<%BANNER%>

A Model of Gentrification: Monitoring Community Change in Selected Neighborhoods of St. Petersburg, Florida Using the An...

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 E20101130_AAAACU INGEST_TIME 2010-11-30T16:36:48Z PACKAGE UFE0010582_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 4661 DFID F20101130_AABWJE ORIGIN DEPOSITOR PATH nesbitt_a_Page_75thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
a6b8d3b03ac1c5ba10c44c182540bfcc
SHA-1
9d11419c624b3496dbe7512118c2ca013f28b372
2004 F20101130_AABWEH nesbitt_a_Page_61.txt
6c61289bd577026f998fcb76b96e29ac
60871fa17ee42207caf46a6304ea379aa02f3d8c
107867 F20101130_AABVXJ nesbitt_a_Page_71.jp2
6db2a5cfc7e8c9d13731b48594794499
9cddc91198c68934ce868d941992bca6e5df8f22
111851 F20101130_AABVNP nesbitt_a_Page_55.jp2
96b801c9e7a49705c15dd7f18dd6e230
70250d7f13bb677ae679d6743bf7ef60168be1f6
1517 F20101130_AABWJF nesbitt_a_Page_76thm.jpg
2d5eb0fc5a5802ff36de075022630c15
b5b63fd05b7ec502ba8f314fa898a09e51976b59
2222 F20101130_AABWEI nesbitt_a_Page_63.txt
bbeaf4775c15065ecd86edbcedbe8794
59ac8b61a36b32f322f8f1dae4145177c706cbf8
76557 F20101130_AABVXK nesbitt_a_Page_72.jp2
2b3bbce6b0edaa2fb59ebebd9ea903c6
23987499df749d523cdc25f5f24f6f75c5baf35d
49459 F20101130_AABVNQ nesbitt_a_Page_58.pro
dbd751c2a4a7a17f364c2c1baab30504
5b1b2e22b422f76d6a27ff52f0504d9d7a9b5f97
12223 F20101130_AABWJG nesbitt_a_Page_77.QC.jpg
8699cdbc5906eefe51327256408a90d0
6f7139afc94e0fae662f0a545e8146f094d5cefa
1865 F20101130_AABWEJ nesbitt_a_Page_64.txt
f3a09fe0dcfb4602966283715ec2be7f
a14b1ccc4ac0c5e4587f918bc9ed60fed51c9915
96409 F20101130_AABVXL nesbitt_a_Page_73.jp2
566bcce950cd73dc0c2a28a6716893e9
46842cac28e9d1f402e6ebb0bb0bdddc66eb3f82
27843 F20101130_AABVSO nesbitt_a_Page_01.jpg
835e6991dfce7bd26691174f40e01ae1
31f1356b8a13212ee43eb37d39066e793f224881
3330 F20101130_AABVNR nesbitt_a_Page_03.QC.jpg
45c2244864e6185c0ffb617788e35c4b
ddd967599ba818b278cda90121441494d6574392
3508 F20101130_AABWJH nesbitt_a_Page_77thm.jpg
89be94740d8dc2a428725c4803726fdc
3b7ae9d428cb13883cda6bdc35d52d2f0ac3da46
2005 F20101130_AABWEK nesbitt_a_Page_65.txt
9ede6af15bb94d40f2e40e8c83713458
82e4712c1678a85c8375949e22700c8a7ac1e8d2
7132 F20101130_AABVXM nesbitt_a_Page_76.jp2
4fbaff9231f658a71e307d7ee518e0c7
08515f6e0416c82ccd2370e33d5c79cca01eff3b
10309 F20101130_AABVSP nesbitt_a_Page_02.jpg
4c9f760958c543ee9899d9a60f145430
d9dfd38f7089a00fa2a375ef0fb35b90d21efb1b
82866 F20101130_AABVNS nesbitt_a_Page_07.jpg
79fe728dfa2f063aed35ff0d20e8f554
186da23a59d9d4e18e096326d8092cf493074823
7530 F20101130_AABWJI nesbitt_a_Page_78.QC.jpg
2d03e78d3ba0591a44b0df6814cdcdc2
c81bc1d26c076ff8a6840610f6e2da7300ce2ace
169 F20101130_AABWEL nesbitt_a_Page_66.txt
143238db5b11f8bc65f850fa414f6014
0718b0bb4dd13b67aff0bf45d6b302263b1729e5
1045599 F20101130_AABVXN nesbitt_a_Page_77.jp2
c4855929c6435e2b965a26b5451aa2b3
b324d62d183fe4f94cd3af5c27cb69655e1cf624
10273 F20101130_AABVSQ nesbitt_a_Page_03.jpg
0bb5f3129952221e3c9998a86d238d8a
f2c7cf5609ba305471d9485fb1c2330633369b79
51053 F20101130_AABVNT nesbitt_a_Page_63.pro
b62160dec4d9e57937735fa34aacbba3
0c413a8ecb870ddb1524794be4d14254bd8183d1
2513 F20101130_AABWJJ nesbitt_a_Page_78thm.jpg
eb9da77d47b9410b70dd522c42b307ba
22808271c0661e8b00cb2a6794ef436b5ae80f75
1765 F20101130_AABWEM nesbitt_a_Page_67.txt
d1a4de674dac268c9aa33a0ae55c171c
bd23186caa0696376f9300321e82196513e6d9bb
19966 F20101130_AABVSR nesbitt_a_Page_05.jpg
91778a9bfe6804f863d10dd1af9bb20b
a44212b89389646eb7794be3c1430267b1792d62
1053954 F20101130_AABVNU nesbitt_a_Page_74.tif
5cc239f618cd22252b875af94d8b71b6
9ffd15bc4ea876320ef55e2e931370b9340b679d
305301 F20101130_AABVXO nesbitt_a_Page_78.jp2
f3c105661f8689e5035292246388d19c
f0000a4066b5b7cb3addc54b237aef90ecd0d9c7
10432 F20101130_AABWJK nesbitt_a_Page_79.QC.jpg
3d85e4bfd1850c092feef1e572a81d0c
23c20855eba5beb26d20644df455682f4389fdc7
1955 F20101130_AABWEN nesbitt_a_Page_69.txt
b1e32ad44d736b2d09bf5df5f9748621
4e03a45631f272889abbd73e2cd743ab6a855661
62129 F20101130_AABVSS nesbitt_a_Page_06.jpg
004bc04c57e0713bfa5b659156437be4
fd321dd3e09f31de1d209cee026d717554a3f766
6501 F20101130_AABVNV nesbitt_a_Page_03.jp2
ead5df6dfed7d41eaed1752332e9a71c
c8260cabff2df137a8dc67411e0fefebbf0b55a4
428251 F20101130_AABVXP nesbitt_a_Page_79.jp2
b25f699baf65c854515378e2fd3148ed
16e857df51e46fa85bc29224f4e759e7d041b0c3
3498 F20101130_AABWJL nesbitt_a_Page_79thm.jpg
d4549d7fa9a1cd7407c91c50e49b0d4c
57ed23db88ee6fae66f82a06ba25c0964818bd7c
1908 F20101130_AABWEO nesbitt_a_Page_70.txt
e884b67457ea08136cd935fa75d7d324
7f1ddd071300209a2aadd06016c1c0fca9af3c0c
58242 F20101130_AABVST nesbitt_a_Page_08.jpg
326d5c4028efd6dbc8c130e317d6fe24
2ef1221c7e0ac73d16e3b35fbbfe0f5ed3d0e4a1
F20101130_AABVNW nesbitt_a_Page_36.tif
a1c2b21e5eec0821d01ffc98c044487c
4fe2f58ef316205c22d5ea1cd1fc9012bdb1f573
113362 F20101130_AABVXQ nesbitt_a_Page_80.jp2
348390b67e60e30e0b8691a1339e0dce
42b1d34b5f052c25ae754ccf098ac3d8500870a0
6176 F20101130_AABWJM nesbitt_a_Page_80thm.jpg
cfc4c0539ee280573b6d4ca74b9c3a88
b8ddec5bc2bd5e50f2d44f1dfa70243172f10698
1369 F20101130_AABWEP nesbitt_a_Page_72.txt
b866790d195ad25341f3412c77bd1167
74312ffb311d63200f2b900f9b533cb07d326899
17823 F20101130_AABVSU nesbitt_a_Page_09.jpg
9ee9038e59004fc015d56ca225c1aad3
2e5b2430b82fd660ad12cbd77e203775844edd08
1546 F20101130_AABVNX nesbitt_a_Page_14.txt
0208a3bc26944848ff313a3f7667f724
6cddf5e9c30b5710f940eccbcd72d78b65cbf21f
121246 F20101130_AABVXR nesbitt_a_Page_81.jp2
9ceb154714849d70c864eb9b1105fb61
6a6971d948b2c1f4a17237351597053c00c84c42
24912 F20101130_AABWJN nesbitt_a_Page_81.QC.jpg
92db3f5db9b900380ac658a16bf39cfe
d67c96a43038620a8611e912a6eb1755d3a978b9
16286 F20101130_AABVSV nesbitt_a_Page_10.jpg
d007e656ca151d31b5f10afabb41394a
d939ef6c594b46d357554034aad01dcdea83e534
F20101130_AABVNY nesbitt_a_Page_47.tif
eac9307c29ac402009dcbc405527b78f
f975f85b12a30397498a2ded5d6404ec9f0d7000
88557 F20101130_AABVXS nesbitt_a_Page_83.jp2
20770baf1f67afe17787d3b35513807c
95ff314b80c9a19af029e1ab058af7f6279d4775
5837 F20101130_AABWJO nesbitt_a_Page_82.QC.jpg
8001e9f8fc9893e5180d1f86fbb34f27
7ab1a7a9429a6e863431693d7af0918527c12d87
2872 F20101130_AABWEQ nesbitt_a_Page_73.txt
fabe3eb95d284f9051095b4a3d05766a
4fae51baaf6bb65b333be0594efd81e2addb4135
57283 F20101130_AABVSW nesbitt_a_Page_11.jpg
7da4ddb4ef9efddeb4649bacae70e7c5
0b6bb041437f9f77c00a26059a6171126cf90d33
F20101130_AABVNZ nesbitt_a_Page_21.tif
ec016a0e501abb1ee6462cf44832e363
28f237c90d67d3d15f15a0d15a65e63958c1d474
51971 F20101130_AABVXT nesbitt_a_Page_84.jp2
20c1d9e03869c3b0aa1c16c349988adb
de04523649fa21d071a70f82f2653aed9a51a3a1
2074 F20101130_AABWJP nesbitt_a_Page_82thm.jpg
90b0c56733bfffc281d93aedeff8baca
5b2f1f3ada92b521c933263996506f98c7b0ee5d
3284 F20101130_AABWER nesbitt_a_Page_74.txt
55ac5f9a42cd9dd12fb636e6e9ea4ffe
01de47133aaaef43bc3e7a65668fa899ad45c98d
47842 F20101130_AABVSX nesbitt_a_Page_12.jpg
423d4de9f3e2cfb2e5847603245cb2fe
61dbb9cf299108db3498bdba1d3b9e39e2a72f3a
F20101130_AABVXU nesbitt_a_Page_02.tif
87890d3151a687b123f1e416da185e63
684ee2bc24c3b4869e66797d67580abc30bbae4d
19369 F20101130_AABWJQ nesbitt_a_Page_83.QC.jpg
924dbb3c6793a3a8674f0989fa72280b
e22d1912e7dbc6bac0d3c6cceac294b6598ca043
1730 F20101130_AABWES nesbitt_a_Page_75.txt
5286ce8251f5bbf5d8f0bbb4f10af7f1
ff0b05f72ecb25d8e35e39ee6cedb33a2d1f6600
66198 F20101130_AABVSY nesbitt_a_Page_13.jpg
9759a223cc34b6c9988436923e6b4595
8bb7cc964babbd93f33abadd19a4f7aa4e500131
F20101130_AABVXV nesbitt_a_Page_03.tif
d951bd32f589d14ad22b7cdbecadc844
0d53a2056db0801d1f9f225a2f777c7ef1b37d27
5570 F20101130_AABWJR nesbitt_a_Page_83thm.jpg
a7c77260d4b5085b99c7c4035bc5e30c
2935a71be09d020226a55f9f1c1db6353254a401
836 F20101130_AABWET nesbitt_a_Page_77.txt
0c9f0cd2a18e783b4704a87549793045
86db2651c3fb7762c1b1336bacd2dad62b1b1952
30450 F20101130_AABVQA nesbitt_a_Page_79.jpg
f6532fa454e9bafdd73c3ed9644d17f8
8f82f4244b972af82636b04d2c7b459a13ed7836
57231 F20101130_AABVSZ nesbitt_a_Page_14.jpg
56f95a537b1dc55d9ab31ae65ac96670
31c7c60b6740a6b3edecbb6db28499233e220d4a
F20101130_AABVXW nesbitt_a_Page_04.tif
0ebc69ce367c4250cd80a60f2c7507a8
e71d2cfbc16eedd8c09978f819cfeb9428dd9b1a
12346 F20101130_AABWJS nesbitt_a_Page_84.QC.jpg
8cc9b1ec525762a9c22c06fc93fb3111
b60216352bd4303f0195d9f945ee415367f28a4c
151 F20101130_AABWEU nesbitt_a_Page_78.txt
e18d2414c28f7e9d8affd143ada4447a
0de5804ab4c2652b1177ffe909269d8356a85a45
F20101130_AABVXX nesbitt_a_Page_05.tif
84921118d075cd3525432dd24407974d
0253b7d1c97862ccabcb3d9b4431da09fa21c911
24247 F20101130_AABVQB nesbitt_a_Page_27.QC.jpg
44cf2b27935078f030b4ea92f0194ce3
7a99f4059b6cab8078303d87ae98827e758004a8
98592 F20101130_AABWJT UFE0010582_00001.mets FULL
1912a2f3473dfaf88591c6bbf8d252ef
514ebdd02b820f1aab1af6de3f778fc88f785205
287 F20101130_AABWEV nesbitt_a_Page_79.txt
dd2b56034d8ac86dbea1162d619bf1ff
6bd77e8c3ebbc97a1d7316e3cba1099c8aecccbd
87248 F20101130_AABVVA nesbitt_a_Page_81.jpg
479dfd151f92fb3ad8abdfb4cca7aeb6
019d441c2a4c026db8b21ed76298cce4d8a53cee
25271604 F20101130_AABVXY nesbitt_a_Page_06.tif
5c5a8060c8d97b5b8a01623faaadd068
4411eb5180a22e91d986a07848fbf716495922d2
3793 F20101130_AABVQC nesbitt_a_Page_76.QC.jpg
36922272128edfe738302fd0deb48252
54cc9f67e6ca2e6c2ef95ce35930e023ad2dfab6
2137 F20101130_AABWEW nesbitt_a_Page_80.txt
4a0c9d691e3d1537c5c5d977df07ca88
ffad26f1c8a12742b2821d944e6bce3e132c396d
19190 F20101130_AABVVB nesbitt_a_Page_82.jpg
199bb8e552405cfa9632aa6ccb7e7889
9bb4b170b68b05e1406c34ba224c5574af05c700
F20101130_AABVXZ nesbitt_a_Page_09.tif
98c337bd8de2f73a581643b05043b262
75474459c17300edc5f74fd8283bd597e655e1dc
34402 F20101130_AABVQD nesbitt_a_Page_72.pro
d83fcfa1247b24c883d2e3906a60d790
84463bde263ac0d9f7784a128c84f9f6cdb40051
373 F20101130_AABWEX nesbitt_a_Page_82.txt
cb05552573ec218689fe00dead7f7d7d
0153377fe8664276b0bddffea07a8b5c5f92fee0
58914 F20101130_AABVQE nesbitt_a_Page_45.jpg
bb563bba8172a4f3d98eef04f93bd090
60f38105b89a20c4405fa9e6808d868cc73d8fe9
1632 F20101130_AABWEY nesbitt_a_Page_83.txt
4374e0570236541a3f9fc04e14a68b34
a6bf979dbaac98fc4b0455523751bc7c5e300daa
50156 F20101130_AABWCA nesbitt_a_Page_68.pro
93af062707e017fe2ca867646613e9d0
c2ec2086be0cd67c68599c115d728cf0a57f9bf5
60124 F20101130_AABVVC nesbitt_a_Page_83.jpg
3b9c6e459ebe67679c49019b5cb461f0
2e571d69803ca91faaa931dcc7cac410690ccf69
62257 F20101130_AABVQF nesbitt_a_Page_67.jpg
3282af2a04e884a5a85a261b467e75a2
4fa08ab314f1c49a60ba7cd35a53efc860248d82
897 F20101130_AABWEZ nesbitt_a_Page_84.txt
b7bb84a5dfdd214ce55580d22f1abe96
acaf8698253f0de4ceeada255bab93a0c45fa53e
48996 F20101130_AABWCB nesbitt_a_Page_69.pro
eff93a1f591835bb6f30fc1a59321326
74a32ef790f57ddd9e2b29df64eb06c34e2acfe3
37803 F20101130_AABVVD nesbitt_a_Page_84.jpg
c993392a6361eb3c8e748dab8dea8231
26ea31e5095a7a8c4555c01b2a308affc897312d
51117 F20101130_AABVQG nesbitt_a_Page_20.pro
eda3d21b06da89098a2b486afe24ecfd
e0dc24b65f77535de1e6dacfac6ee333b82c400e
48315 F20101130_AABWCC nesbitt_a_Page_70.pro
ced7f41048bd7ee97d544a9d1a1120e0
4fa1fab34fd9d7a3183f4627b2a771900a4d76ee
5539 F20101130_AABVVE nesbitt_a_Page_02.jp2
9d29805444893062ac70fc8160e38ad9
805d991270761641eae2bc84155e5b339e0e2783
F20101130_AABVQH nesbitt_a_Page_07.tif
2101bb60343c3efeb209865cce08cd3f
a5b506cfd2d86185973d9a36e364f479da013999
23369 F20101130_AABWHA nesbitt_a_Page_39.QC.jpg
637bc6f29c21f5941fb75fe911f9763b
f1ff17cd6233d13534c160bf43bc68e8cc201d9e
49510 F20101130_AABWCD nesbitt_a_Page_71.pro
12be300bbec0cbb161c4cabed37c71fe
d3819384a3808228ec1e51b0807ffcee2a28599b
91888 F20101130_AABVVF nesbitt_a_Page_04.jp2
bf60b40309faad0b4429013e9dccf690
2365629790fb60ab446c30855e056bd6528127e5
106759 F20101130_AABVQI nesbitt_a_Page_57.jp2
566670859f35cabf8e3416c05408cc77
dad27d36472ff44f6ef7347c1b0e1f7e30acd156
6299 F20101130_AABWHB nesbitt_a_Page_39thm.jpg
35b19047dc1c5fa05d239e39e3c2b310
0f3b3a3a55bb4c3962ffebc90ebe6a12ec122548
67837 F20101130_AABWCE nesbitt_a_Page_74.pro
4fed9fb687cda7a10f2e6dc5872f5471
b17179f372638ef4bbb3556fdd3d1924d3f37778
22275 F20101130_AABVVG nesbitt_a_Page_05.jp2
8d1388673025db0cce235a466160bd60
22304f747334e1f94ad7713b4f81cb10d91e392d
22653 F20101130_AABVQJ nesbitt_a_Page_16.QC.jpg
d9834eeb9dda2d8c1fac08977b404a1a
c55ae1074e931932f408c661c706f740731292b8
20667 F20101130_AABWHC nesbitt_a_Page_40.QC.jpg
f52895e7363c84c3b4bc9b9c02820657
3336f47bbc3590c1dd58202d8948c2ab7e66ab81
36746 F20101130_AABWCF nesbitt_a_Page_75.pro
257f3a09efc9af04c3d63ce37af95d90
fc6c78915dd0cc534ad65f2e6b6aee055555607b
1051981 F20101130_AABVVH nesbitt_a_Page_06.jp2
d7235b134f5efa5feb06df8f85d5b4a3
20d970b5a305a007b006fe3d26bfa27eaf7d0e43
65360 F20101130_AABVQK nesbitt_a_Page_75.jp2
414796b032b41309b09467eb29fe2a30
81a1140c28821dd4918025e6614563ebb02ff783
5757 F20101130_AABWHD nesbitt_a_Page_40thm.jpg
a97ed8cf2e37d42b0d200688a7981b10
202cde73bcae0c2a44e606425a58b81d7ccd346f
2316 F20101130_AABWCG nesbitt_a_Page_76.pro
e3ee430f7e8af646d8c1a9873c049ba8
8589625fc9784ac9e17aa6c8bd18d2893ceae9cb
1051967 F20101130_AABVVI nesbitt_a_Page_07.jp2
87461456dc1e7b9ee317a4704ac2f303
8650b976fbf29c99774f5f8993bebb880e737b1f
F20101130_AABVQL nesbitt_a_Page_50.tif
e0c13469f265339989d2bca47e203ed1
2a69b09efd8f2fba48f86177b7c4e15c82780758
23643 F20101130_AABWHE nesbitt_a_Page_41.QC.jpg
ec39af6237b4964a32c6a1aae76c670a
4c2212d08ffffe45eef9523190758fbe4af86220
14416 F20101130_AABWCH nesbitt_a_Page_77.pro
1f5161de573930c0957175e50e927e31
801387664f55dbb35f56891f441348e9eea03711
1051984 F20101130_AABVVJ nesbitt_a_Page_08.jp2
b00c729e4bfbb9c2827006baa960cb19
e639680e17039ff0da2f58edec5c1bbc72c88f40
F20101130_AABVQM nesbitt_a_Page_77.tif
ea63e525c174ccb61b866f5986abab43
cedb6c3afb3d0311cb626447b4ae1fa8214ba504
6314 F20101130_AABWHF nesbitt_a_Page_41thm.jpg
d81217cf2e84b189c94035fe204917ba
0a6c05708d0c6aec7e478ec3c21e46c4c662e9ed
2354 F20101130_AABWCI nesbitt_a_Page_78.pro
bd546848dcb535c2ec833c0fb2ffedb3
c9fc5049a0cd5df9804691ace1e18677362cd3cf
318586 F20101130_AABVVK nesbitt_a_Page_09.jp2
f01d2eec06c2cfef96bc281bbbe7a3a7
1251f183503230a00f1ed64b0e49816e8b3ded8b
17930 F20101130_AABVQN nesbitt_a_Page_45.QC.jpg
fc512adc173e1f07ee34204a8f861d53
7386534790ef944600dbf3af66ec9929dc891a82
22143 F20101130_AABWHG nesbitt_a_Page_43.QC.jpg
1ea6172acef2d8ca5c4704cb1008294e
f594bd3e3529a59a9eaf14d98839b33f020b55ab
52760 F20101130_AABWCJ nesbitt_a_Page_80.pro
d55cb94c0426c7462a14fb308d47b77d
8bcda754a6c62342ac109d07a761ce055271d32a
227139 F20101130_AABVVL nesbitt_a_Page_10.jp2
7b27c877ecdd5fd17df7481f587a9ee8
67a92f707daada326169a9e011a1871733c26a2b
21516 F20101130_AABVQO nesbitt_a_Page_19.QC.jpg
2bc28605629807291c5f4daaa667bb93
b6ec4522ab329482a0350fd72454548caa20e5b3
6327 F20101130_AABWHH nesbitt_a_Page_43thm.jpg
bfa183bf7627c87e03c05fd6cae56d38
5476020e314476b67ce3823d09a8ce98481072a1
55528 F20101130_AABWCK nesbitt_a_Page_81.pro
6498c8c37ea781f2d3364499fca85199
b736e4340cdbb23da9a926998e859f79f16a514f
80511 F20101130_AABVVM nesbitt_a_Page_11.jp2
cab40e89e89561988ed6a06b178b7587
c94a2837e15ad537670ed9921a0466824a52528f
2008 F20101130_AABVQP nesbitt_a_Page_68.txt
9f22a48d386a129b1d9d15f05279449a
af33278e1a0fd7e363a3d49bbd75fa552e9337f8
5679 F20101130_AABWHI nesbitt_a_Page_44thm.jpg
096ad24cc5645f7259e178115890de95
2ff42ecac2dfbed99defe1f4cbba43ee678e1c3a
8077 F20101130_AABWCL nesbitt_a_Page_82.pro
e2d8585ca732028d764f4c16d1c0c487
b73e1cb08c27138feeb547cda9cbffcbe61e0e86
70129 F20101130_AABVVN nesbitt_a_Page_12.jp2
346a451f8bb6673f1081319f7fd51ce3
95f4f9b9b03bedf5343c4c3aaec07d442614e6ad
72575 F20101130_AABVQQ nesbitt_a_Page_55.jpg
98140cf2e56765a27d83d3bb7d4ea55e
d7f5e38a307851dfc814a1abf2bb71a0842b3e9a
21694 F20101130_AABWHJ nesbitt_a_Page_46.QC.jpg
09b28df6a387145574597d174ddbc25e
fb7ca0fe0382fb7a79006b411a7f9b2760adcde2
22382 F20101130_AABWCM nesbitt_a_Page_84.pro
63b12d20ae2b49a81af1e6e462320ae2
130ba7c4e84d39deb235c820622f854b119c58c0
99114 F20101130_AABVVO nesbitt_a_Page_13.jp2
3ec571336eeec45b08ef6374b4270cfb
7cbdeba77c39d5df167c24e84cc5bf42ceeddb93
6269 F20101130_AABVQR nesbitt_a_Page_18thm.jpg
3554d6fcd6415e51c14e9bed0890496d
fe0f4d357d70959a97087e7f27a9047c25d686f9
6378 F20101130_AABWHK nesbitt_a_Page_46thm.jpg
baa78139f206da26eec1cbf155c1f3e2
fbc5c50761f347a3616658d6e8dd5b6f09feec46
537 F20101130_AABWCN nesbitt_a_Page_01.txt
0201bb926eed77609ec914963f023beb
ce110ced3f186078b39167e688f7fb1a50e42216
92609 F20101130_AABVVP nesbitt_a_Page_15.jp2
6e3b88d271f53ab4dde34cb3520240f9
8bde9bef047701b8c39fb1322af54f85516a583f
9172 F20101130_AABVQS nesbitt_a_Page_05.pro
007ead8cf7f835840fed188721c8972d
926f07f9b876fbc2030b8400b82e9570c84c60f6
21945 F20101130_AABWHL nesbitt_a_Page_47.QC.jpg
c035dccbd0ea5383e9d94d2a11579ece
e819b2d89e39591c97df0532796a3c0a04ca0d45
104806 F20101130_AABVVQ nesbitt_a_Page_16.jp2
48fe05287c539df9683147844662834b
484560a601288a733b0fb9a1f2c3c66b8aad1da0
5102 F20101130_AABVQT nesbitt_a_Page_45thm.jpg
9661cde4f3c0dd5bfdfbd797705b3571
90efd5b0822fdfef30260c23a2a4148729514f29
6226 F20101130_AABWHM nesbitt_a_Page_47thm.jpg
31a7a5689c0874044ab9f1adb8ffdf32
7293a1b9df2d4665c7ace7c33ce40a66e39a1052
77 F20101130_AABWCO nesbitt_a_Page_03.txt
f05bff5ed111b75cf17c0c7857187311
737a9113437b971980174afa3e6712d1f47e5be8
106923 F20101130_AABVVR nesbitt_a_Page_17.jp2
9fa800fbcba9b0ceda1c9e466e1217b3
bdab2d404d353ecec420e060ed38a69e254042bb
1911 F20101130_AABVQU nesbitt_a_Page_18.txt
e476d9f86123078f7acdf38064041f21
ab7b4dacd6ba8ee0e20ad34240efa397f249586d
11238 F20101130_AABWHN nesbitt_a_Page_48.QC.jpg
63c64a0dde384d24ecb71d5d6f63bb8f
0a30b91fd1a8f1424caa98cc0e350a4dc0e6bfdc
372 F20101130_AABWCP nesbitt_a_Page_05.txt
88d42fa7a274d706831e59f45865dd16
c89c23db91bba576ffeece0d66fbd44161fc5856
104578 F20101130_AABVVS nesbitt_a_Page_18.jp2
e6f0f09735b82ce10a33b1f34a06fd81
