violence as process: A developmental theory. In Violence Between Intimate Partners: Patterns, Causes, and Effect, ed. Albert P. Cardarelli. Boston: Allyn & Bacon. McCall, Patricia L., and Karen F. Parker, K. F. 2005. A dynamic model of racial competition, racial inequality, and interracial violence. Sociological Inquiry 75:273-293. Merton, Robert K. 1968. Social Theory and Social Structure. New York: Free Press. Messner, Scott. F., and Richard Rosenfeld. 1999. Social structure and homicide: Theory and research. In Homicide Studies: A Sourcebook of Social Research, eds. M. Dwayne Smith and Margaret Zahn. Thousand Oaks, CA: Sage. Messner, Steven F., Glenn D. Deane, Luc Anselin, and Benjamin Pearson-Nelson. 2005. Locating the vanguard in rising and fall ing homicide rates across U.S. cities. Criminology 43:661-696. National Coalition Against Domestic Violence. 1991. 1991 National Directory of Domestic Violence Programs: A Guide to Community Shelter, Safe Home and Service Programs. Washington, DC: National Coaliti on Against Domestic Violence.

PAGE 125

125 National Coalition Against Domestic Violence. 1999. 1999-2000 National Directory of Domestic Violence Programs: A Guide to Community Shelter, Safe Home and Service Programs. Denver, CO: National Coalition Against Domestic Violence. Nisbett, Richard E., and Dov Cohen. 1996. Culture of Honor: The Psychology of Violence in the South. Boulder, CO: Westview. Osgood, D. Wayne. 2000. Poissonbased regression analysis of aggregate crime rates. Journal of Quantitative Criminology 16:21-43. Osgood, D. Wayne, & Chambers, Jeff F. 2000. So cial Disorganization outside the metropolis: An analysis of rural youth violence. Criminology 38:81-116. Parker, Karen F. 2004. Industrial shift, polarized labor markets, and ur ban violence: Modeling the dynamics between the economic transf ormation and disaggregated homicide. Criminology 42:619-644. Parker, Karen. F., Patricia L. McCall, and Kenneth C. Land. 1999. Determining social-structural predictors of homicide. In Homicide: A Sourcebook of Social Research, eds. M. Dwayne Smith and Margaret A. Zahn. Thousand Oaks, CA: Sage. Paulsen, Derek J., and Victoria E. Brewer. 2000. The spousal SROK revisited: A comparison of Chicago and Houston intimate partner homicide ratios. Gender Issues 18:88-100. Peterson, Ericka S. L. 1999. Murder as se lf-help: Women and intimate partner homicide. Homicide Studies 3:30-46. Puzone, Carol, A., Linda E. Saltzman, Marcie-Jo M. Kresnow, Matie P. Thompson, and James A. Mercy. 2000. National trends in intimate partner homicide. Violence Against Women 6:409-426. Reckdenwald, Amy, and Karen F. Parker. 2008. The influence of gender inequality and marginalization and types of female offending. Homicide Studies 12:208-226. Reidel Michael, and Joel Best. 1998. Patterns in intimate partner homicide. Homicide Studies 2:305-320. Rodriquez, S. Fernando, and Veronica Henders on. 1995. Intimate homicide: victim-offender relationship in female-perpetrated homicide. Deviant Behavior: An Interdisciplinary Journal 16:45-57. Rosenfeld, Richard. 1997. Changing relationships between men and women: A note on the decline in intimate partner homicide. Homicide Studies 1:72-83. Rosenfeld, Richard. 2000. Pattern s in adult homicide: 1980-1995. In The Crime Drop in America, eds. Alfred Blumstein and Joel Wallman. Cambridge, UK: Cambridge University Press.

PAGE 126

126 Russell, Diane E. 1975. The Politics of Rape: The Victims Perspective. New York: Stein and Day. Saunders, Daniel, and Angela Browne. 2000. Intimate partner homicide. In Case Studies in Family Violence, 2nd, eds. Robert T. Ammerman and Michael Hersen. New York: Plenum Press. Schechter, Susan. 1982. Women and Male Violence. Boston: South End Press. Schwartz, Martin D. 1987. Gender a nd injury in spousal assault. Sociological Forum 20:61-75. Shackelford, Todd K. 2001. Partner-killing by wo men in cohabiting relationships and marital relationships. Homicide Studies 5:253-266. Shackelford, Todd K., and Jenny Mouzos. 2005. Partne r killing by men in cohabiting and marital relationships. Journal of Interpersonal Violence 20:1310-1324. Sherman, Lawrence, and Richard Berk. 1984. Th e specific deterrent effects of arrest for domestic assault. American Sociological Review 49: 261-272. Silverman, Robert, A., and S. K. Mukherjee. 19 87. Intimate homicide: An Analysis of violence social relationships. Behavioral Sciences and the Law 5:37-47. Smith, M. Dwayne, and Victoria R. Brewer. 1992. A sex-specific analysis of correlates of homicide victimization in United States cities. Violence and Victims 4:279-86. Smith, Paige Hall, Kathryn E. Moracco, and John D. Butts. 1998. Partner homicide in context. Homicide Studies 2:400-421. Smith, M. Dwayne, and Margaret A. Zahn. 1998. Homicide: A Sourcebook of Social Research. Thousand Oaks, CA: Sage. STATA. 2003. Cross-sectional time-s eries reference manual. STATA Press. Steffensmeier, Darrell, and Da na Haynie. 2000a. Gender, structural disadvantage, and urban crime: Do macrosocial variables al so explain female offending rates? Criminology 38:403-438. Steffensmeier, Darrell, and Dana Haynie. 2000b. Th e structural sources of urban female violence in the United States. Homicide Studies 4:107-134. Steffensmeier, Darrell, and Cathy Streifel. 1992. Time-series analysis of the female percentage of arrests for property cr imes, 1960-1985: A test of alternative e xplanations. Justice Quarterly 9:77-103.