938fd60bdbe146eb5db169ac9ee111a8d0f7cc18
19578 F20101130_AABVQV nesbitt_a_Page_44.QC.jpg
13271b03b82b460b7b53dcf4fe86a9f4
6f8c1021a4d034a0f20deb4dfc2b74414b832afa
3509 F20101130_AABWHO nesbitt_a_Page_48thm.jpg
93f8a0787b498b3a6051f89553cc50eb
bbfb8ea7f843d0afae3633b6f3a1dee87c0d5664
3198 F20101130_AABWCQ nesbitt_a_Page_06.txt
ceb35564f9478752c0d8378af1ff5cfe
2d017d3653cc31ac4e20939b0b0181201f900cb5
94043 F20101130_AABVVT nesbitt_a_Page_21.jp2
6c1027148122facbfd19eaf275e75513
416dd3f10d0f87046ab1b36536af292611f12e2e
69311 F20101130_AABVQW nesbitt_a_Page_64.jpg
d5e0df6426b5444cf90d102f26b867a3
7e6f21eb5185b3f11f18149f2860772edc0029d5
19997 F20101130_AABWHP nesbitt_a_Page_49.QC.jpg
4cf85ee15fa44e117d5cec34a788fe8a
89d693805326b7fc4113672371ff6933991d455b
2580 F20101130_AABWCR nesbitt_a_Page_08.txt
efb594e16905b9e49ba14abd5bea1b96
ff7b73494e27525edc2b5c6021d3a5419c9717dd
115099 F20101130_AABVVU nesbitt_a_Page_22.jp2
f89cc61626352a350402ddf8d8500d2e
3cd15bed674e4557a2dd9365cf2e9300c1e02d1f
15650 F20101130_AABVQX nesbitt_a_Page_12.QC.jpg
f03ea586a659f1e1888a29958e42aa68
7e2969407390d5a30bbcbc5b3e1c0bc71f0a5da8
5595 F20101130_AABWHQ nesbitt_a_Page_49thm.jpg
924c32ba0d5d28db76875a193c0ebe44
1058c423135ae3245b743f72c5e6844380352e11
474 F20101130_AABWCS nesbitt_a_Page_09.txt
a491f888437717974f36bad31748c2b5
ce4efd01ebc1de333a96b0c82540ee689c204317
99182 F20101130_AABVVV nesbitt_a_Page_23.jp2
f5e6368f9ef2b7fd8509b97a52e30111
b2c6e5ca75916943d07ed819bed944465375444f
23093 F20101130_AABVQY nesbitt_a_Page_60.QC.jpg
5e839db3c1e234851c35a0dc94f47e17
76708bd42b5452fbc107b382d40cc6244df9a95a
22651 F20101130_AABWHR nesbitt_a_Page_50.QC.jpg
9e5369485eb0d36490b7e9c16daac48a
a3ff9ff88cb97f88e54ec80d2f1525270a442ac1
477 F20101130_AABWCT nesbitt_a_Page_10.txt
4c95c1049ef91a6b67f6568a8a604b98
ee07fb5788302d096e0f3b9a0e2099746901e346
111985 F20101130_AABVVW nesbitt_a_Page_24.jp2
df5937c09ad08755582bcb54fa08601b
9c555e5c2aa52d81fbccd2eb4e0eaf2bd754bc3f
23404 F20101130_AABVOA nesbitt_a_Page_55.QC.jpg
32532a15c22934d10eb0292479f01820
9a3a1b362b699301fbb97ae8bf1deecdba702eab
F20101130_AABVQZ nesbitt_a_Page_35.tif
faa1c95006cc13b7f9866170da78bc46
10ba74475a3d2948c01bb45c5627420a8527b2b2
6528 F20101130_AABWHS nesbitt_a_Page_50thm.jpg
81b1622ac22289925f5e3764dc0617d9
0c67f84622829d73decbe06376774f2daad71075
1580 F20101130_AABWCU nesbitt_a_Page_11.txt
31a5ed0c8be4304dcf114a77885cf6a1
bead8b7436534ea2826348e172ae8a17da3a48d9
99413 F20101130_AABVVX nesbitt_a_Page_26.jp2
ccef423b77813bccbd30fe5f44712797
a9d31f280837ddd4c5238c69dd379e78ff95a8b7
6684 F20101130_AABVOB nesbitt_a_Page_79.pro
d1df11de6521fdfb774c196ff32178ac
f066bdc290a47c791bf93a814144d4d17a570897
1244 F20101130_AABWCV nesbitt_a_Page_12.txt
7b79b5b82d94a897ed17fb6536806745
23115883a4a8b0947d1ae5c2956752d15f5ed7cb
112853 F20101130_AABVVY nesbitt_a_Page_27.jp2
95d4f48f706fcfda7a46152a6285955c
e501fcaaef582f231aa6cca77fd6082360c6ee49
F20101130_AABVOC nesbitt_a_Page_32.tif
b02cca59584d456c68caaeee8e4bbe4b
acffb4aef5a91b369885be5a450209e9d40edcce
6354 F20101130_AABWHT nesbitt_a_Page_51thm.jpg
058081fec9e6b743325a527e4fa37007
2c47a2e09dd004233759555c557814dc1434e0d8
1883 F20101130_AABWCW nesbitt_a_Page_13.txt
dff76c96fa82c75be7050c778dd21c2d
077fc082dfe7eae9c7c092dbfdb17a9b336f0d27
114503 F20101130_AABVVZ nesbitt_a_Page_28.jp2
e6412755cbb659c3c8d15db5500e8955
2227c01786e6c9a82c1e673b54588b0e66736927
47622 F20101130_AABVOD nesbitt_a_Page_38.pro
a0cdd6af1f2fb4cc09151a4c41258547
7ed9b61b97686c2fe12cb7d07f5dfe26313e0aa0
67466 F20101130_AABVTA nesbitt_a_Page_16.jpg
3870cf77887183c0825d26f8a937686b
a555f16c09b5c2c717be34a296dc417363cb2dc3
22018 F20101130_AABWHU nesbitt_a_Page_52.QC.jpg
114454724f5655f67edcb4edf8e7315e
e93f039848cab24c097f9fc03b801a89581bcd2e
1784 F20101130_AABWCX nesbitt_a_Page_15.txt
e556c875f6ef64712043ea5343371727
bb80cbd2b8f9fff99f25b688382b5693080840d4
59495 F20101130_AABVOE nesbitt_a_Page_73.pro
39f012aed4413b993568bd22d976d2c2
97155529969088d20be91a661f6a0e213ef0f4e7
68873 F20101130_AABVTB nesbitt_a_Page_17.jpg
4709e6396037cb45be38fe61b2766dc0
ed0c5df5a938fc1bdcd52aca6378e1ce2e4c89c3
6177 F20101130_AABWHV nesbitt_a_Page_52thm.jpg
22b837fe90b3485d0710d921e334b79d
8425673603018332237b47e639ba5bf4db1b4f52
1928 F20101130_AABWCY nesbitt_a_Page_17.txt
1b5aa7ca1c40e73145888ad6ae5156e5
d4a6e4967ec3d011fe603449ec0e47de6cd6845b
38659 F20101130_AABVOF nesbitt_a_Page_14.pro
fdec46c86f3973d6209f6adddd5b06dd
e0f49d463ecaf8b3808b0510ba42bfd2b4573934
F20101130_AABVYA nesbitt_a_Page_10.tif
fe10cb03bd2b7f939630d8ab3e94600b
8678ff28dd44963673417c4ef26f4d8a71bec6f0
10537 F20101130_AABWAA nesbitt_a_Page_01.pro
ac05e41833d263ef5b013e408b04a9ac
5b520f1f8a7faeac192085dd783d4f0bf9ad273f
68691 F20101130_AABVTC nesbitt_a_Page_18.jpg
f9ebf76b6891b033a7ed7af9e3a3a60b
7568464ad4edb643a024f3fe6ca045663c9ca82b
23318 F20101130_AABWHW nesbitt_a_Page_53.QC.jpg
cac08ab24c852999db0cd8cea966eb3a
7df97b7b7ecf064f8f860cae64651cd932cf2d09
1843 F20101130_AABWCZ nesbitt_a_Page_19.txt
c3c495e240efb55807f2b7db70f79282
257072c061ae3cd89a693a468e72cdb5608a0006
23543 F20101130_AABVOG nesbitt_a_Page_71.QC.jpg
032402730127a0179140b264aa60b053
508d35e004b688672ea97a43d2cc97ee9ba066fe
F20101130_AABVYB nesbitt_a_Page_11.tif
82c66d9cb3be658eef4b4ede97ebeb11
2f55c01c143fad5f93645eee206041f3f85438e7
1201 F20101130_AABWAB nesbitt_a_Page_02.pro
38cc737d943a30fa281ee4dbf451c9be
1a1e8ecc35e0ed98873a7618868b69f0d2db4d06
68013 F20101130_AABVTD nesbitt_a_Page_19.jpg
1987dd1db24e3a0e3b8da617f801e0c1
56628d8499a8faedb8b541ecbb93893b70de574e
6498 F20101130_AABWHX nesbitt_a_Page_53thm.jpg
9f1884abb4c056346c86905a6c2f478c
7a7571faaa66be32c62b0fae694f2bc947ff6472
35435 F20101130_AABVOH nesbitt_a_Page_11.pro
a4addca74f0dcad0dd1a20b10c9b11d7
7dbb83e839b0545c0300412a2d5f88099b300c9b
F20101130_AABVYC nesbitt_a_Page_12.tif
382645ab5c49cf6036ba13e6c53ba2b1
9d5c5f78f8832100ba97ddd95ffabd13f90bbc49
1535 F20101130_AABWAC nesbitt_a_Page_03.pro
d2d28e184eda11f908a13a8f8b4320e1
5dcf2cd2d076f0b78fde9bc043e5e24bb0a35c82
71488 F20101130_AABVTE nesbitt_a_Page_20.jpg
748e47d237495d6ed38239ee5ea29067
b644900c51494699cc619c598f6a2bc01b6728b9
6917 F20101130_AABWHY nesbitt_a_Page_54thm.jpg
682371e5f063fc89cfef25f289192a9c
9c2a14499e372869dc7c15828231a36cf4f809e9
2600 F20101130_AABWFA nesbitt_a_Page_01thm.jpg
45154cd6292445ad73b5448d580489f6
601f96c4e14d63a40386f2acc98dab0768913133
F20101130_AABVOI nesbitt_a_Page_19.tif
de9213170f9a20f7958ddda4854c5980
c5013244eaee782f8fbfb7406cc9e1584c545bf7
F20101130_AABVYD nesbitt_a_Page_13.tif
35cd8ed244e9ae7bfeb6186e18747180
e0c52f63e12fd8b9976cbf0200a44fbccad4ee02
42525 F20101130_AABWAD nesbitt_a_Page_04.pro
f59b890ee037cb68d3c893dbec89a32d
9f57786643cc5e0a0640f3423d6f6e78e141ae3b
62332 F20101130_AABVTF nesbitt_a_Page_21.jpg
8f413a6c947960da580d029be2bc6dbe
0b4d569e6821dfb00adf8f690e04995d179c9e2c
6638 F20101130_AABWHZ nesbitt_a_Page_55thm.jpg
75f202f1d8bf1251d00c86d42585feb8
80f2d5ca110965bdabce9b01a81029c598b5abd7
3308 F20101130_AABWFB nesbitt_a_Page_02.QC.jpg
e922f96b161ead6d6b6d33d6e9a59db1
db1d72b0957c051564888bd6e0f5ef19de0f7144
20309 F20101130_AABVOJ nesbitt_a_Page_07.QC.jpg
0c3ff18de9d7631923229cc9d1ccc1a9
e793f6b4ec9309d9af5547ab03c5150bb3574243
F20101130_AABVYE nesbitt_a_Page_14.tif
5368e1e3d807a6f401de067c8afbfe25
2c8219f041c371d8fbf5bd2f34b66b6a7bc9d66d
77489 F20101130_AABWAE nesbitt_a_Page_06.pro
6fb1ce724ffe13c8587b05bdb81b15fd
22eadea0fab23cdc61ece9c1806ab23a57e0eeae
65128 F20101130_AABVTG nesbitt_a_Page_23.jpg
b34e0c3207962e86bcf1b09724fc4c89
4511b4b18941a334116b4b48cb6d7400c4884474
1380 F20101130_AABWFC nesbitt_a_Page_02thm.jpg
2f0d82a006080a38e4b4976552b8ce14
c81cd8761966030c6a05193a5ea79150dd818b72
F20101130_AABVOK nesbitt_a_Page_68.tif
1b69755d53e5542d1d6f03476370bb20
d05827e8f532bd701a9a246183049bcdb12c81b3
92703 F20101130_AABWAF nesbitt_a_Page_07.pro
3aa28581ce67bb3cb472eb3434595e37
88a06afde392181203b7fdc3a18ad3519932d62e
73282 F20101130_AABVTH nesbitt_a_Page_24.jpg
ee7ed770959bc7ae8e7cb1028ef23b09
f07fbc308b9d308e6283c199aeccef255ed42a9e
1362 F20101130_AABWFD nesbitt_a_Page_03thm.jpg
374440713844df944462350553246879
2f7989d1a7b6645fe1d8409238044fef051283cb
6396 F20101130_AABVOL nesbitt_a_Page_70thm.jpg
7903596ab610692261d0cb42810e0bd1
d9f68c680fe8a364854b9725f34f9127633719c0
F20101130_AABVYF nesbitt_a_Page_15.tif
ebfa85dbfbf75af00d5a275cd9294922
cd003e63d8092ec119afa9b1111bbe17f7a12fc0
12052 F20101130_AABWAG nesbitt_a_Page_09.pro
3b9155dd3c3f396e99e2064c9f5f0ff1
4b217a2121b959a6a9fff41c1c58d4006c6cc7ac
71607 F20101130_AABVTI nesbitt_a_Page_25.jpg
bc0c15f016f183921f20210fff708f2b
75ed460f148c025c313da262f2f4badf479aa0bd
20577 F20101130_AABWFE nesbitt_a_Page_04.QC.jpg
b62945c434ea0ff52162e475079a5581
2c6150f1d65522fa48f2e6341ad9dc0f886d9a7a
73344 F20101130_AABVOM nesbitt_a_Page_31.jpg
b2a2ce82ffe70228df666544bf4c9fd4
2c79681f4cb49844f08d5a05f9b76b01ef4aa8c3
F20101130_AABVYG nesbitt_a_Page_16.tif
306f2cab5f4bc1ae1c054874d63b36b7
9271b0f9257f018ebbe70a4090dce9ff3209a7e8
9945 F20101130_AABWAH nesbitt_a_Page_10.pro
4c8627993786e3b3baa468b9a5c49d67
eeb8e41d4c575216a7179e2c781231a0edcefc3f
64642 F20101130_AABVTJ nesbitt_a_Page_26.jpg
f9f077e378970ccf7a5bb761f5cfa3d0
54418e8e3e8454073bf9000381e42b0da4182161
2254 F20101130_AABWFF nesbitt_a_Page_05thm.jpg
2428471c4427746028aa9d19ad5e23fb
c937d9530d469a33929a3dd2119fd839be682d7a
1781 F20101130_AABVON nesbitt_a_Page_56.txt
6de71a6a6e85589b836b18a14e500801
9c468f7d9fb1bb5548de5d94fd62bf2972ae5c84
F20101130_AABVYH nesbitt_a_Page_17.tif
fe6f0a6ea2928a13716be01114fdd2e6
92c7b3781db44ebb798c3c286f6af4d9fa2c89c6
30996 F20101130_AABWAI nesbitt_a_Page_12.pro
064bfdff0c01bbf579dd2efea0212f10
5eb0eb6fdbdaf7c4b319d43705cdc90cbad108dc
75179 F20101130_AABVTK nesbitt_a_Page_28.jpg
401489e6f2526f82cb2d70adc2a67494
7276d020988fdae9ecd4e63b85d6206270e9dcb7
16314 F20101130_AABWFG nesbitt_a_Page_06.QC.jpg
7b110704c2891bdcb985908c143789e1
96d8e5bf896707f19f2f5de54b3608bac12255ab
F20101130_AABVOO nesbitt_a_Page_82.tif
bf9e11844bc5e1b0658ec9cb6b5af848
05ce3b199e9b0f77a748b52fbb0945eba413dd0d
F20101130_AABVYI nesbitt_a_Page_18.tif
b3fe0761a9fdda489e717315122fe1e8
18370216bd583b62d85164541f1f090951ba208a
45958 F20101130_AABWAJ nesbitt_a_Page_13.pro
0f02be4c979f407a953e84cc9e5282e7
1b286479535439a49e1d6bee7f7890a375be50b7
72610 F20101130_AABVTL nesbitt_a_Page_29.jpg
8b865b5fc491f881e9a7df67c3faee73
0a45fbd169af327b7345857b14cd55b6fedb9d1f
5137 F20101130_AABWFH nesbitt_a_Page_07thm.jpg
d3e5bf846a20253a71a130ef216c6ba9
636dce0c37ba27675459b6e3bd163eef67ede534
71562 F20101130_AABVOP nesbitt_a_Page_53.jpg
701c5b512f582b0f8fbfcaa2bb0c899d
444ea98324117f440b36a346ef8a4cc461cb283d
F20101130_AABVYJ nesbitt_a_Page_20.tif
5ca8bfea60c63ca5eba1ee47e858604e
4cbea3f594e53318b0a1716bcef5a135aa9c70cc
42453 F20101130_AABWAK nesbitt_a_Page_15.pro
42d6cbe05a7848c8680ce524bb86875c
3886085d9ddb2857b0438242d3d7d8a4ca75a918
65749 F20101130_AABVTM nesbitt_a_Page_30.jpg
81fc029640761f39897717ec7917cf41
cfc82274e6c7a60aa64eb94a1284abcbe0e2292d
17057 F20101130_AABWFI nesbitt_a_Page_08.QC.jpg
789a1c9bcb64606a70384f3e7d969347
7915c6eb604b54776801336fb5bf869a357b2479
85691 F20101130_AABVOQ nesbitt_a_Page_14.jp2
10fda8159dacc0e95699fa806a5d724e
f7de619a83913713d30105374790f2a1c5852c3d
F20101130_AABVYK nesbitt_a_Page_22.tif
714b1a25610e84e588d370a2df175830
2c3f79dc4186a6cb96ca3660ea6d819e08dff82a
50258 F20101130_AABWAL nesbitt_a_Page_16.pro
eea4b3745574da3b74800d72b4a91ab7
227d4fa0ba1346b2555aa68d9759d9dea341e514
73920 F20101130_AABVTN nesbitt_a_Page_32.jpg
f5e9658f4c8d6f4a450c5df8d50fe455
ee4b87be34faa94103a23edda33f05129052979d
4607 F20101130_AABWFJ nesbitt_a_Page_08thm.jpg
f5f736bc2c46c771a35a6aa2b1d9c014
e9e5b15024c36af51d19152c748455127d32222c
F20101130_AABVOR nesbitt_a_Page_70.tif
53450c31807a5c4d533b7d1fa892a683
93052c7d4f632654335c0b3aa711cafa0ebfdddf
F20101130_AABVYL nesbitt_a_Page_24.tif
470926ead4498ed70009faf86999be71
6223c13c6e8b1de2bcec2c40ae0b7f038ee81e5e
72416 F20101130_AABVTO nesbitt_a_Page_33.jpg
fa0ed2d7355faf5766b1fc5dbd94289e
d069cf5f931a4db1acc7fffde65cf63e6ecbfcbc
5301 F20101130_AABWFK nesbitt_a_Page_09.QC.jpg
9e7d8637bfc029556c429a03c3addb63
f096e656fc2d7d1b050c1611ef6d71a379c4127b
39955 F20101130_AABVOS nesbitt_a_Page_83.pro
ab30555f631e950d1269bd3e23a118a3
129559735dadf3536799cf9a5016e43b1af40796
F20101130_AABVYM nesbitt_a_Page_25.tif
4c83356acdc881be7f036b78419e2eb2
fe248111e26e1b6a886b3ffed0edc90a36c9f895
48900 F20101130_AABWAM nesbitt_a_Page_17.pro
445203258531dac3fe3d81a1c8d8c185
e852f995aaa7d9d4596750a504935e0d9a365309
11681 F20101130_AABVTP nesbitt_a_Page_34.jpg
d8f9b7a8a55eead2cbf5b704d04fc4a7
39ace12c4304bc8d5051061ddd138ab226da17af
1837 F20101130_AABWFL nesbitt_a_Page_09thm.jpg
2671d88ecabfbb8f3cbceb6365a96463
9ed7e5ae21e6aa75fe8877f9977d278c62ef672f
22472 F20101130_AABVOT nesbitt_a_Page_51.QC.jpg
cd0107f0dafdf280972608be17d95a97
b6f292cf66c3f068895c3360058f2bc206600f3e
F20101130_AABVYN nesbitt_a_Page_28.tif
1668703cb79358fc918d33224a7e6281
17e0c90606297f70433a7c8f26d65b9d92af86e7
48224 F20101130_AABWAN nesbitt_a_Page_18.pro
bb0a5a1ca5891bc041188fbe0c6de446
7c77549f38f962ab0cfa749d36e601ecc4262c8e
66450 F20101130_AABVTQ nesbitt_a_Page_35.jpg
74a0ac017942a850deb681d7dfa5b72e
8643713998c12ed171ec454f683817c4e5ff39fe
5125 F20101130_AABWFM nesbitt_a_Page_10.QC.jpg
cbde0701253d4fa893525c8ed855ff45
3fbd016a425cd0528db3cbe5b6eec6a8866beaca
5902 F20101130_AABVOU nesbitt_a_Page_42thm.jpg
dd2f62c11426eceb2a9ed2ec404aec69
8b1efb82ac83ddea3950fc1b8f5aca7b14d1cad3
F20101130_AABVYO nesbitt_a_Page_29.tif
ea0b7f210b1eec7c25ce11b07cabcbfa
5c077909326b6da869fb76d1b9469fc075992670
46497 F20101130_AABWAO nesbitt_a_Page_19.pro
421255bdcf9cada53ed7f25fe552aa92
0cf6ac2c1456d538aa36937fb29e308d18a804f2
62045 F20101130_AABVTR nesbitt_a_Page_37.jpg
d60729bc0b3105e0d66096a25aa70052
88b7a2ec7aabed09960293e6282509dd8dd04da0
1919 F20101130_AABWFN nesbitt_a_Page_10thm.jpg
90b2975514f866eeb16e4a65f8ed5b8b
d6b1c2eab0615e4680ff8223477e6443953eb518
19947 F20101130_AABVOV nesbitt_a_Page_62.QC.jpg
f4d6a6f24a1a936f5e1191c8762121a8
aa9760d8685f8017993d461469cf4e047441106d
F20101130_AABVYP nesbitt_a_Page_30.tif
3d183e26bb9448a33c5e5087b7dd4271
982ba55a33e86005fcfe2eb4160f9fbc34e4d972
55459 F20101130_AABWAP nesbitt_a_Page_22.pro
b85fd395df76091fd0dca3453ff6b152
c0739a19a3b366828bb5cfc815e51e1e73dcb70c
70414 F20101130_AABVTS nesbitt_a_Page_38.jpg
87355c42bd9ab7cc26dce27127bcb782
847e0779c16716cc3a8853b68f4884a0b0b21e2b
17399 F20101130_AABWFO nesbitt_a_Page_11.QC.jpg
d8a4358d34c11a72229de71a9e444db7
8c22fe1323d6e9fad80f63950e8acabb55d0787f
F20101130_AABVOW nesbitt_a_Page_46.tif
7a329d8fb52d276b071025482a6c2ecd
cd1740824e2191cb2653d7be9cbd9fd500cad932
F20101130_AABVYQ nesbitt_a_Page_31.tif
37b0502ac66e14d8de0d9f664dd3c7a6
09cc0686abd7b064c9dd38bc55a888e79e878958
45584 F20101130_AABWAQ nesbitt_a_Page_23.pro
299797e78c1ebe1bc5487600e730bfb6
65a347b165628fd8ca6d81df6dbfaf00bc723418
82139 F20101130_AABVTT nesbitt_a_Page_39.jpg
ffbbd325057970550f2b28f04605cbb1
4e16ed0e5403abc3f5ddf98771a0a1e0d45b7951
4916 F20101130_AABWFP nesbitt_a_Page_11thm.jpg
49905403fdc197a3d949daf8e7963791
708a51eeae1193fb20ceb7efcf12d7286c7bb009
3944 F20101130_AABVOX nesbitt_a_Page_34.QC.jpg
3ed78e5af6ccb21e528124afcae9f8b3
eae62778b2682e498e4ac4093592c3551c3711f9
F20101130_AABVYR nesbitt_a_Page_33.tif
cf79b083812a358dd2774dbfcf7c4827
40ed53555b0caa7e1ebc515de7d6154a8ea2525d
51700 F20101130_AABWAR nesbitt_a_Page_24.pro
9dda6c80b36527c02f6f6c29097ace68
5a46d2cf8bc719b6c499a8e1ffc0adfb17db4e32
68919 F20101130_AABVTU nesbitt_a_Page_40.jpg
e59c9c7ae67bf36f87a2e58469bf8777
95ae5cc01279170350aff9f718768b1f0cee9208
4515 F20101130_AABWFQ nesbitt_a_Page_12thm.jpg
170118f5bbc671c8bc21caa66b49cc1b
e68bfe9397aa07b9a5c8b5f7b021c1d03985f420
84260 F20101130_AABVOY nesbitt_a_Page_45.jp2
0a0ad4527570504e4f90542efffd5aa0
d51427bfca080958b650ffa6604e5f421d8218d2
F20101130_AABVYS nesbitt_a_Page_34.tif
3845f7de123c779e7cb37bab901d2972
fa975b238b789afc0ccb033344f2e542770849e9
51230 F20101130_AABWAS nesbitt_a_Page_25.pro
a133bea7b13f6e7c2f8c8c523ae3f084
cbc458feca840170944d2f8d30dade92fdffe628
75095 F20101130_AABVTV nesbitt_a_Page_41.jpg
103750a5b77db4c0f67373ab12cec2ae
df1fba38a9dcb0fcf525c6deaa7d6884fd1110ad
6398 F20101130_AABVOZ nesbitt_a_Page_05.QC.jpg
c68ed9ad680e2e62bf3098b129abcba6
0cbd240f612c4a072ca35065b794ea147a92fb9b
F20101130_AABVYT nesbitt_a_Page_37.tif
cb7fc9efb160de7f6f845d8d3236768a
685cae3c0f2738ac951b219ae4f6ea05eb692d08
47702 F20101130_AABWAT nesbitt_a_Page_26.pro
65881aaeb5c6fde7708b310a7076b13c
192d3eaaf2ec7bdbfaf0d12dc133678dda1accd0
57714 F20101130_AABVTW nesbitt_a_Page_42.jpg
1a47fa960cc14b33c1437fe037e6e0f4
e57031a0b5e7251546a7c49ea6f1bf3307bf95bd
21430 F20101130_AABWFR nesbitt_a_Page_13.QC.jpg
bcf54782aae824dc3292c7fa53b1436b
aa13b80606af1344e5e8487b89bf9f7d827c4c6a
F20101130_AABVYU nesbitt_a_Page_38.tif
e080303b2586ebd34fa9df82951d5bcd
8ed95a1d53669d90b8d87f288d4a4e906e2eb298
52585 F20101130_AABWAU nesbitt_a_Page_27.pro
0176d87f4d1d4605906a6d6726597162
b0f61d55cfa48c0d2658a86858a6a52d067e71b9
72089 F20101130_AABVTX nesbitt_a_Page_43.jpg
c58cd0d5024eacc7a4b49a159a40a343
872ef603e95240841ee994f21033a336fa7c62f0
6012 F20101130_AABWFS nesbitt_a_Page_13thm.jpg
c5940ee3ea1509ebddd54c38c1b00efc
9cc75aa5c35b284b384caab100a4b6fba56ada24
F20101130_AABVYV nesbitt_a_Page_40.tif
01333c57deba452a2fffe0567a415fff
500817bccecd5c2ddd339ebf1338aedfbff73ce3
53514 F20101130_AABWAV nesbitt_a_Page_28.pro
632f053253591263520446d7e3f1c725
80c0e39330b033756b8510316fc01e4ed0a46a62
60145 F20101130_AABVTY nesbitt_a_Page_44.jpg
ffd429ec29b6ea54ac4156ad923aad73
e22cc6818b20f959f848034813b000a4aeb72530
5245 F20101130_AABWFT nesbitt_a_Page_14thm.jpg
864310b9b3dd2d71fd3bd5765f1ccefe
199919af1095775f4f9270b9ca67393e1f139dbc
F20101130_AABVYW nesbitt_a_Page_41.tif
25d372b6243457ead427a07fbacda859
ea5a7a48ca7ab9c6fd6b1217574cb299e568f6e1
54839 F20101130_AABWAW nesbitt_a_Page_29.pro
8d664d8141f4201a40219f68e2e75de6
db18899b7f26b7df493bed7651d9be26ef743d4b
22985 F20101130_AABVRA nesbitt_a_Page_58.QC.jpg
d81aef98fbf260b412158ecc0cf30111
1308a83ca139fa3d887406726894cd22f5f95806
67460 F20101130_AABVTZ nesbitt_a_Page_46.jpg
25b0fffe1f8628e35b532080235083e4
0bffc733d55f4dda037f6e254de6b51222bd07ef
20334 F20101130_AABWFU nesbitt_a_Page_15.QC.jpg
53a965d86d169cc93cdc1411485b9f21
7cfd33eb8cb73a0cae63aa8dd6ec09608a02b50d
F20101130_AABVYX nesbitt_a_Page_42.tif
b8ca25d33ea0806f0558601cf9ef36ad
709c43b3f36fffdd7a9198d144ef748ad08ab878
45208 F20101130_AABWAX nesbitt_a_Page_30.pro
5f7155ec397c42be249121563c64a026
cd70d586581310b11cabb1a28f4cde95b258096f
20394 F20101130_AABVRB nesbitt_a_Page_82.jp2
62d7543e7cd9628e72a019d89402573a
b92b96bb19aafa2f4c03e1387fabdd71d1eec10a
5914 F20101130_AABWFV nesbitt_a_Page_15thm.jpg
eabd2f9a3a9996bc08b2ba4fae503835
cf343e41462417a758be61224d8dd29b6afcc8b7
F20101130_AABVYY nesbitt_a_Page_44.tif
6623f4dd9685decc76af609f0f19ae97
652c7e17da2ab9815881c41f37288d4b8b4dcb3e
52645 F20101130_AABWAY nesbitt_a_Page_31.pro
4e88556708316d37b22673047a40dea7
2e78328281d569fd111dbe53e0d0294d89ba2826
77947 F20101130_AABVRC nesbitt_a_Page_36.jp2
937925bbfe6c64d1065eafaf4d60de5e
3c4d913b7c71b0299ef4f9da508e8770ffeb3f74
112439 F20101130_AABVWA nesbitt_a_Page_29.jp2
7dceddb495f04523c39eccfd2a7f027a
394060e050aac3c0a52adefe7bfdb999b08378b9
6370 F20101130_AABWFW nesbitt_a_Page_16thm.jpg
3734ca287361b3eb64e9f000a8aa9964
da967685de5ab2c2f9c5ad1dd75c3b9857ca24c7
F20101130_AABVYZ nesbitt_a_Page_45.tif
4f848e46ad5ca0503f0275e9f538d1ec
e5c8932d1139e985f7f33e78ad6660848a697ce2
52745 F20101130_AABWAZ nesbitt_a_Page_32.pro
4a8dcddd5ffda466c3fde50386b2cc5b
94b041d530bd30c2b3ac1ab1570bf675a17a1cee
112820 F20101130_AABVRD nesbitt_a_Page_39.jp2
3535857978d60597b077b49437216aac
987839bdd0c8f275cb45c8ebb0a07be7dfe3cd39
97936 F20101130_AABVWB nesbitt_a_Page_30.jp2
ebe6bd62b66e38eb7153f55bc088b08f
7a7fa49ba32e5cee231358ba3d981d2978066846
23061 F20101130_AABWFX nesbitt_a_Page_17.QC.jpg
8b71a0340799fe1d98c00ad60a2ab151
fe663a5235fbba8cc7320f72a2f1180d6663cf3b
70165 F20101130_AABVRE nesbitt_a_Page_69.jpg
34a1749d69c49661ff72b1b6b8d8f229
928d8edf07de771e3e6db2b4ec5bce40af939373
111910 F20101130_AABVWC nesbitt_a_Page_31.jp2
5937614c292fba8a14c2c0f66e8d05d5
c499ac538419aa199aa148d0cc8bceff522922e8
6414 F20101130_AABWFY nesbitt_a_Page_17thm.jpg
5594d2cae9c838cf8f83a44c65fd172f
6de0faa09001b19ed0891ab142edad5fcafe0418
2018 F20101130_AABWDA nesbitt_a_Page_20.txt
9e875b8799bfbab69a7eb041b237d727
c61e3a0ebb80094f60a6e9e6d2b21bd89815f244
23332 F20101130_AABVRF nesbitt_a_Page_59.QC.jpg
fb63ccd6cded7b4af8e2f7a19afd8864
a23bdb57f92d646703091522921aed76733b40e3
22084 F20101130_AABWFZ nesbitt_a_Page_18.QC.jpg
cd800a333184eebe67fcfd8843300017
2def1b84a7379219b4a3e8fe2190ee0da7e332c0
1709 F20101130_AABWDB nesbitt_a_Page_21.txt
93f3e5e90031abc7e3838ff1b9aa9ca1
9a23370f310418f1aa170ed7cb0dc47ab9feb80b
F20101130_AABVRG nesbitt_a_Page_01.tif
9ade9383de69839b058ae02c1ed10f0b
a50c6b6107f8ebb3f01e8f44dc05052e0bd7126d
113417 F20101130_AABVWD nesbitt_a_Page_32.jp2
02eff1ec5b7036429947f81f0faf9510
b4795de9d231adf7cae4d2770dcf32e378b02110
2201 F20101130_AABWDC nesbitt_a_Page_22.txt
b7baab8e781a30a4dafeeb7a87ba6103
0e17e8f8e3bd5a6d11f9914c643c97b39b7dcfe4
5969 F20101130_AABVRH nesbitt_a_Page_26thm.jpg
1a9764237e5520ab679a91762c069a25
45fe642a68220aba62e01bfe79213c71678138ea
109597 F20101130_AABVWE nesbitt_a_Page_33.jp2
87b7487b7c23aa1f10b557871f04f99e
d8c7a0cda057644528fe5c6ce8ef95e58f95deb9
22200 F20101130_AABWIA nesbitt_a_Page_56.QC.jpg
fc106e84b8a108eecba53c68a9e1ab0d
3170582849688d9863982d476c868137480961ee
1854 F20101130_AABWDD nesbitt_a_Page_23.txt
3ec496b5406dd0c8241c2bcd6c7ee0a0
0ad6a88dcd0917d7692ba48545b2f9f882881fd9
F20101130_AABVRI nesbitt_a_Page_43.tif
aa5fcc98510d31b98ac26b5627377004
3ebb5c4843fdae638947730e23aad29a77d4c883
9545 F20101130_AABVWF nesbitt_a_Page_34.jp2
44ac55af9967db2d1c57eda5fdfc4c6a
0cd8cb63a01a6515c95571c4647db7df3e3f75b8
6180 F20101130_AABWIB nesbitt_a_Page_56thm.jpg
97625f153e788a3e66dfc0c85379f742
01dd44ebd3dffe5ea07235cd92d8028c6ddbd292
2034 F20101130_AABWDE nesbitt_a_Page_24.txt
8cd5fcb78141dfdd7914fa9e2106c6dd
78d203fcc59a78a5a4ff60e778f4991cd849d04d
106737 F20101130_AABVRJ nesbitt_a_Page_25.jp2
1caecaf77cc9322b41394596c1c20423
00c395a7d3561f45b170dbe6de4eeb11d3ffcd22
100367 F20101130_AABVWG nesbitt_a_Page_35.jp2
1546237270e25d84b60a33ade32ee304
781eef1f33c67d1511fc9be4e7d95dd77aaecd1f
6625 F20101130_AABWIC nesbitt_a_Page_57thm.jpg
e8dd42b90ce5adda100b688665e045a4
06453bf11998f9c4ae4a1fd10c9177124c7e25ed
F20101130_AABWDF nesbitt_a_Page_25.txt
95bb308b58b1c6b70c3b73b39706ec8d
9b9140d83d953cc9eb58cb3596cb52bae2ae696b
6438 F20101130_AABVRK nesbitt_a_Page_71thm.jpg
9c0a8ae511ef0ade0363ecbdac1e6eef
9db02cd8db74513ea62a196dfeae29e77522ab9c
92784 F20101130_AABVWH nesbitt_a_Page_37.jp2
5925793cb5bc09761949bb5fdcefb51a
7f578bbc690a1fe5618e52582105c68a7aad8256
6682 F20101130_AABWID nesbitt_a_Page_58thm.jpg
ecc523b30d3f91e9c0518e80b924cf1d
371b6043f4e2d6d70d45fe703683cb235463d8c2
1890 F20101130_AABWDG nesbitt_a_Page_26.txt
46b423f7be294f9b4c61272f63430aef
60982004454437de80f2f08970c25d7616a28284
F20101130_AABVRL nesbitt_a_Page_78.tif
8ab73f60137b1e37506d9606163e9685
c7fe45436ee19cad3c63ea8711e4e0c968ddfffa
103663 F20101130_AABVWI nesbitt_a_Page_38.jp2
f2c0a2f4f14944b8d1a6e2cab5190c09
2908a9739e81dc297e7682d1ceda15c630c0a79e
6470 F20101130_AABWIE nesbitt_a_Page_59thm.jpg
724b2b00909bb6b6361ea4aec960925e
1abab72b6307078038d6cde2c0f0d775c00b9f24
2070 F20101130_AABWDH nesbitt_a_Page_27.txt
5683f3397ac86327e96b5027c06316c8
aef244a54eff2007042a22204052ecff09ad0277
2809 F20101130_AABVRM nesbitt_a_Page_40.txt
739407dc2974bdf96358ea1837f88db1
acf0323216aa27116471b1c51018974e930fa7dd
94357 F20101130_AABVWJ nesbitt_a_Page_40.jp2
f90925d9dceed675627d868e78da43e5
59a915c02af5c0528afedbaa27ce3b9aec42efb1
23270 F20101130_AABWIF nesbitt_a_Page_61.QC.jpg
3f655932dd226a9eade85a52526a5520
227b9fa3351db68e49a44195df023432925a91b6
2099 F20101130_AABWDI nesbitt_a_Page_28.txt
f73820b8f790a330f9650cedb37f00cd
79ff15f0a6269d467e0055ec0d628ec9e8cf8650
F20101130_AABVRN nesbitt_a_Page_60.tif
f8535f9df0f0c790490ab6afd1045395
f10f8e4975050c64ac38a822d86fb158631e6989
98378 F20101130_AABVWK nesbitt_a_Page_41.jp2
c45696083d1ba614dfed7946ffff5d55
1baf672378b05cda4cbfae295e7ae95ddefe2684
6703 F20101130_AABWIG nesbitt_a_Page_61thm.jpg
180fda018cead20ab8af6150a1b87f05
2c7eb0702cab6cfc91fae7e61bf3ac120de0243d
2154 F20101130_AABWDJ nesbitt_a_Page_29.txt
451aa348907de501442d4217e014761e
6233cdee018664369efe1b215046565e731439ff
23796 F20101130_AABVRO nesbitt_a_Page_33.QC.jpg
9638d43a0b3056bbba9596149477cbd1
b5a6fc91e9db5181ce6fce6a79834b7377e45359
90342 F20101130_AABVWL nesbitt_a_Page_42.jp2
eec6758f5ff4768f0f7fd7fe530f8af9
6748e52422b0350164802bcce52ef5da2516a109
5588 F20101130_AABWIH nesbitt_a_Page_62thm.jpg
6b161970f9d366e2ceb1b139361d2fb4
09e9c7b1af58525cdab479c2d5c727e482b5c01d
F20101130_AABWDK nesbitt_a_Page_31.txt
aa2af5ce9a703406e6ab0ca19dcf9cf6
be8591124743986545496436d67a809b644cda8f
63320 F20101130_AABVRP nesbitt_a_Page_08.pro
2ca871785070a42b48ed2d37a09a2341
097ea2562be0e45c8ceb7ebb746c2cb7c2e3c9c9
117307 F20101130_AABVWM nesbitt_a_Page_43.jp2
719878766bcb3dc97d8e7d909e85b613
53f0d1189c6da058951edea72d7d0935a845a0b3
20415 F20101130_AABWII nesbitt_a_Page_63.QC.jpg
f27ef853ef6168f19933567d410b5d8f
eb19b8ea8f163f6c651a8a0853ba70c0f0b87ec9
160 F20101130_AABWDL nesbitt_a_Page_34.txt
2bf52260bc71d48bbb32abf78c9e483c
f0c672ec9c5fd08ad445d894f9d4445f3ba11fdd
52485 F20101130_AABVRQ nesbitt_a_Page_36.jpg
935c118353c7d3610fab31daf85599df
8ad650bf661ede07ba19694959a3c4828e9b1e5f
93099 F20101130_AABVWN nesbitt_a_Page_44.jp2
1f394656c81c034f3a29a5545d35d1c7
7ed180d57a11e6a2a5bcd397e876d5703d624db2
5770 F20101130_AABWIJ nesbitt_a_Page_63thm.jpg
a70155d7cf8cdc90dc21c891a8de742a
0de913d82768caccc2371b6b30042cff50a5fe07
1915 F20101130_AABWDM nesbitt_a_Page_35.txt
6c5fa3d8e990e44954ccfc38cc23ae5c
dc47bc2e00d2141cfa5ba254c70c1be5e46c4f70
22945 F20101130_AABVRR nesbitt_a_Page_57.QC.jpg
90a1ed0597208e2827566582258eada4
17215422cc990810cf49d008c69253a96d05204e
97869 F20101130_AABVWO nesbitt_a_Page_46.jp2
4768f786e5dbd7ee78fa6519acdaa973
a8e525325cbdcf4a47c14ee98022cb6ae5e15ddb
20170 F20101130_AABWIK nesbitt_a_Page_64.QC.jpg
fdb858180c8dcf0f5b4f3a716f7b8f0d
bbd49c24318da651057fbeeecb8ed0bb5319795a
1407 F20101130_AABWDN nesbitt_a_Page_36.txt
2f8964e0e69fdc664923a9b854ad180a
9545e160fef5da7d0686b3dec0cddd691331114e
5908 F20101130_AABVRS nesbitt_a_Page_04thm.jpg
d149b9ac7b755fe126c19e3260c56eb6
3affd20ed0eda0da0ca6a0421d8bac1242aea16e
105365 F20101130_AABVWP nesbitt_a_Page_47.jp2
92bb8a8b6752b6237e29f82252d4aa0c
20fa0ea2b66d53eaab2919954ba1f374fffcf52b
5898 F20101130_AABWIL nesbitt_a_Page_64thm.jpg
cad04e8b705f4abc400e35266f3697f7
dcd614d19154e420a1d7fd648f81741024a16d32
1805 F20101130_AABWDO nesbitt_a_Page_37.txt
cc9d92c7192476adbdc571994e1f0db6
0ad9d16f0eaa88f95c2a670e72eca32fb30af1a9
F20101130_AABVRT nesbitt_a_Page_23.tif
0321368d50f2732d88a60e0ed345a725
e1803e622f91f3669e6119fa39e70538ac8f0f2f
48064 F20101130_AABVWQ nesbitt_a_Page_48.jp2
54cad1784f10280b9b32951794dac8b2
d05f033ef9bf652a69fbc5979f44bd95fdb75299
23617 F20101130_AABWIM nesbitt_a_Page_65.QC.jpg
d169c4914214455cb9a4d1af02619c99
a38d79b842eb2ec43f71c7fd3522af5c99616724
3744 F20101130_AABVRU nesbitt_a_Page_07.txt
b30ea4231f4d82ee03f0a4e8bd52f460
776253806d603dbf19727c7e03e08f3cd173352e
90969 F20101130_AABVWR nesbitt_a_Page_49.jp2
a03ebac4ec6f9e02afd61ea20391f505
fffaeeebbc790c3fd1d608fb1d5ce4161d6b7545
6639 F20101130_AABWIN nesbitt_a_Page_65thm.jpg
983a748905dbdee4790b0158693655e2
2ac512075632d3b30c8b90b3f77337fd8dfbbeaa
2123 F20101130_AABWDP nesbitt_a_Page_38.txt
e2be0e6c34b5b0a111ab9fb8fef42c53
8f687af99fee4a86e27e7dde65a64635f10c5d39
1816 F20101130_AABVRV nesbitt_a_Page_52.txt
7e5095af1efb8df3ee1b9cd2ab5b7422
e8a912198b6284311050a359a4a902470ecb280f
107112 F20101130_AABVWS nesbitt_a_Page_50.jp2
ac64954485022dd83ec63041c7f525a7
e3275726b3c4a69707031057971f0ba3af14e53c
4054 F20101130_AABWIO nesbitt_a_Page_66.QC.jpg
18a2a8ba89e33a7ba9a5a9d676a5676a
cb711fa0ce4dd14e1fe7db713d4a043d6ded8b64
3471 F20101130_AABWDQ nesbitt_a_Page_39.txt
a990b0b599dc12fa3d4c28d9e8e396a5
b20d22bcd96957008d07cc767d2fb3185f2bffb5
104174 F20101130_AABVWT nesbitt_a_Page_51.jp2
e6d891f4460c280cb8935201293e0f43
2b47ab7219b7a8c23b33ce06e06da1bc93b68d54
109084 F20101130_AABVRW nesbitt_a_Page_20.jp2
368d8292e18ad7454845ac35389fc31d
5d133b587cd5f85a35c23d1d0a3fd088691c9921
1554 F20101130_AABWIP nesbitt_a_Page_66thm.jpg
67d51c5f5844ddc6823e37a50e1baa98
7d34a86ba5c9bd2918bc26587e4868c743a94b55
2098 F20101130_AABWDR nesbitt_a_Page_41.txt
3ca4bc414cf30c5c6e25fbd33beb7f82
8abf5d93e2bdf38c0232dbe025f6c82f04966a45
40149 F20101130_AABVMZ nesbitt_a_Page_42.pro
eb4a00ba66e3cd24183705959efb09ef
05794dc4ac9ed1d812e4294578fd90166be0cdc8
103886 F20101130_AABVWU nesbitt_a_Page_52.jp2
c62210d5f64891ab71fbfedb2ab7cf8a
d6291a78767a37e90a73cc85ab32dea2170bb28f
8553 F20101130_AABVRX nesbitt_a_Page_01.QC.jpg
a30358a615fc7650f05ba53a3bb51f17
8fe140d4bf11f44a2535c62dde783a1e8d3aad34
19705 F20101130_AABWIQ nesbitt_a_Page_67.QC.jpg
efdf1d5e5197badca97e7d1ccb064171
235e373f59132b3511db7aa3b9e04f9dd7a345e5
1601 F20101130_AABWDS nesbitt_a_Page_42.txt
78977de004a8c4fc8aedc2085bea4843
628cf5a0645e55cc09ab92bde2f46f849fda48fd
117084 F20101130_AABVWV nesbitt_a_Page_54.jp2
be6c3fb2cd94e4fe67b88c969698a044
8c9a3a29ee01b41bbdb00c65f630e18997530578
25059 F20101130_AABVRY nesbitt_a_Page_54.QC.jpg
bf21592c32c5ebe58e035fa6515d51d1
e0126bcffc0aafb739a4ac38566bc5a9dfa96f88
5611 F20101130_AABWIR nesbitt_a_Page_67thm.jpg
f2f97bd58bfeb160cf30cde1ded6b459
1aa857d9153ac08653a1aef681621c9a1b7db390
1767 F20101130_AABWDT nesbitt_a_Page_44.txt
9b1585e9f91d20c63c74416f577badc1
d571fcc3bbfa1a963d90ccd9622b03ac9f8597f3
102321 F20101130_AABVWW nesbitt_a_Page_56.jp2
361127308c78c237bcefd1b836ef0d74
1d23ab070e50182876f14a986ffc73d411348173
42858 F20101130_AABVPA nesbitt_a_Page_21.pro
e9bf6c1b5bed9d1280c336b2401ca64c
40cebe3b5324dd9616a0265047b59b02d5a5179f
51358 F20101130_AABVRZ nesbitt_a_Page_33.pro
8416d0dd42b5992ee0eb63c1b8128fa2
a03c6930207cfe673ce3093ead334dca0fe79baa
23323 F20101130_AABWIS nesbitt_a_Page_68.QC.jpg
92b8e6da217a9fdf44b3ec94ceb4eca8
1f7f05addabac5a153d3b10ccf3e786575d72f62
1842 F20101130_AABWDU nesbitt_a_Page_45.txt
4797390ee58ca637d3750d63fc9478c0
2fc2165a63f10c408f6b87690d0992896423e710
106375 F20101130_AABVWX nesbitt_a_Page_58.jp2
494d4edb91c882cfde7ce53f7204e282
83df8fec69e52373c58b8fe3291b1bc30f42413d
6323 F20101130_AABVPB nesbitt_a_Page_19thm.jpg
43c89ca66ba768c292aa431504cfb594
e78ef7a943b9743751999a2807816485d30809f8
6481 F20101130_AABWIT nesbitt_a_Page_68thm.jpg
cc27403cfbca16f72e8b923c47a2eedd
98f433fb8412d5221fc0951f88dd9d12c6516dce
1994 F20101130_AABWDV nesbitt_a_Page_46.txt
ed3fe9dcc61b00c781681d4e524c12bc
63536dd5d2bf988a16d60fe27d51e5e31827767b
68560 F20101130_AABVUA nesbitt_a_Page_47.jpg
db1fe90455bbad3f7de49d8bd31b1df4
101ee33dacdebb5af7f250efa70ec4510cd984db
108672 F20101130_AABVWY nesbitt_a_Page_60.jp2
d1b62b5cec88169cbff33b23ba8b313a
08ffd592d9e7ae7cf1674795204c12cc2828cb02
20771 F20101130_AABVPC nesbitt_a_Page_26.QC.jpg
14c956625ccc3f1bee8bc3cedf12220e
00b9252d2964bb21b913554b09f9a3614b483496
1905 F20101130_AABWDW nesbitt_a_Page_47.txt
97a6b36a9600aec103de0006d4b65433
c83d6a7db492b89fb22f5a497d299f7df4d165ac
108282 F20101130_AABVWZ nesbitt_a_Page_61.jp2
2350893d6579d7352f1182633a989ce6
85b6b1b8c124bfcf27ad17018b353292b3cfac9d
23486 F20101130_AABVPD nesbitt_a_Page_20.QC.jpg
1ae3a2d4c653297c5da7dfe853fa8ad4
9dbd225447b539f9cde86658d9a4e128fc553b37
22993 F20101130_AABWIU nesbitt_a_Page_69.QC.jpg
85585a231b5467622b9e4e178f5205fb
a313464d4754cb94e5608e1fd269bb46b7a639e2
838 F20101130_AABWDX nesbitt_a_Page_48.txt
e4be2f9f7c5d7e90ec8366dbce8b3d54
c10d280db5b48b546e3ffe037d99a8913802454f
34662 F20101130_AABVUB nesbitt_a_Page_48.jpg
667728afbde0ccd4ef86f1f7bbdfa957
ed3efaff5dff77cc778e100fe5ececf2b512fb53
22683 F20101130_AABVPE nesbitt_a_Page_25.QC.jpg
d5411f4d24eb8dae8e55e59232eb53bd
14a1cb5d95ff5bbc346510fea920fc89a9819fa1
6385 F20101130_AABWIV nesbitt_a_Page_69thm.jpg
d33ebd52c4b11257d25357cc87009249
111541e581c026ace82572fa4d7391bb28515ab1
1744 F20101130_AABWDY nesbitt_a_Page_49.txt
1617dc643e6c82718f047d0723034488
5cf7d5edc25369361606e9c535d0e97fac773ca2
60465 F20101130_AABVUC nesbitt_a_Page_49.jpg
0ce7eff0e95051c8f727c8cba26e4563
8f376ccf8f4ebac513adb9232f54983bd4f7574e
63913 F20101130_AABVPF nesbitt_a_Page_62.jpg
14d2494aad609410d2a3c64e9f784dc2
15e695f8cea442567ce487e87a4a9977e6dc93ba
F20101130_AABVZA nesbitt_a_Page_48.tif
499c3be183de7dca479c86f3a34517d0
72fa4453750baf68bb36083942319e800ac03e87
2862 F20101130_AABWBA nesbitt_a_Page_34.pro
502c47fdc0f7b97a047908c3b5c7f515
93d86ddc4f3d5cf6db15fcc78226b6a8935658f8
22655 F20101130_AABWIW nesbitt_a_Page_70.QC.jpg
bf0e71bf3262dee4b0786c5165e611b8
5386e016377bcf30c3ae6c58494c3d38a131a643
1896 F20101130_AABWDZ nesbitt_a_Page_50.txt
e90b6a7e34a9680af9af41bc10083c3e
99ee29aaf74a4745a958cca2ba23ac952ecb0be2
68559 F20101130_AABVUD nesbitt_a_Page_50.jpg
931e4ec16358bac9c1c478e12189a076
a4148fe346b22fe8e8eff6fdbd551baeefddf6b6
3682 F20101130_AABVPG nesbitt_a_Page_84thm.jpg
42020131ae4655a08921bf6352a7a188
04f6f2031c6f90bc6bd2117b9fe67c36a155d38a
F20101130_AABVZB nesbitt_a_Page_49.tif
fd86d65b49f26c69b3f8e73428d6bb16
b298596dd7bdf19f7624d6dcd68fca301d4a2843
46678 F20101130_AABWBB nesbitt_a_Page_35.pro
f84ff06fc8f338901082f507e5c69ba7
91316c581f82d96600bceea432f82010d406c32d
17294 F20101130_AABWIX nesbitt_a_Page_72.QC.jpg
d0e4abe3d3bbd2af5c98b64e4aff760a
a32c3df57414149ad83b1c90bb9db2f1ea226d7d
69019 F20101130_AABVUE nesbitt_a_Page_51.jpg
4a82c5e477c0f38340c50026b5680a37
57cfee0572491b4beb6147bf40ed3995ad9e27dc
F20101130_AABVPH nesbitt_a_Page_27.tif
8451839219dae7dd1dd2586c654c6e10
bfea0010f43b9014c2ad4eac2f14ffd5e351fd00
F20101130_AABVZC nesbitt_a_Page_51.tif
757209a3b3073298e3a904af0bebe7ee
1e9c565bbc29f028115bc23a38202f067c65ff20
35108 F20101130_AABWBC nesbitt_a_Page_36.pro
5a58629f0d882f46a713ac0ed134a9af
fbc0ab751900173bc223d8629a124f69abb45a17
4909 F20101130_AABWIY nesbitt_a_Page_72thm.jpg
c6369425604d3082248900ca55fd48f2
8cf8c4609040652e1101d5ffddcab5dbd6774882
6511 F20101130_AABWGA nesbitt_a_Page_20thm.jpg
7ca620aa9ff935a978e6e7042ad8b2e9
0d6e138ec3da6babc3622694ff476077a055f231
67295 F20101130_AABVUF nesbitt_a_Page_52.jpg
c420de7d43a7747252c8fda84128df7d
4714b76101783602065178fe33445dd9dfc99246
F20101130_AABVPI nesbitt_a_Page_26.tif
718388335adec685ef13ce439067ac19
1e3cbd5c818e1eed7aa55f488eb7ebd71cdd021d
F20101130_AABVZD nesbitt_a_Page_52.tif
37afd7834ec1cae5074f6b835d5db7a8
81ce6bc804127c503d810f34f1383e0aa6fb4d48
42958 F20101130_AABWBD nesbitt_a_Page_37.pro
7854254733d690d88259cce5147771f9
84a67be656cd2f7ba265bce2b21d6c6522d8a310
23779 F20101130_AABWIZ nesbitt_a_Page_73.QC.jpg
9d07708e8dc00d4abf498b195b33babc
818d8806586d5e5f171a2beb666fd04b489550da
19969 F20101130_AABWGB nesbitt_a_Page_21.QC.jpg
cd574dad3bc5d36bafe7e999b683e856
9846077130bb2ca84c257eb4da96294f2e11a5f3
77139 F20101130_AABVUG nesbitt_a_Page_54.jpg
4f7d74338d417c313d68c1a090470539
ee086daddfacde1bb3b622f026d224b0502cfaed
120 F20101130_AABVPJ nesbitt_a_Page_76.txt
952635c2f28df9fbdb5f4b268c109ad5
709d7bbea65226f29f86b491ce88ae8a72fb76db
F20101130_AABVZE nesbitt_a_Page_53.tif
f8e3a4f84ad95007430094547c83b429
986a52a4b96bc03217f9047af700e4536a6915a2