PAGE 127

127 Stets, Jan E., and Murray A. Straus .1990. Gende r differences in reportin g marital violence and its medical and psychological consequences. In Physical violence in American families: Risk factors and adaptations to violence in 8,145 families. Murray A. Straus and Richard J. Gelles. New Brunswick, N.J.: Transaction. Stout, Karen D., and Patricia Browne. 1995. Le gal and social differences between men and women who kill intimate partners. Affilia 10:194-205. Stark, Evan, Ann Flitcraft, and William Frazier. 1979. Medicine and patriarchal violence: the social construction of a "private" event. International Journal of Health Services 9:461-493. Straus, Murray A. 1993. Physical assau lts by wives: A major social problem. In Current Controversies on Family Violence, eds. Richard J. Gelles and Donileen R. Loseke. Thousand Oaks, CA: Sage. Straus, Murray A., Richard J. Gelle s, and Suzanne K. Steinmetz. 1980. Behind closed doors: Violence in the American family. Doubleday, Garden City, NY. Totman, Jane. 1978. The Murderess: A Psychological Study of Criminal Homicide. San Francisco, CA: R & E Research Associates. U.S. Bureau of the Census. 1994. Census of P opulation and Housing, 1990. Summary Tape File 1a and 3a. Washington D.C.: U.S. Department of Commer ce, Bureau of the Census, 1991. U.S. Bureau of the Census. 2003. Census of P opulation and Housing, 2000. Summary Tape File 1 and 3. Washington, D.C.: U.S. Departme nt of Commerce, Bureau of the Census, 2002. U.S. Department of Justice, Federal Bureau of Investigation. Uniform Crime Reporting Program Data [United States]: Police Employee (LEOKA) Data, 2000 [Computer file]. Compiled by the U.S. Dept. of Justice, Federal Bureau of Investigation. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor], 2002. Vieraitis, Lynne M., Marian R. Williams. 2002. Assessing the imp act of gender inequality on female homicide victimization across U.S. cities: A racially disaggregated analysis. Violence Against Women 8:35-63. Villarreal, Andreas. 2004. The Social Ecol ogy of Rural Violence: Land Scarcity, theOrganization of Agricultural Producti on, and the Presence of the State. American Journal of Sociology 110:313-348. Walby, Sylvia. 1986. Patriarchy at Work. Minneapolis, MN: University of Minnesota Press. Whaley, Rachel Bridges, and Steven F. Messner. 2002. Gender equality and gendered Homicides. Homicide Studies 6:188-210.

PAGE 128

128 Williams, Kirk R., and Robert L. Flewelling. 1987. Family, acquaintance, and stranger homicide: Alternative procedures for rate calculations. Criminology 25:543-560. Williams, Kirk R., and Robert L. Flewelling 1988. The Social Production of Criminal Homicide: A Comparative Study of Disaggregat ed Rates in American Cities. American Sociological Review 53:421-431. Williams, Joyce E., and Karen A. Holmes. 1981. The Second Assault: Rape and Public Attitudes. Westport, CT: Greenwood. Wilson Martin. 1989. Spousal conflict and vi olence in evolutionary perspective. In Sociobiology and the Social Sciences, eds. Robert W. Bell and Nancey J. Bell. Texas Tech University Press. Wilson Margo, and Martin Daly. 1992a. Till death do us part. In Femicide: The Politics of Woman Killing, eds. Jill Radford and Diana E. H. Russell. Boston MA: Twayne. Wilson, Margo, and Martin Daly. 1992b. Who kills whom in spouse killings? On the exceptional sex ratio of spousal homicid es in the United States. Criminology 30:189-216. Wilson, Margo, Martin Daly, and Christine Wr ight. 1993. Uxoricide in Canada: Demographic risk patterns. Canadian Journal of Criminology 35:263-291. Wilson, Margo, Holly Johnson, and Martin Daly 1995. Lethal and nonlethal violence against wives. Canadian Journal of Criminology 37:331-361. Wolfgang, Marvin E. 1958. Patterns of Criminal Homicide. Philadelphia: University of Pennsylvania. Yang, Bijou, and David Lester. 1988. The participation of females in the labor force and rates of personal violence (suicide and homicide). Suicide and Life-Threatening Behavior 18:270-78.

PAGE 129

129 BIOGRAPHICAL SKETCH Am y Reckdenwald grew up in Ithaca, New Yo rk. After graduating from high school she moved to Pittsburgh, Pennsylvania where she obta ined her bachelors degree in psychology with a double major in statistics from Carnegie Mellon University. After completing her undergraduate degree in 2001 she moved to Tampa, Florida and later re ceived her masters degree in criminology from the University of South Florida in 2004. In 2008 she completed her Ph.D. in criminology, law and society at the Univer sity of Florida in Gainesville, Florida. After graduating from the University of Florida, Dr. Amy Reckdenwald decided to remain in Florida and accepted a position as an Assistant Professor at Florida Atlantic University, in Boca Raton, Florida.