55760 F20101130_AABWBE nesbitt_a_Page_39.pro
4a105048c62d2ae0c16f3d37cb64ba45
861427d172d909df0392015cb18c750c46860959
5852 F20101130_AABWGC nesbitt_a_Page_21thm.jpg
9d24f485dbf2e321a2ee1af148055a4b
f2a1b112f38c3013aab8937e701d12327bf684b4
66821 F20101130_AABVUH nesbitt_a_Page_56.jpg
53091c80c0fff0ff8baf87dacdb78083
a61b13928cf2ce83cb0f9a2e9d34dc74551c2404
62549 F20101130_AABVPK nesbitt_a_Page_04.jpg
5867ffa57d934f403746c92165abe371
f6490cd2318105349cb07cfb64a1cbbc37eeae99
F20101130_AABVZF nesbitt_a_Page_54.tif
d697fd5db4b543a4f75b40b6903ee6f0
21c5fcf7334a1dc496bbaeee9372783d896cdfe8
46433 F20101130_AABWBF nesbitt_a_Page_40.pro
88e9b77acd5c887bb1c013f95717d9bf
5fc9e443e02355e3ab65d0a2262ae77f338087ab
22265 F20101130_AABWGD nesbitt_a_Page_22.QC.jpg
110d5c5fd07cbe26028c8cf47036e49e
4e8fad14953203f4282560b4a1f9238838a16ef6
F20101130_AABVUI nesbitt_a_Page_57.jpg
737b2178733152b9d074405bc08ffa84
297837b9f1eef70319056e526e78e3f8301232a3
105316 F20101130_AABVPL nesbitt_a_Page_74.jp2
38bcbc325eb72273055b65764ed0ec97
4ca40e9eba3ce4f81d07ca6494b480e04581189c
46596 F20101130_AABWBG nesbitt_a_Page_41.pro
603cc240b53efd951e04e30e40e01158
ddf5cb348894bb6a20e701c7c55a1a25fac4da7e
21099 F20101130_AABWGE nesbitt_a_Page_23.QC.jpg
8758290549d5b7683c8ed9f3362164f7
a543e16cc87ffc176ffe0b08a4858545ad7eb6cd
69667 F20101130_AABVUJ nesbitt_a_Page_58.jpg
31301c6af88a5b1ff751d8c1de1c37f1
20b6b5f48953756e310b2b9ba16387f1e16ccfbc
110733 F20101130_AABVPM nesbitt_a_Page_59.jp2
68701f35b4b062b45b7bd20b989205b4
42412b12eaa6399e3938afba0a26dd7c7755e7f4
F20101130_AABVZG nesbitt_a_Page_55.tif
872e7e52346621b82e5571e738f50810
5efdf7d7a5bd8054b395c949148ebb649aeaf82a
62867 F20101130_AABWBH nesbitt_a_Page_43.pro
e1aaab740a9b555d1642e6a52607bef6
3231a567812831378fcfc514b9f5b934551bfbf8
6077 F20101130_AABWGF nesbitt_a_Page_23thm.jpg
dbc0f7bdb11480a4cdd74489ba6dc06b
79d159f3e6d54605528f0ec9c571f6f869262095
72166 F20101130_AABVUK nesbitt_a_Page_59.jpg
93b9e7821c190a3d136fe0465c62b932
882b72177157ffe8177199789d10076d71fcdcbe
111 F20101130_AABVPN nesbitt_a_Page_02.txt
8b72a67adee4cc75774eea900f126340
af2d0dd34957702fad3964c780d1c3c78f8f6203
F20101130_AABVZH nesbitt_a_Page_56.tif
c2f7312980726a2c0c3380f689f1503f
b761a2d74adc930ccad295891fdd1b55bed3ae41
43608 F20101130_AABWBI nesbitt_a_Page_44.pro
b1d666a02d034a589f1484efa1995a7e
81bb25a0a5c5f5f3b56e1d8f88ad4f02383c2687
23955 F20101130_AABWGG nesbitt_a_Page_24.QC.jpg
f47b0ca1705400f9848e5ac61e6b16e7
9e46ebe013064be1459d255a3b3c5e555cbcca9c
70586 F20101130_AABVUL nesbitt_a_Page_60.jpg
4afaf2b0a2ea7b0365272eb21216bf0f
6bc36f159024bc67bc66cf49af636b114778e839
55055 F20101130_AABVPO nesbitt_a_Page_75.jpg
4032d269bf505312e2f62b365aba2c3b
2fff8ed9062fc78e52ba9aea53698180c27b3e58
F20101130_AABVZI nesbitt_a_Page_57.tif
ebefe8a6ce15bc491ae7ca40071c685e
128e01c04c9adc91315215836fe0bc6c1639ea51
42313 F20101130_AABWBJ nesbitt_a_Page_45.pro
b1df79a16530ea410ad1b3224a660017
72eca5b7a6b764a07b1d38af2e4927c248fe0719
6592 F20101130_AABWGH nesbitt_a_Page_24thm.jpg
d0a04f8433cd72ad8879d8efe3b7e301
c748052c23ca8b8d59857ccac4bfeac5155fbb80
70960 F20101130_AABVUM nesbitt_a_Page_61.jpg
7d8fb8ce3248f96f90ed9cb247bab776
121c740911043145453bdadcd68571083fdee724
47602 F20101130_AABVPP nesbitt_a_Page_64.pro
bf7fb35043338ef448f7aa048849688c
4711ee2266ed32efa5bcf7a00212abf036b5b6f6
F20101130_AABVZJ nesbitt_a_Page_58.tif
0285450f0da295de96b68f87d2923547
8840df02d824c0669f6e27a7d6aafdcc362dcac1
45631 F20101130_AABWBK nesbitt_a_Page_46.pro
4ea310a6213747d876fcb09510ff910a
57e30ee9649915db4ee921d2f30472c693ed7516
6506 F20101130_AABWGI nesbitt_a_Page_25thm.jpg
e491f870806c48c61de1e97079492346
f276b43ceee0e1a3f49eb5b75ed468fa5891cf7e
64253 F20101130_AABVUN nesbitt_a_Page_63.jpg
e6f7e2c23451afe5b44003d43967130e
bef687c0b647fc921908dc80fd1823b5605277aa
F20101130_AABVPQ nesbitt_a_Page_30.txt
4607d63f83d1f9c5d082a84eaf9f6971
86c0321a045689b654db6375ec9e31f652765361
F20101130_AABVZK nesbitt_a_Page_59.tif
e359037d1f0edf5b801f684f229cce85
2783eaf84555c7f66dbbeee8193ad633eeed0dd9
48035 F20101130_AABWBL nesbitt_a_Page_47.pro
d791a3896d55f1281f6117ef89bda119
10714f2ea8f0239014010ec860e29080429d8391
F20101130_AABWGJ nesbitt_a_Page_27thm.jpg
81d1e59e713b8a9c4e88042bd9631abe
bad7ed6454ed0e97513309aa37a0144d0a80c40e
73578 F20101130_AABVUO nesbitt_a_Page_65.jpg
8f4500ae5f7a82c4d7257e2b421e51d6
8095b49a6ecca388b360a89c44794b6b6b7ee4f8
109933 F20101130_AABVPR nesbitt_a_Page_53.jp2
f5b2b3bd8855d38c112e25a0be9db134
bdd3dcccf2314639bf0780b442c8a0a1cda1acbf
F20101130_AABVZL nesbitt_a_Page_61.tif
0930065778a51f770f99ba7694120040
2c95872e1bc4e5627882a3b108aea5f641368565
20708 F20101130_AABWBM nesbitt_a_Page_48.pro
e203957aba5b7251739e4a10cfd669ce
a3222751003f3e1bc4a0ed631183d5bde1a5115b
24458 F20101130_AABWGK nesbitt_a_Page_28.QC.jpg
23c7ab9fed46c367104ad54b97acb473
ba495e9b7e759c3b240e8d392a4c5a80c24e6be9
11917 F20101130_AABVUP nesbitt_a_Page_66.jpg
a175813a4dd17a3c56b52d7ab239633a
4e216a8972150874eda9e57d15799195959c5d71
2480 F20101130_AABVPS nesbitt_a_Page_43.txt
61de8480303d10c8550e3dccfdebb9a4
0323cc0ba8c008a96956add60b6a8faceb319f5c
F20101130_AABVZM nesbitt_a_Page_62.tif
8b527ba46fc4a72886937091f35701d2
0d2181aae51646602be83872c8874dfe3b6b5e9e
6817 F20101130_AABWGL nesbitt_a_Page_28thm.jpg
73ca8409788fd1b0188e2072463ce4d1
a3cbae92784e5ea4df1ebb0ae314a2bdfd46e883
71734 F20101130_AABVUQ nesbitt_a_Page_68.jpg
087af2a059c079c89ed73ac932dfb7c1
80d1b12c6c8a94680753f00614e89012f85e4af3
21197 F20101130_AABVPT nesbitt_a_Page_30.QC.jpg
6532e571474ffcb96f84e287fde19e89
082b2fcffb7f65579776c8eca8a8ab613cc11dc4
F20101130_AABVZN nesbitt_a_Page_63.tif
ed85c031d91e1352a1b5f980d37cb43e
5eed132fed342e1aaf3f42c80047c9a48dd44f93
41379 F20101130_AABWBN nesbitt_a_Page_49.pro
9bf203c02fda98b00ff511d7c44f1de3
05900ee11851bd92edfd48aebbe900255eed5603
22964 F20101130_AABWGM nesbitt_a_Page_29.QC.jpg
3b69e052af9afc4ef5340b44bd0a9d5b
ab23df7d2053acf05c95a8606f42945ea97dfe8f
69569 F20101130_AABVUR nesbitt_a_Page_70.jpg
19a6a9d9d09e659a8c05574566692b67
10258bf82ffb9c6bfa872b24a809854f7ec68600
F20101130_AABVPU nesbitt_a_Page_08.tif
1fc4fbec2ee577605069de07a567eb5b
969c1cb812cce70615af728f8021348d2358ddd0
F20101130_AABVZO nesbitt_a_Page_64.tif
382dd27db5a7f05b7ff333e2e0830e9f
e4d05d879f28998ebdaca588ef24e76d4f4a3915
47643 F20101130_AABWBO nesbitt_a_Page_50.pro
f44828fa99e946668d57fadc8d6bb4b4
325047e3f00d1026280b8c4eac853d1ff5d742eb
6115 F20101130_AABWGN nesbitt_a_Page_30thm.jpg
d88215719b2655ed49bc28fc262cb278
24574690184b51408e01d4c974a399f21852b73c
24073 F20101130_AABVPV nesbitt_a_Page_31.QC.jpg
00594d233c783437c98c76333fac2013
1c1579dde4f46a0865edcc68b365d37bd437185f
F20101130_AABVZP nesbitt_a_Page_65.tif
9a6298339d993dace71dc9d1158e8e1f
cddfdb95ed572ab9dadb578b52472ba4d8744a83
46834 F20101130_AABWBP nesbitt_a_Page_51.pro
6e9dcb3d97ab30654d442526a7c5f3f0
02f505591ed29660a24ff23881d8e345076f7cc0
71479 F20101130_AABVUS nesbitt_a_Page_71.jpg
437cc8f3d7c89a50d94ea27c363d6716
a0c2adae93fa34093176519d198487e2908690c6
6568 F20101130_AABWGO nesbitt_a_Page_31thm.jpg
dd31ee6e12ad30a3c5e477e2a2f5b8ea
d524cd9d7b2af42e94f7ebc48e927639775437f3
41920 F20101130_AABVPW nesbitt_a_Page_67.pro
a4f6ba1f035bf41aa847715188291497
32d989059a7e09a70e4ce7b323354f9cdceb41ef
F20101130_AABVZQ nesbitt_a_Page_66.tif
da4fb2a5dedc10e225c8fc566cbd529c
2003c48938db69055f53f2e39cc9e32b4fb81e41
48521 F20101130_AABWBQ nesbitt_a_Page_53.pro
5b06a1d59da2b472eadabbb2ca7f076d
a08d980e4a8efdc47e0ab4463e5aeb4ea8dca899
51906 F20101130_AABVUT nesbitt_a_Page_72.jpg
7a41bca01b67fddba6514e489054b4e2
df7021d88f8b25859a61ff4eae16ecf7acad3a7b
24280 F20101130_AABWGP nesbitt_a_Page_32.QC.jpg
878d90b2c34e3e17cc0296cfcd0c9b48
ff992508c86a4df0641c47943d8d68a1105ac082
45650 F20101130_AABVPX nesbitt_a_Page_52.pro
8648587ca28eed7d61872fc1ded10717
abe3cdf7a786e8e777ca844798235d7bb5ab3ac8
F20101130_AABVZR nesbitt_a_Page_67.tif
496d913fd867e8dd110e160d0e6909b8
e0245e50c79c1264bd2d97214595e35b88e3476d
52946 F20101130_AABWBR nesbitt_a_Page_54.pro
ccae45678dc8d64a499967675027e2a1
5b36219d331cb9807c7b742741f4c8ed2bf3abd8
80857 F20101130_AABVUU nesbitt_a_Page_73.jpg
b3eb5b79d6a1678358bebed74c5c863f
5d9eddf6e47caece4ec466e470b7f067c94ffd86
6648 F20101130_AABWGQ nesbitt_a_Page_32thm.jpg
fc93848570d93776b580b6ba91d97a2b
eeab61f3dc5612c95c022c10ca0b15e4ed0d563e
6784 F20101130_AABVPY nesbitt_a_Page_81thm.jpg
85b10db5b9011d10fff4e189a756641c
7b872aa6922658abd3faad0fc0b59f9d47f067ab
F20101130_AABVZS nesbitt_a_Page_69.tif
e4a464481024fd2e0229394c653be31e
653e8fa37623a9304e9e79e7394b3dbb00ec9dd7
49505 F20101130_AABWBS nesbitt_a_Page_55.pro
d357b0928cad00e73371c7bf82a04c9d
e12615ceebfa53834751902f3a2bf4cf02a5cb01
89053 F20101130_AABVUV nesbitt_a_Page_74.jpg
7353916d2c612af7ebcde428ad2c5df6
736a6e14b3d748eba5fecc58df8ef5b9759079f6
6393 F20101130_AABWGR nesbitt_a_Page_33thm.jpg
2db22de21427031a4f6d1d9ee760a0f2
0ea40f1f3fb566fcded092efa116eb0c3b769585
F20101130_AABVNA nesbitt_a_Page_79.tif
648bfdab2dad615e382dec63d1f51a4a
adda7d99ce1d34d324780056dc7ccf8bb601d379
F20101130_AABVPZ nesbitt_a_Page_71.tif
6f78c375d5dbe5daec704e07e862a535
97da64c5627d30ec1dcaa6f10b70c921663c31f9
F20101130_AABVZT nesbitt_a_Page_72.tif
f954889f39d7debc53a35b54f5e36c6f
d33c6454ceebdc240afab62d3117e082a4d4cc21
44986 F20101130_AABWBT nesbitt_a_Page_56.pro
4be3d50135d5f0fabfa81438ba1ae71d
9d9f150190e8bdd763f0ec0686de501697bd37ac
11920 F20101130_AABVUW nesbitt_a_Page_76.jpg
0a24b0323c15ac83eb8d364747da515b
998463d069969332a6487a42cbd38e8d8379756b
6507 F20101130_AABVNB nesbitt_a_Page_60thm.jpg
b264559088f8ed27b5af1a1e31794c11
28965552ce5d19faa2ea3e56e65e2b3626751870
F20101130_AABVZU nesbitt_a_Page_73.tif
7c3ee309ac835260a641dcd5bdb5df37
ada345dbd830fcb56f6176eedf70d4bcd91490e2
48427 F20101130_AABWBU nesbitt_a_Page_57.pro
c47c28dd80dc70198b0708290f290dd3
b946e9baac447d29f03bb896219fa4ddfea76ff3
44652 F20101130_AABVUX nesbitt_a_Page_77.jpg
f3f553831f74835a01ab029a6ecd5f65
777062df8c0dd7fdadb1c808c45f021d36c1603a
1550 F20101130_AABWGS nesbitt_a_Page_34thm.jpg
eccb7b6c96bd3e5c54d47410557d0473
4b7a6fd03ef4897835aa9cada980565ba56a07df
4503 F20101130_AABVNC nesbitt_a_Page_06thm.jpg
c1b1e60319579c9de550fe5babf314de
c6cbd2c96ac1f7aea6e692981ddbc0f20598db43
F20101130_AABVZV nesbitt_a_Page_75.tif
9502a74fcd7fb334465a04be7e763b28
fca7254fc7da05d07a081acefe0d03c43c645dee
51215 F20101130_AABWBV nesbitt_a_Page_59.pro
334871e9c0137273f0442c112fae2535
eebed39c937253be9876eef40779c02b1cea0399
22260 F20101130_AABVUY nesbitt_a_Page_78.jpg
9999b5cbb9494e98a7595e7528f13f94
c85988cbf8a3626c0891bd765036a3d5fdd3fb4b
21624 F20101130_AABWGT nesbitt_a_Page_35.QC.jpg
eb4cb7458926bc1b68988a507a757b60
f8ac15181ee647f43d04d9ec8e36017106f533d2
73132 F20101130_AABVND nesbitt_a_Page_27.jpg
2d9854ccdb74bff7276794c58edd6889
99361bad45112b1e9d84441474824ead79619fa5
F20101130_AABVZW nesbitt_a_Page_76.tif
f7c5fe721de5be911352ff2134df1c78
dbea560f27490eba97b1a819c29b08da8cf2a263
49954 F20101130_AABWBW nesbitt_a_Page_61.pro
86d31546934993a883535dd19b2e1714
2b5002f6437445a1cc2c1419ae7f61e640eed831
F20101130_AABVSA nesbitt_a_Page_39.tif
d21abb11e1db335ceea5065906b51a30
9c555c7f07d8fca5b93cd050037bb047e00d1bb8
72189 F20101130_AABVUZ nesbitt_a_Page_80.jpg
9f714dde7d1b85fe377e3b16773287a8
c9305f30067b2442b01a2a8e8b5022324c45290f
6043 F20101130_AABWGU nesbitt_a_Page_35thm.jpg
239ff60642f6081d39a7e02753aa1352
6d06f6569dead945009bb74b1c2ad92edb5e459a
F20101130_AABVNE nesbitt_a_Page_80.tif
e753020bcd3df93ebbd4afcaa077f166
92e86bcd0d4dcceb174794acf291395425a036a7
F20101130_AABVZX nesbitt_a_Page_81.tif
13525ef97f4dd5e2c43f11971f0aeda7
f03ea2d5e8d8b9cbecbb118e77ec4b1cf646d0f4
37608 F20101130_AABWBX nesbitt_a_Page_62.pro
ca05ece8219f50c011e0ef177aa1b336
ba9d8bef94ac049fc5672cf28b90e2112b10563a
1750 F20101130_AABVSB nesbitt_a_Page_62.txt
99c0cb913325ab765e01b1e65a6e1b02
d0a281ad359342c6a79b190f295e5b8b4e1d1729
16898 F20101130_AABWGV nesbitt_a_Page_36.QC.jpg
c34a5c3f60f12f69a74910c2f16c0a6e
f107d08aac23a7f1b2103ca82b580f54b6b59eb5
81949 F20101130_AABVXA nesbitt_a_Page_62.jp2
e3c7dafae73bdea02146becfe0c43f6c
a3bd042a928adb63d41440e92bbe51c43b37d702
F20101130_AABVZY nesbitt_a_Page_83.tif
bbc7cc4802b2bb33f6a7cb6aaa4acfaf
ce528f42ad4a6720d970b77079745bf660c10151
51004 F20101130_AABWBY nesbitt_a_Page_65.pro
bc9760b1ce368c7bc399edf9355bd4ef
7535eeebccc33a9178239f4d2951d83aba2fb5ab
6256 F20101130_AABVSC nesbitt_a_Page_22thm.jpg
fd051e0375d7eb192a708265233f9adb
35d22e55f01d37a92c776b7cb9ed4d7b8611fa00
19699 F20101130_AABVNF nesbitt_a_Page_42.QC.jpg
aeaa7db3e0f16bd5c2e18eebdcd720bc
73d28eadf419d592efe9762587d661ae4efcbb33
5058 F20101130_AABWGW nesbitt_a_Page_36thm.jpg
b96264abd7729a91a77e862c6372bc69
d142b5c2303eed25aeaea2c97462fabca0700d03
97127 F20101130_AABVXB nesbitt_a_Page_63.jp2
6ca8dd2be6218dea897819ad8c1bfc37
41d40a1581a85d4074e2ccfa92f03fb31131e51e
F20101130_AABVZZ nesbitt_a_Page_84.tif
85a60ec7fec2866e2374a24bfc07b7d3
2cd50e64f2ef68145b962b4f7f2a9132d3f1fbbc
3092 F20101130_AABWBZ nesbitt_a_Page_66.pro
4f440edfee030abcc3acde0756688895
9c7487e08b4f51731e0263ccfd990ef2cae4b2d5
1985 F20101130_AABVSD nesbitt_a_Page_16.txt
f6ca6a33f88798641e94be81dfc09ef5
c88bd9859ae93c2b77eca63e92711f33f3b7f221
29659 F20101130_AABVNG nesbitt_a_Page_01.jp2
743712374e69c357a5db69463bdfb65e
1b886a9d75a8ec9c9f7c659e02c93092ce9a98ce
5772 F20101130_AABWGX nesbitt_a_Page_37thm.jpg
8338f7268a3caca10970de3a0023d0e3
56526f18adb495320c5a614a993fc3ec2d630428
103089 F20101130_AABVXC nesbitt_a_Page_64.jp2
0509c1437a47067bc46cd08679ec6a3f
379e509ca11f266a018024f1809e8062028e3180
1956 F20101130_AABVSE nesbitt_a_Page_71.txt
4f4a6dc6b8812d132ebb07ac9c9436cb
05e3b4b8afb968b08b458e5b3e8f8eca1c3c91d3
2055 F20101130_AABVNH nesbitt_a_Page_33.txt
47ddea9985e03439628ea64d99c89218
ae753307e89157a328c1d952842114a659b22e46
22122 F20101130_AABWGY nesbitt_a_Page_38.QC.jpg
be5aa3cc46fbd1a6f9073b36a5a8d621
389dc695a80f84868456f54edf2070eb0240eeb6
F20101130_AABWEA nesbitt_a_Page_51.txt
b1a0693c48c36e421d6ef5b5ca824c54
2009c54e51cde18d67a576a7f890d08494525979
110813 F20101130_AABVXD nesbitt_a_Page_65.jp2
7b163213e112c2ed6d86d700f12b5349
1d598b25084362455319b6f11b22c6c441345b8c
6452 F20101130_AABVSF nesbitt_a_Page_29thm.jpg
7e720f706aca0560da1bea27895efa02
41f70e4abeda2bb3ed64960f65c2caf54f573e08
598400 F20101130_AABVNI nesbitt_a.pdf
a6cd3f9ed2b14c1e4a184e5ebe964f1b
1092e5cf42ef3e102fbef639cbac526859df73d2
5989 F20101130_AABWGZ nesbitt_a_Page_38thm.jpg
414ce4fdf52e15bf8e3251b69141433a
bb936805ccf67e1e94b21a7458bcc93ec61f2094
1913 F20101130_AABWEB nesbitt_a_Page_53.txt
8fff799765e79109cd529963f997eb34
717167736668b41c220a14738ebdca0b310dbed7
50256 F20101130_AABVSG nesbitt_a_Page_60.pro
26e4797dd21f5bc599c8456795111261
f34618eacea2a72c1e09897adbd12870ea0a3860
73785 F20101130_AABVNJ nesbitt_a_Page_22.jpg
26413f2be65d1c8e30b2fd8cec666156
8cc01905382a010135bd9dfd7333d2ac385ba977
2084 F20101130_AABWEC nesbitt_a_Page_54.txt
ea678a8d0bbe88b79f4bf8b3b1cd6b5e
31de39717ae48d900664558f3abf36eec5b1b6bf
9404 F20101130_AABVXE nesbitt_a_Page_66.jp2
eed57b8ffd99bcd0a542b15a289fedfa
c310f492a0e17e5303026c375cde119f7e3b6df6
20238 F20101130_AABVSH nesbitt_a_Page_37.QC.jpg
6461c729b6eb1792e5a01af1250731fb
a6abf394ecb1db7e1f448a5418be1c0ba724e47a
18421 F20101130_AABVNK nesbitt_a_Page_14.QC.jpg
a1b9858191fa555220975994ff5e61bc
f5231ba49b6d4d5eb42083c4889f5ffda61575de
6335 F20101130_AABWJA nesbitt_a_Page_73thm.jpg
acf4a203341af197a6ce8f5c47cd0a17
894d4e5334d663392931ba9260d2353bc11690b4
1957 F20101130_AABWED nesbitt_a_Page_55.txt
a764a4f4762664fbd2555bd2222ecb49
28bd1d2e76f28567871bb31f486830b64d179054
92149 F20101130_AABVXF nesbitt_a_Page_67.jp2
2e90e0f6c4bf5fd7eac571d8a6d127b5
c38bcfb28e1e479b7b56dcef6cb57b58e43482d0
22691 F20101130_AABVSI nesbitt_a_Page_80.QC.jpg
3a564c9dec2fe1403e9fd8fa31895434
9947e555fde85f8f96278346359ba12b8010c93e
100820 F20101130_AABVNL nesbitt_a_Page_19.jp2
9ee45b3d8f8d3a08083f63bfceaa6d92
1d6c32062f21dc09dd80a1ceeec3a6365e5fdacc
26453 F20101130_AABWJB nesbitt_a_Page_74.QC.jpg
bf8589aa1a350be2d51d29092fd46974
114ec143245950684028555396c40bdc699b06a4
1937 F20101130_AABWEE nesbitt_a_Page_57.txt
c288e0a81ce66ccec43498c552dfc821
b0820fb2fed0584ba70fbeda45a6db6a6f6d0041
109146 F20101130_AABVXG nesbitt_a_Page_68.jp2
7cb342b9c20a3f6a90b72a95b7f86fe7
78ec1c1cc85a8a6d91797a9ca4412bd446803ea1
2255 F20101130_AABVSJ nesbitt_a_Page_81.txt
485eb31247f2a9d7ddca84153df7d9a0
ddbc7b9cfe9a30f6187d76224740c7d2f35a6afa
2044 F20101130_AABVNM nesbitt_a_Page_59.txt
de6fb002bf21de4c88015a5cbc9e582e
87a72153e2071bff20991b22441d7d39982aad69
6922 F20101130_AABWJC nesbitt_a_Page_74thm.jpg
2a13d9df10dd33349329bb52f5606bb0
c1af155a2a123e92b2e3197d03675cecff221c21
1958 F20101130_AABWEF nesbitt_a_Page_58.txt
f5a9f2095768891fc15ec78dcceb8524
126c8c2ee3672aa5ba8020576362b8ed574b5ad9
106574 F20101130_AABVXH nesbitt_a_Page_69.jp2
77214bb2ff3a630aa46e0c3bab22b155
c82bd943d2f1e19583e389feb9abd676e7ee2c54
1726 F20101130_AABVSK nesbitt_a_Page_04.txt
67e593410ba0a519351d738270184fef
9635fe1253f63212296cffbff0e4eeaf77aa9408
2073 F20101130_AABVNN nesbitt_a_Page_32.txt
e5193a98fbbcc0f900a7cb1f8ccc16d7
7aafa97fb9a57c8ed7e1cefed54fa2695efe5a15
16432 F20101130_AABWJD nesbitt_a_Page_75.QC.jpg
eb4d3b0be2f7f1db8aa61690d872e7e3
a20724627880d82972d7fdbfe4f62a1ea233d690
1977 F20101130_AABWEG nesbitt_a_Page_60.txt
fcc5637ca3e3882fb8254b559a3ad116
9460a04002a8562f946d47840edbd81a59a40ed7
104424 F20101130_AABVXI nesbitt_a_Page_70.jp2
aceb10483641aba31c7ed8f363536337
fbdc305dd17048b42d768b71c908a947ccf57e96
127589 F20101130_AABVSL UFE0010582_00001.xml
cf3cc8f860f31c8fac18c6365b83aad6
f64cdb544a156cbe9b12be8cf74b33e2608910c9
61633 F20101130_AABVNO nesbitt_a_Page_15.jpg
0e70d6596d15fe5ead633f2cf502a5d1
14be977c05edb36332b2fba00c34d8fa81b1331a



PAGE 1

A MODEL OF GENTRIFICATION: M ONITORING COMMUNITY CHANGE IN SELECTED NEIGHBORHOODS OF ST. PETERSBURG, FLORIDA USING THE ANALYTIC HIERARCHY PROCESS By ASHON J. NESBITT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN URB AN AND REGIONAL PLANNING UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Ashon J. Nesbitt

PAGE 3

To all housing and economic development planners

PAGE 4

ACKNOWLEDGMENTS First, I acknowledge the Most High for extending His grace, and giving me the knowledge, passion and strength to complete this study. Secondly, I acknowledge my committee members. I thank Dr. Kristin Larsen for all of her advice, assistance and encouragement throughout the completion of this study. Her passion for housing has truly affected me, and caused me to take even greater interest in this area of planning. Also, her skills and talents as an author, mentor, professor, planner, researcher and scholar are admirable and have benefited me immensely in this process. I thank Dr. Paul Zwick for his expertise and his belief in me as a scholar. I thank Dr. Rhonda Phillips for providing me with exclusive access to her publications and for her financial support through the Center for Building Better Communities. I thank Dr. Marc Smith for balancing my views on housing and pointing me to other sources of information and research I wouldnt have otherwise thought to look in. I thank them all for your input and support. In addition, I would like to thank Karen Freggens, database manager with the City of St. Petersburg Building Department, Connie Clark, Secretary with the City of St. Petersburg Budget Office, and the Data Processing department of the Pinellas County Property Appraiser. I thank them for working with me and providing me with important data for this study. Finally, I acknowledge my family, close friends, and all of my fellow students who provided input or gave an encouraging word. I thank them for their kindness and love. It did not go unnoticed. iv

PAGE 5

This thesis is more than a demonstration and testament of my knowledge of an Urban Planning issue and my ability to conduct scholarly research. It is also a reflection of all who played a role--big or small--in the process of completing this study. I thank them again, and I hope they benefit from it, as much as I have benefited from them. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ...........................................................................................................viii LIST OF FIGURES .............................................................................................................x ABSTRACT .......................................................................................................................xi CHAPTER 1 INTRODUCTION...........................................................................................................1 2 LITERATURE REVIEW................................................................................................3 Gentrification................................................................................................................3 Origin and Introduction.........................................................................................3 Location and Scale................................................................................................4 Who and Why........................................................................................................5 Displacement.........................................................................................................9 Researchers Definition.......................................................................................10 Indicators....................................................................................................................11 Definition and Applications................................................................................11 Gentrification Indicators......................................................................................15 Thresholds...........................................................................................................16 Geographic Information Systems (GIS).....................................................................18 Introduction.........................................................................................................18 Definition.............................................................................................................18 Functions and Applications.................................................................................18 Summary.....................................................................................................................21 3 STUDY AREA..............................................................................................................23 4 METHODOLOGY........................................................................................................25 Explanation of Model.................................................................................................25 IdentifyingtheIndicators.......................................................................................26 Data Collection...........................................................................................................28 DevelopingtheEquation........................................................................................32 vi

PAGE 7

WeightedSuitabilityModel ...................................................................................34 5 FINDINGS AND RESULTS.........................................................................................37 Regional to Local Comparisons.................................................................................37 Professional Job Growth......................................................................................37 Change in Population..........................................................................................38 Change in Housing Units.....................................................................................38 Change in College-Educated Population.............................................................39 Change in Age 25 through 34 Population...........................................................40 Change in Age 55 through 64 Population...........................................................40 Change in Area Median Income..........................................................................41 Change in Median Single-family Unit Value......................................................42 Change in Housing Vacancy...............................................................................43 Change in Owner-Occupancy..............................................................................43 Unit Size..............................................................................................................44 Change in Commute Times.................................................................................45 Neighborhood-Specific Indicators..............................................................................45 Percentage of Housing Constructed before 1950................................................45 Proximity to the Central Business District and Interstate Highways..................46 Historic Designations..........................................................................................46 Major Relationships....................................................................................................47 Results.........................................................................................................................49 Weights................................................................................................................49 Values..................................................................................................................50 6 CONCLUSION..............................................................................................................55 Universal Applicability...............................................................................................55 Policy Implications.....................................................................................................56 Recommendations for Future Research......................................................................57 APPENDIX A DATA TABLES...........................................................................................................61 Regional to Local Comparison Indicators..................................................................61 Neighborhood-Specific Indicators..............................................................................63 B AREA MAPS................................................................................................................65 C GENTRIFICATION INDEX........................................................................................67 LIST OF REFERENCES...................................................................................................68 BIOGRAPHICAL SKETCH.............................................................................................71 vii

PAGE 8

LIST OF TABLES Table page 4-1: Regional to neighborhood comparison indicators......................................................26 4-2: Neighborhood-specific indicators..............................................................................28 4-3: Sources for regional to neighborhood comparison indicators....................................29 4-4: Sources for Neighborhood-specific indicators...........................................................29 4-5: Pairwise comparison matrix.......................................................................................33 4-6: Pairwise comparison matrix value pattern.................................................................34 5-1: Regional to neighborhood comparison indicators......................................................50 5-2: Neighborhood-specific indicators..............................................................................50 A-1: Total population.......................................................................................................261 A-2: Housing units.............................................................................................................61 A-3: Professional job employment....................................................................................61 A-4: College-educated population.....................................................................................61 A-5: Age 25 through 34.....................................................................................................61 A-6: Age 55 through 64.....................................................................................................62 A-7: Area median income..................................................................................................62 A-8: Single-family home value..........................................................................................62 A-9: Mean commute time..................................................................................................62 A-10: Housing vacancy......................................................................................................62 A-11: Owner-occupied housing.........................................................................................62 A-12: Rooms......................................................................................................................63 viii

PAGE 9

A-13: Housing pre-1950....................................................................................................63 A-14: Proximity to central business district.......................................................................63 A-15: Proximity to transportation corridors......................................................................63 A-16: Historic designations...............................................................................................63 ix

PAGE 10

LIST OF FIGURES Figure page B-1 Neighborhoods....................................................................................................65 B-2 Census Tracts......................................................................................................66 C-1 Gentrification Index............................................................................................66 x

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts in Urban and Regional Planning A MODEL OF GENTRIFICATION: MONITORING COMMUNITY CHANGE IN SELECTED NEIGHBORHOODS OF ST. PETERSBURG, FLORIDA USING THE ANALYTIC HIERARCHY PROCESS By Ashon J. Nesbitt May 2005 Chair: Kristin Larsen Cochair: Rhonda Phillips Major Department: Urban and Regional Planning Gentrification has emerged as a major issue in urban and regional planning, particularly in the central cities of large metropolitan areas. As more middle-class and upper-class residents begin to choose city life and reject suburban living, many older neighborhoods, once occupied exclusively by very-low income and low-income residents, are being re-inhabited by more affluent residents. Research on this topic is extensive, and several researchers have come to the same conclusions on the indicators of gentrification and the characteristics of the gentrifyer. However, there have been few attempts to develop methods to identify neighborhoods more likely to gentrify and monitor change in neighborhoods toward gentrification, which would allow planners and policy-makers to be proactive in their approach to preventing many of the negative affects of gentrification. xi

PAGE 12

In our study, we developed a model for monitoring gentrification based upon the indicators of gentrification identified in previous studies on the subject. The model uses St. Petersburg, FL as its base region and identifies four neighborhoods as potential areas of gentrification. The model uses statistics derived mostly from census data and converts them into spatial data using geographic information systems, and calculates a gentrification index based upon the indicators it identifies as most important to identifying gentrification. We found that two of the neighborhoods are indeed more likely to gentrify, and perhaps the process has already begun. Two neighborhoods may be likely to gentrify in the near future; while one may be likely in the distant future. The results of the analysis and gentrification index suggest policy changes and program implementation. Moreover, our study demonstrates that indicators, statistical analysis and the spatial analysis capabilities of geographic information systems can be used to identify complex planning issues and monitor community change related to those issues so that appropriate policy responses can be established. xii

PAGE 13

CHAPTER 1 INTRODUCTION As urban development returns to formerly distressed neighborhoods, gentrification emerges as a significant planning issue. Much of the discussion and research on this issue relates to affordable housing in general, and the plight of very-low, low and moderate-income families in terms of housing options for these income groups. As more middleand upper-class households choose urban instead of traditional suburban living, how can cities maintain affordability for lower-income households that do not possess the financial resources to allow them to choose where to live? Planners and researchers continue to struggle with solutions to this problem. While dealing directly with the affordable housing issue and striving to solve such a complex problem, planners and researchers have learned much about gentrification. They know much about the profiles of these middle and upper-income households that would potentially choose urban, or central city, living over suburban living. They also know the attributes these households look for in urban neighborhoods. In addition, research on gentrification identifies the major indicators of gentrification and establishes a basic understanding of each indicator in determining gentrification. However, with all of this knowledge, very few studies have sought to create a method of synthesizing quantifiable data related to these indicators in order to identify neighborhoods likely to gentrify and to monitor community during, and even prior to, the gentrification process. Our aim was to develop such a method by applying community indicators, the analytic hierarchy process and weighted suitability modeling. Thus, proper steps can be taken by 1

PAGE 14

2 planners and policymakers to mitigate the negative effects of gentrification before the process occurs. Developing the model involved several steps. First, we reviewed the current body of literature on gentrification to determine its major indicators. We examined information on community indicators and their application to planning as well as methods of spatial analysis and deterministic modeling currently available, yet typically unused in the field of housing planning. Second, we examined background information on St. Petersburg, Florida, the test city, and the five neighborhoods in St. Petersburg to justify the use of this area and to demonstrate implementation of the model. Finally, we discussed the findings related to each indicator; outcomes of the model; overall applicability of the model and recommendations for improvements and future research. Our study focused on identifying gentrification specifically. We also intended to demonstrate a useful application of spatial analysis and generate discussion and further research into its use to create a more proactive culture in the field of urban and regional planning as opposed to the reactive means of operation that presently characterizes much of professional practice. Geographic modeling can be a powerful tool in planning and policymaking. Our study demonstrated its particular usefulness in housing planning, and how indicators and spatial analysis can be applied to a real planning issue.

PAGE 15

CHAPTER 2 LITERATURE REVIEW Our study assessed three planning issues often considered separately. Specifically associated with housing, these issues are that of gentrification, community indicators and applications of the spatial analysis capabilities of geographic information systems (GIS). Much has been written in planning journals and other related publications about all three subjects. Researchers and practitioners continue to disagree on the true meaning of gentrification. Several articles and books have been written on the effectiveness of indicators in determining a communitys economic direction. The application of GIS to community, housing planning research and practice is still in its infancy; however, researchers and practitioners are beginning to look for ways to use this powerful software to examine such planning activities. Gentrification Origin and Introduction According to Atkinson (2003), Ruth Glass originated the term gentrification in the United Kingdom in 1964. The word is derived from gentry, referring to the middle and upper class households that are seen to displace local working-class groups. According to Glass, this displacement causes a change in the area. This change is the action referred to in the term gentrification, or the process of becoming a place for the gentry. This urban phenomenon has been studied and analyzed for forty years, since the inception of the term. Many definitions and ideas as to the causes of gentrification have been presented and debated over time. In this section, these definitions and ideas will be 3

PAGE 16

4 explored and discussed. Throughout the discussion, recurring themes, as well as key points most relevant to our study will be highlighted. This section will conclude with a definition of gentrification framed by the researcher. Location and Scale Perhaps a good place to begin a discussion of gentrification would be to define where it occurs and at what scale. According to the literature, gentrification is defined as an urban phenomenon, 1 occurring in large metropolitan areas. Most of the studies on gentrification have been done in large cities, and the process was first observed in London. In the United States, studies have been done on such cities as New York, Boston, Washington, DC, San Francisco, Atlanta, and Cleveland, Ohio. Further, gentrification is typically attributed to central cities. However, there are cases in which older suburbs in large metropolitan areas are experiencing change often associated with gentrification. Examples of this are Vallejo and East Palo Alto, California (Kennedy and Leonard 2001a). In addition, questions have arisen as to whether gentrification is truly limited to large metropolitan areas. Could gentrification also occur in smaller cities? A study done by the City of Gainesville, Florida Community Redevelopment Agency looks at the possibility of gentrification occurring there in an economically distressed community west of central business district known as the Pleasant Street neighborhood. These examples challenge the notion that gentrification is only a central-city issue and perhaps speaks to the future of gentrification studies (ADP, Inc. 2002). 2 1 Some recent studies show that gentrification is also taking place in small towns and rural communities 2 As the poor are being pushed out of central cities to inner ring suburbs, these older suburban areas are now seen as a possible location for future waves of gentrification.

PAGE 17

5 Gentrification is a process denoted by the middle and upper class reinvesting into the housing stock of poor inner city neighborhoods with high levels of abandonment, disinvestment and vacancy. Although gentrification only occurs in neighborhoods with specific attributes within cities (Gordon, Goudie and Peach 1996; Lang 1982), it is a significant phenomenon that is happening in an ever-increasing number of cities (Wyly and Hammel 1999; Wyly and Hammel 1999). For instance, in the United States the return to the city trend, which started in the larger, older metropolitan areas, particularly in the Northeast and Midwest, has begun to filter down to more recently urbanized areas in the South and West. Even with gentrification occurring in more and more locations, these neighborhoods have yet to outstrip the suburbs as the primary residential area for the middle and upper classes. One reason for the process of gentrification lagging behind suburban expansion is that, in most cases, neighborhoods prone to gentrification are not large enough to meet all the housing needs of a metropolitan areas growing middle and upper classes. Also, these areas tend to be attractive to a certain subset of the middle and upper class population based on such attributes as neighborhood location, urban character and architectural style. Who and Why Now that we know where gentrification occurs, what are the characteristics of gentrifyers and why do they choose to live where they do? Although there appears to be a general consensus on what the characteristics of gentrifyers are, there are some differing ideas on why gentrification occurs, and why in these areas of urban decay. Following a description of gentrifyers, this section will broadly discuss why gentrification occurs and specifically why in these urban neighborhoods.

PAGE 18

6 Since gentrification generally occurs in cities, in order to understand gentrifyers, one should understand why people like city living. City dwellers like the privacy specialization, and the hundreds of one-of-a-kind shopsthe excitementthe heterogeneity, the contrasts, the mixture of odd people. (Land, Hughes, Danielsen 1997, p. 437). However, most people identify city dwellers as less affluent or poor. In opposition to that perception, gentrifyers, also part of this city-dwelling population, are generally moderate to upper income households normally associated with suburban communities. In What Makes Gentrification Gentrification?, Redfern describes the gentrifyer as being other to the suburbanizing middle class. (Redfern 2003, p.2355) What makes the gentrifyer different from their suburban as well as their urban counterparts? First, unlike other city dwellers, such as the inhabitants of public housing complexes and working class households who cannot afford a house in the suburbs, gentrifyers can choose where they live. Second, gentrifyers are often highly educated professionals. Third, gentrifyers tend to be untraditional households. Gordon, Goudie and Peach (1996) identify gentrifyers as often being young, unmarried and childless as opposed to the typical two-parent, two-child household found in the suburbs or working class neighborhoods for that matter. Another population of gentrifyers includes empty nesters, those older couples or individuals who no longer have children living in the house with them. Other groups associated with gentrification are artists and gay and lesbian households. Often called urban pioneers, these are usually the first groups to move into a deteriorating area, rehabilitate the housing, and make the area attractive again (Solnit

PAGE 19

7 and Schwartzenberg 2000; Wyly and Hammel 1999). Whats interesting is that these groups often become the victims of what is called a second gentrification where these urban pioneers having proven the worth of a neighborhood, are subsequently displaced by investors and more affluent households. (Solnit and Schwartzenberg 2000; Wyly and Hammel 1999) Land, Hughes and Danielsen (1997) describe potential city dwellers, referred to in our study as gentrifyers, in the context of the environments from which they originate. They describe two different types of gentrifyers: suburban urbanites and urban suburbanites. These descriptions provide more insight into what gentrifyers seek in a neighborhood based on the urban context of the metropolitan area as a whole, and will thus help determine a neighborhoods potential for gentrification. The suburban urbanite is defined as a suburban resident with a similar lifestyle to a central-city resident. Suburban urbanites are found in the inner suburbs of Northeastern and Midwestern cities. Cities in these regions tend to be smaller in land area, denser, and surrounded by high-density suburbs that have central-city-type neighborhoods. (Land, Hughes, Danielsen 1997, p.441). Because they already live in neighborhoods that have similar characteristics of central city neighborhoods, they are more likely to choose central city living. In contrast, urban suburbanite would most likely be found in the suburbs of Sunbelt cities. These cities tend to be larger in land area with less dense urban cores as well as suburban-style subdivisions within the central city. These individuals are looking for areas that offer all of the advantages of urban living with all of the comforts of the

PAGE 20

8 suburbs. Therefore, in different urban contexts, gentrifyers seek different characteristics. The presence of these characteristics in a neighborhood affects its gentrification potential. The distinction between suburban urbanite and urban suburbanite is an interesting and significant one that bears importance in this particular study. The neighborhoods in our study are located in St. Petersburg, Florida. Although it is not an extremely expansive city geographically, its development pattern fits the Sunbelt City mode, with its less dense urban core and suburban-style subdivisions within its city limits. Therefore, gentrifyers in St. Petersburg would probably have the qualities of the urban suburbanite. In addition to the socioeconomic status of the gentrifyer, another, perhaps more controversial attribute of the gentrifyer is addressed in the literature race. Suburban expansion is associated with the term white flight, which refers to the exit of the white population from the central city to surrounding suburban communities. Gentrification counters this trend, with white residents returning to the city, sometimes going right back to the same communities they fled decades past. Still, gentrifyers are not necessarily white. For example, in certain areas of Atlanta affluent blacks are returning to the city (Kennedy and Leonard 2001a). Therefore, although gentrifyer usually has a clearly white racial identity, sometimes the term includes members of minority races. Gentrification occurs in regions where the housing market is tight (Kennedy and Leonard 2001a, 2001b; Lang 1982). When new housing demand outpaces the production of new housing, the price of housing will escalate. Thus, investment in the existing housing stock becomes an option considered by those with means (Nelson 1988, p. 15). Typically, areas chosen for investment have the greatest opportunity for reinvestment due

PAGE 21

9 to high levels of abandonment, disinvestment and vacancy. However, these attributes dont always guarantee a high potential for gentrification. Gentrifyers also choose areas characterized by their architectural style and high historic value of the homes as well as location near cultural amenities and/or the traditional central business district employment center (Lang 1982; Nelson 1988; Redfern 2001). Because these neighborhoods are so undesirable at the time of initial investment, the housing is cheap. In fact, Nelson (1988) argues that cheaper housing and the perceived profitability is more important than being fashionable. The reality of the situation most likely involves affordability, architectural style and profit. Thus, a gentrifyer is a middle or upper class, nontraditional household that prefers urban living. Gentrifyers are usually affluent whites, although this is not always the case. Further, gentrification is the result of a tightening housing market, making cheap inner city housing appear more desirable due to its affordability, profitability, location and style. Displacement One major issue of debate regarding what defines gentrification involves the issue of displacement. As more is invested in an area and property values rise, the poor and working class households that comprise the original residential population of a neighborhood will no longer be able to afford to stay there, resulting in displacement. While such displacement may be of economic benefit to cities overall as the rising property values increase the tax base (Kennedy and Leonard 2001a, 2001b), many view it

PAGE 22

10 as an unavoidable, socially detrimental consequence that overburdens the original residents, particularly renters in the neighborhood(Lang 1982, LaPeter 2004). 3 Many definitions and studies of gentrification require displacement to occur in order for an area to be declared gentrified (Kennedy and Leonard 2001a, 2001b). However, Wyly and Hammel (1999) speak of urban pioneers, the initial investors, as possibly displacing the original residents and oftentimes displaced by a second group of gentrifyers. Lang (1982) also uses the word often to describe displacement in the gentrification process (Lang 1982, p.6). Freeman and Braconis (2003) study of New York found that significant displacement does not have to occur for gentrification to take place. For instance, if the abandonment and vacancy rate is extremely high, then the likelihood of displacement is very low. Similarly, a study done by the City of Gainesville, Florida for its Pleasant Street neighborhood found that abandonment and vacancy were high enough for reinvestment to occur without large numbers of residents being displaced (ADP, Inc. 2002). Researchers Definition Based upon the various characterizations of gentrification explored in previous studies and their applicability to our study, we offer the following definition for gentrification: Gentrification is the process by which the socioeconomic status of a neighborhood populated mostly by lower-income households is substantially elevated by renewed interests and investments by higher-income households, including homebuyers, renters and commercial interests from outside the neighborhood so as to change the overall character of the neighborhood, and usually results in widespread 3 Gentrification changes the character of a neighborhood. The new middle and upper income residents not only upgrade the housing stock, they also bring with them new consumer demands, which affect area amenities, such as public spaces and retail offerings. Sometimes businesses are displaced as well as residents. However, this study has a residential focus.

PAGE 23

11 displacement of the lower-income residents already living in the neighborhood as well as the businesses they support. This definition includes the social as well as economic implications of gentrification. It also addresses both the residential and commercial aspects of gentrification. Although our study and previous studies on the subject tend to focus on the residential, the commercial component of gentrification is worth mentioning in any definition or discussion. Indicators Often used in community planning and economic development planning, community indicators evaluate social and economic change in an area. Different types of indicators function on different scales. Gentrification definitely has economic ramifications, thus certain types of indicators are typically present when it is occurring or likely to occur in a given area. This section defines indicators and outlines those relevant to gentrification. These specific indicators will become the basis of the gentrification model. Definition and Applications Phillips (2003) defines indicators as measurements that provide information about past and current trends to assist planners and community leaders in making decisions that effect outcomes (p.1). These measurements quantify the social, environmental and economic factors that work together to create change in a community or region. She describes them as gauges that document how much progress is being made toward reaching a certain goal or to show what a community or region is likely to become according to data gathered on the indicators. According to Hart (2003) and Oleari (2000), combining several indicators together to create a measuring system, or model, can

PAGE 24

12 provide (useful) information about past trends, current realities and future direction in order to aid decision making (quoted in Phillips 2003, p.2). Two basic types of indicators are defined in the literature. They are system (descriptive) indicators and performance indicators. System indicators condense individual measurements that describe multiple characteristics of a specific system in order to communicate the most pertinent information to decision-makers (Phillips, 2003; Hardi et al. 1997). System indicators work best with painting a picture of the current state of a system and are used to guide policy writing. Performance indicators are similar to system indicators in that they are both descriptive. However, performance indicators are also prescriptive. This type of indicator has a goal, reference value or target attached to it and measures how much progress is being made toward reaching that goal or target. Performance indicators are good for policy or program evaluation; therefore, these indicators can guide policy or program changes. Our study accurately describes the current situation in a neighborhood and assesses where the neighborhood is headed if the current trends continue, which will guide decision-making and policy writing. Therefore, performance indicators are most appropriate for our study. Indicator studies comprise three basic categories: economic, environmental and social. Indicators are most often employed in economic studies, which is what our study is. Of course, environmental studies assess ecosystems. An example of a social indicator study is the School Readiness Pilot Study for a Social Infrastructure Network completed by the Hillsborough County Planning Commission in 2003. This study measured several indicators derived from research in the field of education, and formulated a model that determines the likelihood of school readiness in neighborhoods throughout Hillsborough

PAGE 25

13 County, Florida. Although it is a social study, it provides a helpful example of how to use indicators in building a model for monitoring a community. Another important aspect of indicators is their scale. Phillips (2003) defines four levels of indicators in her publication. They are national and multinational, regional, local, and neighborhood indicators. National and multinational indicators measure trends on a national or international level. Regional indicators may exist on many different levels, as regions are defined in different ways. A region could be one state or a large section of a state, encompassing many different cities, towns and metropolitan areas. It could be a group of states, or it could be just one metropolitan area. Therefore, the scope of regional indicators is defined based on how the region is defined. Local indicators deal with specific municipalities. However, they assess the municipality holistically. Just like regional indicators, local indicators have varying scopes. They could be for one small town, a large city or an entire county. Neighborhood indicators look at the conditions in individual neighborhoods within cities or towns. For our study, regional to local comparisons as well as neighborhood-specific indicators will be used to develop the model. 4 In order to build a model that produces meaningful results, the proper indicators must be used. Phillips (2003) lists several criteria for the successful selection of indicators. Those criteria are: validity, relevance, consistency and reliability, measurability, clarity, comprehensiveness, cost-effectiveness, comparability and attractiveness to the media. Validity involves insuring the indicator is based on accurate data. Relevance is making sure the indicator relates directly to the issue at hand. 4 More specifics on the indicators and their justifications will be given in the Methodology chapter of this thesis.

PAGE 26

14 Consistency and reliability relate to the ability to collect the same quality of data over a period of time. Measurability addresses the ability of the indicator data to be collected directly from the neighborhood, locality, region or nation(s) being studied. 5 Clarity concerns how well the indicator is understood. Comprehensiveness measures the ability of one indicator to cover a wide range of issues yet retain the focus of the overall model. Cost-effectiveness reflects how much money (or time) must be put into collecting the data. Comparability involves how effectively the indicators can be used in different communities. Attractiveness to the media deals with how well the indicators and model are accepted by the press. 6 Although the aforementioned criteria are important in selecting indicators for monitoring community change, Phillips (2003) states that the true test of the success of an indicator or a model is whether or not the data collected in relation to that indicator or the results of the model prompt government officials to take action. However, out of all the criteria previously discussed, perhaps the most emphasis should be placed on the validity or accuracy of the data. In order for proper action to take place, the data associated with the indicators must be accurate. Indicators and models can then produce meaningful information that decision-makers can work with to affect proper change. Producing results that support proper shifts in policies and programs is the aim of our study. 5 Lindley Higgins Gathering and Presenting Information About Your Neighborhood published in 2001 by the Local Initiatives Support Corporation provides useful advice on collecting data (how and where). 6 In this case, the press would be journals and other respected publications.

PAGE 27

15 The use of indicators has a strong foundation in economic development planning and research. Most applications have targeted sustainable development, which is defined as development that seeks to meet the needs of the present without compromising the needs of the future. Most indicator projects evaluate community progress. However, indicators research presents very little on how individual indicators can be evaluated together to monitor community change. Our study creates a model for monitoring gentrification that involves the use of several indicators evaluated together. Gentrification Indicators The literature describes several indicators of the likelihood of gentrification. Some are regional; others are local or relevant at the neighborhood level. Further, gentrification is notoriously difficult to measure and the results (of the model) are sensitive to the indicators chosen, the time periods over which the indicators are measured and how neighborhoods are defined (Wyly and Hammel 1999, p. 726). Kennedy and Leonard (2001a) identify rapid job creation, a regional indicator, as the most significant indicator of potential gentrification. Rapid job creation provides more opportunity for those already living in the region as well as attracts new residents. Second on the list comes the supply of housing units in relation to demand. As more residents move to an area and current residents earn higher incomes, the demand for housing increases. If the current supply of housing cannot meet the demand, then housing prices will increase to curb demand. Thus, cheaper inner city housing becomes a viable alternative to more expensive, suburban housing. Other regional or local indicators include increased commute times, growth in certain population groups and nontraditional households and public investments. At the neighborhood level, the historic value of the housing stock, level of abandonment and percentage of owner-occupied housing are all

PAGE 28

16 indicators. For our study, these indicators and several others were chosen based on the literature. They will be identified and explained in the Methodology chapter of this paper. Thresholds Galster, Quercia and Cortes (2003) define threshold as the critical value of an indicator that triggers more rapid change. Another way to view a threshold is the point when change is completely apparent and cannot be easily stopped or reversed. Knowledge of the correct indicators is important to monitoring community change. Just as important is knowledge of the threshold related to each indicator. Thresholds are not arbitrary values. Accuracy in determining the threshold value plays a huge role in determining the success or failure of a model for monitoring change. Quercia and Galster (1997) describe four aspects of thresholds: geographic scale, absolute or relative impacts, time of impacts and pattern of relationship. Geographic scale is the area over which each variable is measured, and the corresponding threshold applies at that geographic scale. For instance, the threshold for a regional indicator should apply in the same manner throughout the region; whereas, the threshold for a local indicator will only apply to that specific locality. Absolute or relative impacts reflect, respectively, thresholds measured by absolute numbers or by percentages. For example, does the growth in the number of people from the ages of twenty-five through thirty-four have to increase by ten thousand in order to indicate change, or does it have to increase by ten percent? Time of impact addresses whether change has to continue for a certain period of time before rapid change occurs. For instance, does job growth have to continue for a certain number of years before there is a surge of interest in companies wanting to add jobs to an area? Finally, observing a pattern of relationship helps determine how the threshold of each indicator relates to those of other variables. For instance, how does job

PAGE 29

17 growth relate to population growth? Do job growth and population growth increase at the same rate all the time? Or, is there some point when jobs are increasing at such a rate as to cause an exponential increase in population from in migration? Is this job growth rate related to a rapid decrease in housing vacancy in the same manner as it relates to population growth? Data on each indicator should be tested against all other variables to determine the best value for each threshold. Several articles have been written on thresholds that relate to the study of gentrification Quercia and Galster (1997) determine that there is a threshold of middle-class households that must be reached before significant benefits, such as increased property values and retail demand. Downs (2002), Peng and Wheaton (1994) study the effects of restrictive land supply on housing prices, finding the point at which the amount of developable land available begins to effect housing price; however, housing output remains fairly constant. Chapple et al. (2004) study the effects of job growth on housing prices, finding that rapid job growth (particularly in certain industries) begins to effect housing prices over a certain period of time in certain locations depending on the structure of the metropolitan area. 7 The last example of threshold-related literature is Goodman and Thibodeau (1995) who found that the relationship between the age of housing units and price is a nonlinear relationship. All of these examples demonstrate that thresholds exist, they are very specific, they vary by indicator, and they possibly vary by location. Therefore, gentrification can be measured by the value of each indicator in relation to its threshold. 7 Growth in industries with the potential for rapid expansion, such as technology-based industries, could indicate the potential for a high rate of job creation over a short period of time in a region, creating new wealth and drawing new residents at a rapid pace. This results in a tightening housing market, leading to gentrification.

PAGE 30

18 Geographic Information Systems (GIS) Introduction Due to its spatial applications and analysis capabilities, a geographical information system (GIS) is a critical component of our study. The following paragraphs define what GIS is, examine the functions of GIS, and review how GIS has and can be used in real estate research. Some of this information is similar to the material presented on indicators. These overlaps will also be highlighted. Definition Luc Anselin (1998) defines GIS as a powerful set of tools for collecting, storing retrieving at will, transforming and displaying spatial data from the real world for a particular set of purposes (p. 116). Most people associate GIS with specific software packages. Generally, GIS synthesizes value information with locational and topological information into a searchable database. Value information, or attributes, include the price or size of a housing unit. Locational and topological information include the address or census block where the unit is located. Functions and Applications Anselin (1998) also outlines the four major functions of GIS: input, storage, output and analysis. Of the four functions, analysis, or spatial analysis, is the focus of our study. Spatial analysis has four sub-functions. They are selection, manipulation, exploration and confirmation. Selection involves obtaining information relating to certain variables specific to a certain location from a spatial database. Data manipulation involves the creation of spatial data and is done through attribute values (averaging, summation), spatial information (coordinates) and data integration (combination of attribute values and spatial information).

PAGE 31

19 The next two capacities of spatial analysis are exploration and confirmation. These two are considered the heart of spatial analysis. Exploration, or exploratory spatial data analysis (ESDA) is described as being a body of techniques used to describe and visualize spatial distributions, find patterns of association (spatial clustering), identify extremely unique observations (outliers) and suggest different spatial regimes or other forms of spatial instability (nonstationarity) (Anselin, 1998 p. 120). ESDA identifies two classifications of indicators of spatial association. They are global and local. Most of the recent research and literature has focused on the use of local indicators of spatial association (LISA). These indicators can detect patterns of association as well as test a specific patterns uniformity. LISAs are well suited for map visualization, and overlaying LISA maps of different variables is very helpful in deciding variables that should be used in models. For these reasons, our study focuses on LISAs how they illustrate patterns and are used to build models. Confirmation, or confirmatory spatial data analysis is described as model-driven. It involves four steps: model specification, estimation, diagnostics and prediction. These four steps imply an iterative process in which models are tested until the best one is found. As mentioned in the previous section on indicators, studies such as this one should result in recommendations for government action based on the results. Therefore, it is important to find the best model for studying and producing the most meaningful results for the issue at hand. Also, in the discussion on confirmatory spatial analysis, Anselin (1998) addresses regression models and their usefulness in predicting values. One previous study uses a regression model to predict rental rates in several markets and geographically illustrate their results for Atlanta and Boston. This model incorporates the

PAGE 32

20 physical attributes of apartments and their relation to price based upon previous research. The model illustrates geographically how rents are likely to vary in relation to the average rent based on location and demonstrates how variables, or indicators, can be analyzed using GIS to graphically display a neighborhood reality. The aforementioned study testifies to the effectiveness of regression analysis, demonstrating how the interaction of variables can be assessed to accurately display and monitor an issue. Our study uses a deterministic model involving the pairwise comparison method to determine the weight of each variable associated with gentrification. This method, developed by Saaty in 1980, involves comparing each variable to the other variables individually, creating a ratio matrix that outputs the relative weights of each variable. This method was chosen based on the knowledge of the general effects of each indicator on the likelihood of gentrification expressed in the literature as well as research as well as its compatibility with the spatial analysis functions of GIS. The application of GIS, and its spatial analysis capabilities, to housing research has been very minimal. According to Can (1998) this lack of research is due to ignorance of available tools; difficulty in obtaining the updated, detailed and accurate information required for GIS-based analysis; and the relatively recent availability of special processing requirements for housing research. These reasons are valid, particularly the availability of data to make using GIS worthwhile and meaningful. Most of the specific data collected on housing is done through the census. Some data is estimated on a yearly basis, but these estimations are generally not done at the census block level (Can 1998, p. 69). However, some information not available in its most recent version may be available through other non-traditional sources such as the local Property Appraiser or Chamber of

PAGE 33

21 Commerce. In fact, it is possible to get more specific information from a source such as the Property Appraiser down to the parcel as opposed to census data, which only measures down to census tract for certain types of data. One important issue to consider when gathering information from a variety of sources is consistency. While accuracy is very important, ensuring that all data for all variables relates to the same year and is measured at the same geographic level is equally important when using GIS to conduct research and build models. Despite the challenges, GIS is an appropriate tool for housing research. The visualization capacity of GIS allows researchers to see patterns and trends that might not be evident just by examining tables and graphs (Ghose and Huxold, 2004, p. 19). Also, its analysis capabilities allow for the examination of several forces and indicators at one time to determine their effect and guide policy action. Summary The goal of this review of the literature was to establish a working definition of gentrification and examine indicator studies and GIS tools to show their application to the study of gentrification and the creation of a model for monitoring gentrification. The review discussed the major issues and debates in the study of gentrification, resulting in a definition of gentrification for use in our study. Next a discussion of indicators outlined how they have been used (particularly in economic development planning) and how they can be applied to the study of housing and model building. Finally, an overview of GIS and its application to housing research continued to build on themes offered in the discussion on indicators as well as demonstrated the practicality of GIS in relation to housing research and community monitoring. In all of these discussions, important points were highlighted and analyzed in their relation to our study. The next two chapters

PAGE 34

22 describe the specific geographic area used for our study and the specific details of our model.

PAGE 35

CHAPTER 3 STUDY AREA Our study focuses on St. Petersburg, Florida as the test region due to its growing population, rapid job growth, geographic constraints, dwindling availability of large developable parcels, and growing affluence. With a population of nearly 250,000 residents, St. Petersburg ranks as the fourth largest city in the state of Florida, and functions as one of the urban centers in the Tampa Bay metropolitan area the states second largest metropolitan statistical area and one of its fastest growing. St. Petersburg is located in Pinellas County, a densely populated, nearly built-out county along the west coast of Florida. The county itself is a large peninsula, surrounded on three sides by water. St. Petersburg, at the southern end of the county, is also surrounded by water on three sides. Also like the county, St. Petersburg is nearing build-out in terms of undeveloped land. Due to its geography, no outward expansion can take place, including typical large-scale, suburban-style developments that characterize current development in much of the rest of Florida. Moreover, the city is experiencing significant job growth, particularly in high-paying financial services and technological-oriented jobs, attracting thousands of new residents in recent years. Therefore, as these trends continue, we contend some St. Petersburg neighborhoods are bound to experience gentrification. Our study identifies four neighborhoods as probable targets for gentrification: Bartlett Park, Old Southeast, Roser Park, and Crescent Lake. Although each neighborhood is unique, they all share aspects that attract gentrifyers. All are located immediately adjacent or within 1.5 miles from the central business district. All are among 23

PAGE 36

24 the oldest neighborhoods in the city. Roser Park, Old Southeast, and a portion of Crescent Lake called Round Lake are designated historic districts on the national level, local level or both. One neighborhood, Uptown, has been identified as the control neighborhood. This neighborhood features many of the same characteristics of the four neighborhoods identified as gentrification targets. It is a historic district and sits directly adjacent to St. Petersburgs central business district. However, it does not receive the same attention from officials, planners, residents and the press as the other neighborhoods in terms of the characteristics of and potential for gentrification. Therefore, our study asserts that change occurring in Uptown will most accurately reflect the overall change taking place in the city of St. Petersburg. The national trend of central city redevelopment has not missed St. Petersburg. In fact, St. Petersburgs central business district has been recognized several times as an example of successful downtown redevelopment. As the central business district generates more activity, we hypothesize that the identified four surrounding neighborhoods will begin to feel the effects of eminent gentrification. The model developed for our study will prove or disprove the correctness of that hypothesis.

PAGE 37

CHAPTER 4 METHODOLOGY Gentrification literature describes the various measurable indicators of gentrification. It also describes the difficulty in reversing the negative effects of gentrification, most notably the displacement of residents. Since the indicators are known, gentrification must be measurable. However, no attempts to quantify these indicators and relate all of them empirically to some index of the likelihood of gentrification occurring in a neighborhood have been found in previous studies. This chapter describes the method created for monitoring gentrification in our study, determines specific indicators outlined in the gentrification literature using common statistical methods and GIS technology, and tests the model on the five neighborhoods described in the previous section. Explanation of Model Building the model for monitoring gentrification involved four basic steps, each of which contained smaller steps. The first basic step was the identification of the indicators of gentrification to be used in the model. The second basic step involved collecting the appropriate data for those indicators, converting that data into usable statistics, and mapping those statistics for each indicator using GIS independently. The third step involved determining relationships between the indicators and the threshold values for each indicator. The fourth and final step established an equation for a gentrification index based on the statistics and thresholds to determine the likelihood of gentrification occurring in the study area and mapped the results of the equation using GIS. 25

PAGE 38

26 IdentifyingtheIndicators This first step in developing the model identified the appropriate indicators. Perhaps the most important step in the process, choosing the right indicators to use, greatly determined the effectiveness of the model. Our study considers sixteen indicators based upon gentrification literature and the researchers definition of gentrification. The majority of the indicators chosen use census data and other data readily available to researchers, demonstrating the accessibility of the model for practicing planners. We divided the indicators into two groups: regional to neighborhood comparisons and neighborhood-specific indicators. Regional to neighborhood comparisons describe conditions that exist or changes in regional demographics that should reflect on all areas of the metropolitan region. For instance, if area median income (AMI) increased by a large percentage for the region, one expects to find a large increase in the AMI of each neighborhood in the region. Neighborhood-specific indicators describe conditions and qualities specific to a particular neighborhood. A neighborhoods location would classify as a neighborhood-specific indicator. We chose twelve regional to neighborhood comparison indicators and four neighborhood-specific indicators (Tables 4-1 and 4-2). Table 4-1: Regional to neighborhood comparison indicators Name Description Justification Change in Professional Employment The change in the number of people working jobs requiring post-secondary education (AA, AS, BA, BS, MA, MS, Ph. D., technical certificate) as a percentage of overall employment These tend to be higher-wage jobs. An increase in the number of higher-paid workers increases area median income (AMI), driving up housing costs.

PAGE 39

27 Table 4-1 Continued Name Description Justification Change in Population The change in the total population A rapid population increase usually relates to a growing job market, one of the leading indicators of gentrification. Change in Housing Units The change in the total number of housing units A slow growth in the number of housing units with respect to population and job growth leads to rising housing costs. Change in college-educated population The change in the percentage of the population that is college-educated One of the characteristics of a likely gentrifyer; tend to have higher incomes and affinity for city amenities. Change in Age Cohort 25-34 The change in the percentage of the population in this age range This cohort relates to one of the characteristics of a likely gentrifyer (high-wage, young, single or married w/ no children). Change in Age Cohort 55-65 The change in the percentage of the population in this age range This cohort relates to one of the characteristics of a likely gentrifyer (empty-nester; active lifestyle). Change in area median income (AMI) The percentage change in AMI Growing AMI usually relates to a growing job base, increased educational level of residents, and relates to an increase in housing costs. Change in Median Owner-Occupied Unit Value The percentage change in the value of owner-occupied single-family residential units attached as well as detached. Rising housing costs signifies increase demand for housing, a leading indicator of gentrification. Change in Average Commute Times The number of minutes commute times have increased/decreased over time One main reason residents are choosing to move back to central cities relates to increased commute times. % Housing units occupied The change in the percentage of housing units that are occupied by either renters or their owners Higher occupancy in combination with high demand raises housing prices.

PAGE 40

28 Table 4-1 Continued. Name Description Justification % Owner-occupied units The change in the percentage of housing units actually occupied by their owners Rising homeownership tends to reflect a greater amount of income within households as well as growing neighborhood stability an attractive quality. Unit Size The number of rooms in a housing unit Larger homes tend to attract higher-incomes. Therefore larger homes in older areas are likely to attract gentrifyers. Table 4-2: Neighborhood-specific indicators Name Description Justification % Housing Built Pre-1950 The percentage of all the housing units built prior to 1950 The historical value of the houses is part of the allure of inner-city neighborhoods to gentrifyers. Proximity to Central Business District (CBD) The number of miles the census tract is from those tracts making up the CBD Part of the attraction is the closeness to CBD, where jobs, culture and entertainment are located. Proximity to Major Transportation Corridors (Interstate Highways) If interstates run through city, the number of miles to the nearest interchange; if not, the number of miles to the nearest major corridor Easy access to corridors leading to CBD as well as suburban markets one of the important factors to gentrifyers. Historical Designations Number of historic structures or if entire tract is within historic district Designations curtail demolition, encouraging renovation; historic value attractive to gentrifyers. Data Collection Most of the data collected comes from the United States Bureau of the Census (Census). However, some data was collected from other sources.

PAGE 41

29 Table 4-3: Sources for regional to neighborhood comparison indicators Name Units Source Change in Professional Employment Percentage Census Change in Population Percentage Census Change in Housing Units Percentage Census Change in college-educated population Percentage Census Change in Age Cohort 25-34 Percentage Census Change in Age Cohorts 55-65 Percentage Census Change in AMI (area median income) Percentage Census Change in Owner-Occupied Unit Value Percentage Census Change in Average Commute Times Percentage Census % Housing units occupied Percentage Census % Owner-occupied units Percentage Census Unit Size Number Census Table 4-4: Sources for Neighborhood-specific indicators Name Units Source % Housing Built Pre-1950 Percentage Census Proximity to Central Business District (CBD) Number Scaled street map of city Proximity to Major Transportation Corridors Number Scaled street map of city Historical Designations Percentage City Government, National Register of Historic Places In order to gauge change and show a clear trend, data collection encompassed a 20-year period (three decennial censuses) for each indicator whose source is the Census (2000, 1990 and 1980). Data gathered on other indicators also spanned the same twenty-year timeframe where available. If data was available only over a shorter time period, data collection began with the earliest year available. Collecting data in this manner kept the intervals the same to establish trends over the same number of years as the indicators

PAGE 42

30 based on the Census. In addition to consistency in time intervals, the values must also be geographically consistent. Thus, data not available from the Census was appropriately scaled or proportioned to match the census tracts used for the neighborhoods analyzed in our study. We defined the region as the city where the neighborhoods are located St. Petersburg, Florida. The neighborhood refers to each of the five neighborhoods analyzed in our study area separately. The boundaries of each neighborhood matched up almost perfectly with the boundaries of their respective census tracts (Figures B-1 and B-2). Census data generally comes as a simple count (integer) or where appropriate, as a dollar amount. However, in this research, percentage change bears more relevance. For instance, the median income in the city could increase by more absolute dollars than a neighborhood, but the neighborhood could show a higher percentage increase, reflecting a greater rate of change. Therefore, the counts for each regional to neighborhood comparison indicator were transformed into a percentage change value using the following formula: Percent Change = [(X Y)/Y ] 100 where X = Value from 2000 Census or most recent available, and Y = Value from 1980 Census For neighborhood-specific indicators, no rate of change was measured between 1980 and 2000, as they reflect neighborhood characteristics at their present state based on the 2000 census, demonstrating potential based on current conditions.

PAGE 43

31 Most of the indicators are dynamic and measured by percentage change. However two indicators describe static conditions and carry number measurements -distance to central business district and distance to major transportation corridors. It is quite possible for distance to major transportation corridors to change due to construction of new corridors. 1 Yet, we determined that no new transportation corridors affecting these neighborhoods were constructed during the study period. Also, the locations of the traditional central business district (downtown) and the location of each neighborhood remain stationary. For these reasons, a number value is the appropriate measure for these indicators. Each indicator is then mapped using ArcGIS 2 according to the percentage or integer value associated with each. First, the GIS shape files for the appropriate city boundary and the census tracts are downloaded from the Florida Geographic Data Library 3 into GIS creating the base map. Then the attribute table for the census tract layer was edited to include the fields for the values relating to each indicator. Next, the values in each of these fields were converted from vector attributes to raster attributes. 4 These values 1 If new major transportation corridors are constructed, then the distance from a study area to a major transportation corridor may change; thus making this a dynamic variable that may be more appropriately measured by percentage change. 2 ArcGIS is a GIS software package from ESRI most often used by planners, developers and researchers 3 The Florida Geographic Data Library is an electronic resource providing free access to GIS shape files for all counties in the State of Florida and their corresponding attribute tables and metadata files. 4 Vector data associate attributes with each feature point, line, and polygon; whereas raster data represents surfaces as grids of equally sized cells that contain attribute values and location coordinates. With raster data, groups of cells that share the same value represent the same type of geographic feature. For instance, all census tracts would be represented with the same color regardless of their associated rate of population increase when displayed as vector data; whereas, with raster data, only tracts with the same rate of increase in population would share the same color on the map.

PAGE 44

32 are then reclassified using the binary system of 0 and 1 according to their value in relation to the regional percentages. 5 The reclassification assigned a value of 0 to all values less than the regional percentage, and assigned a value of 1 to all values greater than the regional percentage in most cases. In a few instances, the reclassification was based on the opposite relationship. For example, a reclassification value of 1 was assigned to tracts with a change in vacancy rates less than the regional rate. The reclassified values were converted to individual shape files and added to the base map as separate layers. The purpose of doing this was to spatially and visually reinforce the change occurring in the study area in relation to each indicator. DevelopingtheEquation The equation used to analyze the five neighborhoods utilizes deterministic neighborhood value analysis in combination with weighted suitability analysis to determine a gentrification index. The following sections outline this process Deterministic Neighborhood Value Analysis Since monitoring gentrification engages several indicators, the study used deterministic neighborhood value analysis to weight the values of several variables to get one final index for gentrification. Deterministic neighborhood value analysis uses the following equation: = C 1 X 1 + C 2 X 2 + C 3 X 3 + + C n X n where = index C weight of the first indicator X 1 C weight of the second indicator X 2 5 Since the current body of literature establishes no generic thresholds for these gentrification indicators, the most appropriate measures of change are the regional percentages.

PAGE 45

33 C weight of the third indicator X 3 C n = weight of the nth indicator X n The weights for each value were determined using the pairwise comparison method established by Saaty in 1980 described in the literature review. 6 This method determines the weight of variables in decision-making using the comparison matrix (Table 4-5), testing each variable against all other variables individually: Table 4-5: Pairwise comparison matrix Variable X 1 Variable X 2 Variable X 3 Variable X n Variable X 1 1 X 2 :X 1 X 3 :X 1 X n :X 1 Variable X 2 X 1 :X 2 1 X 3 :X 2 X n :X 2 Variable X 3 X 1 :X 3 X 2 :X3 1 X n :X 3 . . . . . . 1 . Variable X n X 1 :X n X 2 :X n X 3 :X n :X n 1 Comparisons were done on a scale of 1 to 9 using the following descriptions: 1 = equally important 2 = slightly more important 3 = somewhat more important 4 = moderately more important 5 = more important 6 = much more important 7 = significantly more important 8 = very much more important 9 = extremely more important When comparing variables to themselves, the value always equals one. If the comparison of variable X 2 to X 1 yields one value, then the comparison of X 1 to X 2 yields 6 An alternative to the researcher developing the weights would be to survey local professional planners with housing expertise as well as area residents using the same criteria and develop the weights through a method of consensus building an iterative process by which all those involved would come to an agreement on the value of each indicator to the whole equation.

PAGE 46

34 the reciprocal value. For example, if variable X 2 is significantly more important than X 1 (value =7), then variable X 1 is significantly less important than X 2 (value = 1/7). Table 4-6: Pairwise comparison matrix value pattern Variable X 1 Variable X 2 Variable X 3 Variable X n Variable X 1 1 1/X 1 :X 2 1/X 3 :X 1 1/X 1 :X n Variable X 2 X 1 :X 2 1 1/X 2 :X 3 1/X 2 :X n Variable X 3 X 1 :X 3 X 2 :X 3 1 1/X 3 :X n . . . . . . 1 . Variable X n X 1 :X n X 2 :X n X 3 :X n :X n 1 These comparison values were then normalized by the following equation: Normalized Value = Comparison Value (1/ Total of all values in column). Then these normalized values were summed up by column. This total became the weight, or coefficient C, assigned to each indicator. After establishing the C values for each indicator, the deterministic neighborhood value analysis equation uses reclassified values for each indicator described in the previous section as values to measure their total effect. For each neighborhood, the study analyzed the regional to neighborhood comparisons and neighborhood-specific indicators separately, providing a total for both to be used later in the weighted suitability analysis. Although the study analyzed the five neighborhoods separately, it used the same equations for each, employing the same C values. Using the same equation demonstrates the regional applicability of this analysis. The uniqueness of the totals for a neighborhood would come from its values. WeightedSuitabilityModel The weighted suitability model is a method of spatial analysis often used in real estate development to determine the suitability of a site for a specific type of

PAGE 47

35 development targeting a specific demographic. It assigns weights to multiple groups of variables in the same manner that multivariate regression applies weights to individual variables. Since our study uses two categories of indicators, the weighted suitability model effectively illustrates the relationship between the two sets of indicators and their effect on the overall decision-making of potential gentrifyers. The weighted suitability model is used to establish the equation for the final index of the likelihood of gentrification, G. For our study, regional to neighborhood comparison indicators Regional carried a coefficient of 0.8, accounting for 80% of the result, and neighborhood-specific indicators Neighborhood carried a coefficient of 0.20, accounting for 20% of the result. We derived these proportions from the gentrification literature that identifies the major indicators for gentrification as increasing commute times, rapid job and population growth, and changes in demographics of age and income, all issues accounted for in the regional to neighborhood comparisons. Neighborhood-specific attributes, such as proximity to the central business district and architectural character, also bear much significance. However, according to the gentrification literature, these characteristics carry less importance than the regional to neighborhood comparisons. For this reason, the 80% to 20% ratio applied well to the model, giving the regional to neighborhood comparison indicators the majority of the weight without marginalizing the effects of the neighborhood-specific indicators. Using the weighted suitability model, the data accurately produces a gentrification index (G) for each neighborhood in the study area with the following equation: G = 0.8 Regional + 0.2 Neighborhood Where

PAGE 48

36 Regional = deterministic neighborhood value analysis of regional to local comparison indicators, and neighborhood = deterministic neighborhood value analysis of neighborhood-specific indicators. The Raster Calculator in the Spatial Analyst menu of ArcGIS calculated the G values for each neighborhood and added their graphic representation to the base map as a separate layer. The G values were measured on a scale of 0 to 1, with 0 equal to 0% likelihood of gentrification and 1 equal to 100% likelihood of gentrification. This process outlines a method for empirically measuring and graphically displaying the potential for gentrification. It provides a means to quantify physical and social attributes of an area and relate them mathematically to describe neighborhood change.

PAGE 49

CHAPTER 5 FINDINGS AND RESULTS This thesis focuses on the use of census and other relevant data to reveal long-term patterns of change and use them to monitor gentrification in a neighborhood. The following chapter will report the findings for each indicator separately, looking at overall trends from 1980 to 2000 as well the differences between the rate of change in the 1980s and the rate of change in the 1990s. Although our model does not use the rates of change from 1990 to 2000, the trends they reveal are worth discussing. Regional to Local Comparisons In many cases, indicators in the local areas (neighborhoods) were consistent with the general trend in the region. However, in some cases, the local areas and region registered opposite trends. Overall, the findings for these indicators revealed that although these neighborhoods share common characteristics, such as their geographic locations, they are each unique; therefore, lending themselves to a range of possibilities in their likelihood for gentrification. Professional Job Growth Between 1980 and 2000, the city of St. Petersburg experienced a 10.09% increase in the number of residents with professional jobs. Further analysis reveals that the majority of that increase occurred between 1990 and 2000, a 7.38% increase. From 1980 to 2000, all five neighborhoods in the study area register an increase in the number of residents with professional jobs. Two neighborhoods, Roser Park and Crescent Lake, show an increase much higher than the city. With a 19.82% increase in 37

PAGE 50

38 professional jobs, Roser Parks rate of increase is nearly twice that of the city. Crescent Lakes 16.38% increase is also significantly higher. This shows the strong appeal of these neighborhoods to professionals. Bartlett Park, Old Southeast and Uptown also showed increases of 5.2%, 9.82% and 8.36% respectively, perhaps implying a growing interest, but not yet on the level of the other two neighborhoods. Change in Population The census reports that the population of the city of St. Petersburg increased from 238,547 in 1980 to 248,232 in 2000, a 4.02% increase in population. Further examination shows that the majority of this population increase occurred between 1990 and 2000, as the census reports a population of 238,629 in 1990. The trend of increasing population for the city of St. Petersburg as a whole does not hold true in any of the neighborhoods in the study area. In fact, some neighborhoods experienced a sharp decline in population. The Crescent Lake neighborhood, represented by Census Tract 235, had the smallest change, with a 0.94% decrease in population from 1980 to 2000. In ascending order, Old Southeast (Tract 204) shows a 3.31% decrease, Uptown (Tract 234) shows a 9.6% decrease, Bartlett Park (Tract 205) shows a 18.26% decrease, and Roser Park (Tract 213) shows a 51.0% decrease. Considering the increase in city population, these neighborhood-level decreases are unexpected. On face value, these decreases in population could represent disinterest and disinvestment. However, this population decrease may be explained by trends relating to other indicators. Change in Housing Units Between 1980 and 2000, the number of housing units in the city of St. Petersburg increased 4.3%. However, over both censuses, all five neighborhoods report a decreasing

PAGE 51

39 number of housing units. Still, Roser Park shows a strikingly high decrease in housing units, reporting a 78.46% decrease. The second-highest decrease occurred in Uptown, reporting a 24.84% decrease. Bartlett Park ranks third, with an 18.26% decrease, followed by Crescent Lake and Old Southeast, with 16.35% and 10.56% decreases respectively. These decreases in housing units may be explained by conversion of housing units to office space. For instance, due to its location near a large hospital district and university campus, some housing units in the Roser Park neighborhood may have been purchased by those institutions for future expansion or by businesses wishing to be close to them. Another explanation could be the conversion of large structures back to single-family uses that were formerly rented as multiple units. Change in College-Educated Population From 1980 to 2000, the number of persons with Bachelors, Graduate and Professional degrees in the city of St. Petersburg has increased 8.25%, from 14.57% in 1980 to 22.82% in 2000. This increase appears to be steady, with 4.19% occurring between 1990 and 2000. All five neighborhoods also report an increase in the number of residents with four-year degrees or higher. Three neighborhoods show a rate of increase higher than that of the city. They are Old Southeast, Roser Park and Crescent Lake, with 19.82%, 8.36% and 16.09% increases respectively. These larger increases imply that these are clearly neighborhoods of interest for college-educated persons. Bartlett Park and Uptown report increase of 5.2% and 6.12% respectively. Although these represent a gain in college-educated residents, the smaller values indicate these neighborhoods arent as popular as the other three.

PAGE 52

40 Change in Age 25 through 34 Population From 1980 to 2000, St. Petersburg shows a slight increase in the number of residents from the age of 25 through 34 with an overall increase of 0.74% from 13.02% of the population in 1980 to 13.76% of the population in 2000. There was a larger increase from 1980 to 1990, going from 13.02% to 14.96%, then decreasing in 2000 to 13.76%. The population in this cohort increased during the twenty-year period in two of the neighborhoods and decreased in the other three. Uptowns increase of 0.95% is slightly above the citys rate of increase. Crescent Lake experienced a more significant 3.86% increase. However, Bartlett Park, Old Southeast and Roser Park all experienced decreases 6.19%, 3.78% and 5.24% respectively. Although the rate of increase appears slow for Uptown and Crescent Lake, both are gaining residents of this age faster than the city, indicating an attractiveness of these neighborhoods to younger adults. The decreases in Bartlett Park, Old Southeast and Roser Park imply an unattractiveness of these neighborhoods to younger adults. Change in Age 55 through 64 Population The population aged 55 through 64 has decreased in St. Petersburg from 12.15% in 1980 to 9.17% in 2000, a 2.98% decrease. The majority of this decrease occurred between 1990 and 2000 when the 55 to 64 population decreased 1.69% from 10.86% to 9.17%. Two neighborhoods registered an increase in this age group, whereas the population in this age group declined in three of the neighborhoods. Bartlett Park experienced an increase of 3.72% from 1980 to 2000, the majority occurring between 1980 and 1990 (2.87%). This slowing increase may imply a developing disinterest in the

PAGE 53

41 area from this age group. Old Southeast reports an overall increase of 0.38%. Although the population in this age group decreased between 1980 and 1990 from 9.79% to 8.43% of the total population, it increased again between 1990 and 2000 to 10.17%. This indicates that the Old Southeast may be developing into a neighborhood of interest for this age group. Roser Park, Uptown and Crescent Lake report decreases of 1.17%, 2.05% and 3.91% respectively. In all three cases, the majority of decrease occurred between 1980 and 1990. This slowing decrease may also indicate increasing interest in these three neighborhoods for this age group. Change in Area Median Income The area median income has increased dramatically in St. Petersburg, going from $11,798 in 1980 to $34,597 in 2000, a 193% increase, or nearly tripling in twenty years. The majority of that increase took place between 1980 and 1990, when median income experienced a 146.26% increase from $11,798 to $23,577. This significant increase in median income could be explained by an increasing number of two-wage earner households and the greater upward mobility of women during this time period. All five neighborhoods experienced significant increases in median income. Crescent Lake experienced the largest increase (234%), going from $6,964 in 1980 to $23,225 in 2000. Not far behind with a 200% increase is Old Southeast, rising from $10,386 in 1980 to $31,163 in 2000. Uptown experienced a 169% increase from $8,466 in 1980 to $22,768 in 2000. The smallest increases were in Bartlett Park and Roser Park, reporting 135% and 158% increases respectively. Bartlett Park increased from $8,135 to $19,125, while Roser Park increased from $7,584 to $19,531. Just as with the city, all five neighborhoods experienced their greatest gains between 1980 and 1990.

PAGE 54

42 Although all five neighborhoods have gained significantly, their median incomes still lag behind that of the city of St. Petersburg as a whole. However, with gains of 200% and 234%, incomes in Old Southeast and Crescent Lake are growing at a faster rate than the citys rate of increase, indicating interest in these areas from higher-income households. Moreover, of the five neighborhoods, Roser Park is the only neighborhood in which a higher rate of increase in income occurred from 1990 to 2000 than the citys rate during that same period an increase of 69.76% for the neighborhood compared to 46.74% for the city, implying that Roser Park has caught the attention of higher-income households. Yet the overall numbers from 1980 to 2000 reveal that there still remains a large presence of low-income households in the neighborhood. Change in Median Single-family Unit Value From 1980 to 2000, single-family homes in the city of St. Petersburg increased in value by 126%, going from $35,800 in 1980 to $81,000 in 2000. This increase mostly took place during the 1980s, when values increased by 96.81%, or nearly doubled. Both Bartlett Park and Old Southeast experienced similar rates of increase 122% and 125% respectively. Values in Bartlett Park grew from $20,600 in 1980 to $45,800 in 2000; whereas values in Old Southeast grew from $37,900 in 1980 to $85,400 in 2000. The three other neighborhoods saw values rise at a higher rate than the city. Roser Park and Crescent Lake experienced the greatest increase in single-family home values. In Roser Park, values rose an impressive 255%, more than tripling from $19,200 in 1980 to $68,100 in 2000. Likewise, Crescent Lake values grew by 211%, also more than tripling from $28,700 in 1980 to $89,200 in 2000. Although not as high, Uptown values rose 170% from $29,000 in 1980 to $78,200 in 2000. In addition, all three neighborhoods had higher rates of increase between 1990 and 2000 than the 29.19% rate of the city, with

PAGE 55

43 Roser Park reporting a 51.33% increase, Crescent Lake reporting a 50.42% increase and Uptown reporting a 48.95% increase. Of these three neighborhoods, values in two Roser Park and Uptown still lag behind the regional median value. Still, the rising values generally relate to rising demand, implying specific interest of homebuyers in these three neighborhoods. Change in Housing Vacancy Interestingly, from 1980 to 2000 the city reports an overall increase in vacancy of 2.24% from 1980 to 2000. However, the vacancy rate decreased by 3.74% between 1990 and 2000, indicating increased absorption of housing units in the city overall. Four of the five neighborhoods followed similar patterns. Bartlett Park experienced the highest increase in vacancy, rising from 17.02% in 1980 to 28.77% in 2000. Vacancy in Crescent Lake rose 6.67% over the same time period. In Uptown, the rate grew 3.77%. Roser Park reported the smallest increase with 0.36%. However, all four experienced decreases in their vacancy rates in the 1990s. Crescent Lake reports a 10.04% decrease during that decade. Roser Park had the second-highest decrease of 6.9%. Uptown and Bartlett Park experienced decreases of 2.16% and 0.02% respectively. Old Southeast is the only neighborhood to experience an overall decrease in vacancy from 1980 to 2000. Vacancy decreased by 2.41%, going from 15.97% in 1980 to 13.56% in 2000. Still, all five neighborhoods continue to have higher rates of vacancy than the city as a whole. However, with vacancy rates decreasing at a faster rate than the city between 1990 and 2000, both Roser Park and Crescent Lake appear to be neighborhoods of interest. Change in Owner-Occupancy Surprisingly, owner-occupancy decreased over the twenty-year period by 1.17% in the city of St. Petersburg from 57.04% in 1980 to 55.87% in 2000. However, the rate of

PAGE 56

44 owner-occupancy increased by 2.8% between 1990 and 2000. Only one other neighborhood followed a similar pattern Bartlett Park. Here, owner-occupancy decreased by 2.16% between 1980 and 2000, but it increased by 5.04% between 1990 and 2000. The other four neighborhoods experienced growing owner-occupancy over both time periods. Ownership in Roser Park grew 9.64% from 1980 to 2000, with 95% of that growth taking place in the 1990s. Old Southeast, Uptown and Crescent Lake also experienced an increase in ownership from 1980 to 2000, with increases of 2.54%, 0.68% and 1.23% respectively. However, these neighborhoods saw greater rates of increase in the 1990s than over the twenty-year span of 1980 to 2000. Old Southeast reports an increase of 9.78% during the 1990s. Uptown and Crescent Lake saw increases of 4.3% and 5.15% respectively. With the exception of Bartlett Park, owner-occupancy increased faster in the neighborhoods than in the city overall from 1980 to 2000. However, ownership increased faster in Bartlett Park than the city overall from 1990 to 2000. Both trends imply a growing number of homeowners, associated with a stabilizing neighborhood. Moreover, these rates indicate the growing appeal of these neighborhoods to homebuyers. Unit Size The median number of rooms in owner-occupied units in 2000 was 5.5 rooms for the city. Of the five neighborhoods, Old Southeast and Roser Park had a higher median number of rooms, with 6 and 7.4 rooms respectively. Bartlett Park homes tend to be smaller than that of the city, with a median of 5.3 rooms. The same applies to Uptown, with a median of 5.2 rooms. Crescent Lake reflects the citywide median of 5.5 rooms.

PAGE 57

45 The larger homes of Old Southeast and Roser Park lend themselves to greater attractiveness; whereas, the smaller homes of Bartlett Park and Uptown may not be as attractive. As the homes of Crescent Lake tend mirror the city as a whole, other indicators would have a greater effect on the likelihood of gentrification taking place there. Change in Commute Times Over the twenty-year period the average commute times increased in all instances. The city average commute time increased 5.64% from 19.5 minutes in 1980 to 20.6 minutes in 2000. Uptown reports the greatest increase in commute times, rising 37.84% from 14.8 minutes in 1980 to 20.4 minutes in 2000. The second-largest increase happened in Old Southeast, with a 24.57% increase from 17.5 minutes in 1980 to 21.8 minutes in 2000. Crescent Lake, Roser Park and Bartlett Park experienced increases of 6.96%, 7.21% and 1.39% respectively. If gentrification is happening in these areas, then these commute times are still low enough to attract new residents. An alternative explanation may be that a change in commute times is not a significant indicator of gentrification. Neighborhood-Specific Indicators Percentage of Housing Constructed before 1950 All neighborhoods have relatively high percentages of housing units built prior to 1950. Two neighborhoods, Uptown and Crescent Lake, have maintained the majority of their older residential units, reporting that 57.47% and 56.04% of their units were built prior to 1950. However, the three of the four neighborhoods believed to be targets of gentrification reported the lowest percentages of old homes. Bartlett Park reports in 2000 that 41.16% of its units were constructed before 1950. The percentages for Old Southeast and Roser Park were 44.08% and 42.17% respectively. It appears that Uptown and

PAGE 58

46 Crescent Lake did a better job of preserving historic character over the years than has Bartlett Park, Old Southeast and Roser Park. If these three neighborhoods are gentrifying, this data may counter the hypothesis that gentrifyers are generally attracted to the architecture of older neighborhoods. Proximity to the Central Business District and Interstate Highways Roser Park, Uptown and Crescent Lake are directly adjacent to the business district, and are all bordered on at least one side by an interstate highway. In all cases, the bordering interstate highway is the divider between the neighborhood and the central business district. Bartlett Park and Old Southeast are located further away one mile and 1.5 miles respectively. However, they are both within a five minute drive of the central business district. Their proximity to the central business district and the interstate highways, which provide access to suburban job markets, make these neighborhoods attractive to gentrifyers looking for shorter commutes to the central business district or who dont mind the reverse commute to the suburbs in exchange for easy access to the cultural and entertainment amenities of the central business district. Historic Designations Old Southeast contains the greatest number of historic designations with a local historic district designation and three individual historic structure designations, two national and one local. Crescent Lake follows with a portion of the area designated as the Round Lake national historic district and one historic structure. Lastly, Roser Park is designated a national historic district. Both Bartlett Park and Uptown have no historic designations. According to previous studies (Redfern, 2001; Nelson, 1988; Lang, 1982), maintenance of historic character makes an area more attractive to gentrifyers. Historic

PAGE 59

47 designations in a neighborhood or the designation of an entire neighborhood as a historic district attest to the neighborhoods commitment to maintain that character. Therefore, two of the four neighborhoods believed to be targets for gentrification Old Southeast, Roser Park are likely to succeed; whereas, Bartlett Park and Uptown may not attract as many gentrifyers as they are not designated like the other two. Major Relationships Examination of these statistics revealed some relationships between indicators. There were some expected correlations, such as that between population and housing units. However, some relationships didnt follow usual patters, such as that between housing vacancy, number of units and value. The following paragraphs will discuss relationships found between these indicators. Overall, the number of housing units in the city increased at the same rate as population increase, indicating that housing production in the city has generally kept pace with population increase. However, although population has decreased in the neighborhoods, the number of housing units has decreased at a much higher rate in all cases except Bartlett Park. Although the citys growing population may be redistributing itself in other areas, there still remains interest in these neighborhoods in 2000, perhaps by larger households than had previously occupied them in 1980. This theory runs counter to how gentrification research identifies a gentrifyer -described as a nontraditional household (young, single persons or unrelated individuals), or a married couple with no children living in the house (younger couple or older yet active, empty-nest couple). The theory of growing household size is further supported by the overall decrease in population of the age cohorts generally associated with these two demographics ages 25 through 34 and ages 55 through 64. An increasing household

PAGE 60

48 size may also indicate that gentrification does not necessarily relate to growth in those demographics, but could possibly relate to growth in families with upwardly mobile householders; thus, adding another dynamic to ideas of how gentrification manifests itself in different cities. Likewise, as the number of residents with bachelors degrees or higher increases, the number of residents with professional jobs increases. In most cases, the number of professional workers has increased at a higher rate than the number of college-educated residents. This, perhaps, indicates an increasingly competitive job market that continues to attract new, highly-educated residents. In addition to possibly reflecting an increasing number of two-income households, the increase in area median income in all geographic areas also relates to the growing number of highly-educated professional workers as demonstrated by the statistics gathered for this research. This increase in income and percentage of college-educated residents supports the hypothesis that these neighborhoods are targets for gentrification, as previous studies on the subject indicate that job growth, particularly professional job growth, is the major indicator of the potential for gentrification. Finally, interesting relationships exist among the statistics relating directly to the housing units. As the number of units decreases, one expects the vacancy rate to also decrease. Conversely, as the number of units decreased, the vacancy rate increased in nearly all instances. Despite an increasing vacancy rate, the value of single-family units continued to rise. This increase in value probably relates to the general increase in owner-occupancy, which also supports previous gentrification research that points to increasing home-ownership as a sign of gentrification. In addition, the two neighborhoods with the

PAGE 61

49 largest homes, Old Southeast and Roser Park experienced the highest rates of increase in homeownership. Roser Park, with the largest homes, experienced the highest rate of increase in home value, while Uptown and Crescent Lake, with the largest collection of homes constructed before 1950, experienced the second and third-largest increases in home value. Moreover, these three neighborhoods immediately adjacent to the central business district Roser Park, Uptown and Crescent Lake experienced the highest rates of home value increase. This supports gentrification research on the attractiveness of large, older homes close to the central business district to gentrifyers. Results Using the model described in the previous chapter the results strongly support the hypothesis in one neighborhood. In other neighborhoods, the results counter the hypothesis. The following paragraphs will describe the application of the statistics developed from the census data, the relationships discovered among the statistics related to each indicator in the model, and the resulting gentrification index. Weights The weights for each indicator were calculated using the pairwise comparison described in the methodology chapter. Each indicator was compared to the other indicators individually based in part on their ranking of importance as expressed in the literature on gentrification and in part on their specific relevance to gentrification in St. Petersburg. For instance, the change in commute time is a major indicator of gentrification according to the gentrification literature, as neighborhoods experiencing gentrification should register decreasing commute times. However, four of the five neighborhoods report commute times increasing at a higher rate than the region (the city of St. Petersburg). Therefore, in fitting with the hypothesis, change in commute times

PAGE 62

50 carries a smaller weight with neighborhoods in St. Petersburg. Tables 5-1 and 5-2 display the weights calculated for each indicator: Table 5-1: Regional to neighborhood comparison indicators Name Weight Percent of Total Weight % Change in Population 0.0864 8.64% % Change in Housing Units 0.1684 16.84% % Change in Professional Jobs 0.1875 18.75% % Change in College Educated Population 0.0712 7.12% % Change in Age Cohort 25-34 0.0362 3.62% % Change in Age Cohort 55-64 0.0439 4.39% % Change in Area Median Income 0.0630 6.30% % Change in Single-Family Unit Value 0.1062 10.62% % Change in Commute Time 0.0379 3.79% % Change in Housing Vacancy 0.1141 11.41% % Change in Owner-Occupancy 0.0419 4.19% Unit Size 0.0380 3.8% Table 5-2: Neighborhood-specific indicators Name Weight Percent of Total Weight % Housing Pre-1950 0.43175 43.17% Proximity to Central Business District 0.26025 26.03% Proximity to Major Transportation Corridors (Interstate Highways) 0.2076 20.76% Historic Designations 0.3478 34.67% Values For use in the equation, the model reclassified the statistics for each indicator using the binary system values of 0 and 1. The regional (city) values were used as the

PAGE 63

51 thresholds to determine how indicator value was reclassified. Since gentrification literature gives neither universal thresholds nor any direction on how to stratify the reclassification of values based on preset thresholds, reclassification based on the city values using the binary system was the most appropriate and effective means of evaluating each indicator. The reclassification for each indicator is as follows: % Change in Population % Change in Housing Units 1 = Tract > 4.02% 1 = Tract < 4.3% 0 = Tract < 4.02% 0 = Tract > 4.3% % Change in Professional Employment % Change in College-Educated Pop. 1 = Tract > 10.09% 1 = Tract > 8.25% 0 = Tract < 10.09% 2 = Tract < 8/25% % Change in Age 25-34 Population % Change in Age 55-64 Population 1 = Tract > 0.74% 1 = Tract > -2.98% 0 = Tract < 0.74% 0 = Tract < -2.98% % Change in AMI % Change in Single-Family Home Value 1 = Tract > 193% 1 = Tract > 126% 0 = Tract < 193% 0 = Tract < 126% % Change in Commute Times % Change in Housing Vacancy 1 = Tract < 5.64% 1 = Tract < 2.24% 0 = Tract > 5.64% 0 = Tract > 2.24% % Change in Owner-Occupancy Unit Size 1 = Tract > -1.17% 1 = Tract > 5.5 Rooms 0 = Tract < -1.17% 0 = Tract < 5.5 Rooms % Housing Pre-1950 Proximity to Central Business District 1 = Tract > 0% 1 = Tract = 0 miles (directly adjacent) 0 = Tract = 0% 0 = Tract > 0 miles Proximity to Transportation Corridor Historic Designations 1 = Tract = 0 miles (directly adjacent) 1 = Historic designations present 0 = Tract > 0 miles 0 = No historic designations present This reclassification was done using the reclass function in the Spatial Analyst menu of ArcGIS. The resulting equation for the gentrification index (G) was

PAGE 64

52 G = 0.8 [(0.0864 in population) + (0.1684 in units) + (0.1875 in professional jobs) + 0.0712 in college-educated) + (0.0362 in age 25-34) + (0.0439 in age 55-64) + (0.0630 in AMI) + (0.1062 in single-family value) + (0.0379 in commute time) + (0.1141 in housing vacancy) + (0.0419 in owner-occupancy) + (0.0380 unit size)] + 0.2 [(0.43175 housing pre-1950) + (0.26025 proximity to CBD) + (0.2076 proximity to transportation corridors) + (0.3478 historic designations)] This equation used the reclassified values for each indicator to calculate the gentrification index G. We used the trends from 1980 to 2000 to establish the values for each indicator in the gentrification index calculation. This equation was inputted into the Raster Calculator in the Spatial Analyst menu of ArcGIS, which inputted the reclassified values into the equation and yielded gentrification indices with the following values: Bartlett Park = 0.1559 Old Southeast = 0.4577 Roser Park = 0.7358 Uptown = 0.4072 Crescent Lake = 0.6277 Multiplying those values by 100 more clearly communicates the relative likelihood of gentrification: Bartlett Park = 15.59% Old Southeast = 45.77% Roser Park = 73.58% Uptown = 40.72% Crescent Lake = 62.77% Both Roser Park and Crescent Lake show the greatest likelihood for gentrification with gentrification indexes (probabilities) of 73.58% and 62.77% respectively. Old Southeast and Uptown have lower likelihoods of gentrification, with indexes of 45.77% and 40.72%. Bartlett Parks index comes in substantially lower than Uptown at 15.59%. These indexes strongly support the hypothesis with Roser Park and Crescent Lake, moderately support the hypothesis with Old Southeast, and disprove the hypothesis for

PAGE 65

53 Bartlett Park. With a likelihood of 40.72%, Uptown proves not to be representative of the city of St. Petersburg and should be re-evaluated in its role as the control neighborhood. Clearly, Roser Park and Crescent Lake are experiencing the most rapid change, and likely would gentrify before the other neighborhoods in the study area. Perhaps, the process has already begun in these two neighborhoods. What differentiates these two neighborhoods from the others that explain this higher likelihood? Geographically speaking, Roser Park, Crescent and Uptown are adjacent to the central business district. However, Roser Park and Crescent Lake are closest to the core of the central business district where most of the activity takes place. Both neighborhoods showed great increases in the percentage of residents in professional employment, the only two with higher rates of increase than the city. Uptown and Crescent Lake both have high percentages of older housing, Uptown with the highest of all neighborhoods in the study area. However Crescent Lake homes are larger, equal to the city average. Similarly, Old Southeast has a slightly larger collection of older homes; however, single-family homes are significantly larger in Roser Park than in Old Southeast. Neither Bartlett Park nor Old Southeast are directly adjacent to the central business district. However, Bartlett Park has shown the smallest increase in professional employment and college-educated residents; its average home size is smaller than the city average, and it has the smallest collection of older homes of all the neighborhoods in the study area. While these explanations do not address every indicator, they begin to explain why Roser Park and Crescent Lake exhibit high potential for gentrification and Bartlett Park trails so far behind. Perhaps, the process has already begun in those neighborhoods, with Old Southeast and Uptown poised to

PAGE 66

54 follow them in a second wave of gentrification and Bartlett Park in the distant future if ever at all.

PAGE 67

CHAPTER 6 CONCLUSION In our study, we identified several indicators of gentrification according to previous research on the subject and used them to develop a model that monitors community change and assesses the likelihood of gentrification with a deterministic statistical analysis method and a weighted suitability analysis that uses the spatial analyst capabilities of geographic information systems. Our hypothesis defines four neighborhoods as targets of gentrification (Bartlett Park, Old Southeast, Roser Park and Crescent Lake) and one control neighborhood (Uptown). The results are mixed. Our model proves our hypothesis correct for Roser Park, Crescent Lake, and arguably Old Southeast. Our hypothesis is proved wrong for Bartlett Park, found not to be a target of gentrification (yet) and Uptown, found to be more of a target than expected. However, our study demonstrates the capabilities of statistical analysis and geographic information systems to address housing issues in a proactive manner by anticipating the likelihood of gentrification. Universal Applicability Since gentrification manifests itself in accordance with the unique dynamics of a local housing market, it is impossible to develop an equation with coefficients that can be used for analyzing any neighborhood in any city. However, the indicators of gentrification are generally the same everywhere. Therefore, in order to apply our model to other cities, the coefficient values associated with each indicator should be adjusted to reflect how they interact in that specific market. 55

PAGE 68

56 Policy Implications Any model for monitoring a planning issue should produce meaningful results for use in the development of policies and programs. Our deterministic model of gentrification allows planners to accurately identify those neighborhoods more likely to gentrify and use that information a basis for changes to or the creation of new policies, programs and planning initiatives. Planning, overall, has developed into a reactionary practice. More proactive planning needs to take place. However, in order for planners to work proactively, they must be equipped with the tools necessary to provide solid analysis on which to base their recommendations. Our model provides an excellent example of how common planning tools and resources can be used for analysis of a complex planning issue gentrification. The results of the model can be used to guide the implementation of specific programs, such as tax credit and grant programs for rehabilitation or new construction to encourage a mix of incomes and discourage the displacement of low-income residents that often occurs with gentrification. Implementing such programs before gentrification begins in earnest will increase the effectiveness of the programs by intervening before any negative effects can occur. For St. Petersburg specifically, efforts should focus affordable construction and rehabilitation dollars in neighborhoods such as Roser Park and Crescent Lake immediately, as developers and speculators will surely start to purchase properties, if they have not already. The same should be done in Old Southeast and Uptown as they both will likely follow the same path of gentrification as Roser Park and Crescent Lake. As for Bartlett Park, perhaps the city may want to encourage the development of more middle-income housing to strengthen the neighborhood. However, realizing Bartlett Park shares

PAGE 69

57 many things in common with gentrifying areas, policies should be written to prevent the neighborhood from falling victim to its own success. For instance, amendments to the housing and future land use elements of the city of St. Petersburgs Comprehensive Plan could be written to specifically address the possibility of gentrification in Bartlett Park and similar neighborhoods. In addition to policy changes, programs such as a community land trust, municipal purchase of residential properties or tax increment financing for affordable housing could be implemented to insure that low and moderate-income households will continue to have housing opportunities in the neighborhood. Recommendations for Future Research Overall, our model appears to be effective in calculating a gentrification index and establishing a model for monitoring community change based on trends over long time periods. However, specific aspects of the model could be adjusted to increase its effectiveness, calling for additional research: Studying the change in the same indicators over a shorter period of time. In several cases, the statistics revealed different trends between 1980 and 2000, and 1990 and 2000. Although comparing changes in values and statistics associated with the indicators over a longer period of time gives a broader base of knowledge, examining the short term trends may help to balance the perspective in assessing the likelihood of gentrification. Since real estate markets can be very volatile, it may prove beneficial to run this deterministic gentrification model based on ten year intervals. For instance, in addition to obtaining the index with a base year of 1980, the gentrification index could be calculated using 1990 as the base year instead. Based on the data collected, the results would probably be somewhat different.

PAGE 70

58 Projecting beyond the census. Reliance on census data lends itself to inaccuracy as years pass. For example, the 2000 census could describe 2001 and 2002 demographics fairly accurately. However, the 2000 census would not reflect 2005 demographics accurately. The overall effectiveness of the model depends upon the accuracy of the statistics inputted. Therefore, one may consider calculating projections of the census data, such as those done by the Bureau of Economic and Business Research at the University of Florida, for each indicator to more accurately relate the current situation to that of the base year. Use of other indicators in addition to those measured by the census. Previous research on gentrification identifies several other potential indicators that are not used in this model. However, some data was collected on these indicators. One major indicator of gentrification is increased sales activity. According to the Pinellas County Property Appraiser, Bartlett Park had 33 sales in 2000 as opposed to 10 in 1980. Crescent Lake had 125 sales in 2000 as opposed to only 13 in 1980. Comparison of these rates of increase to the rate of change in the citys sales activity would strengthen the model more. Other indicators include the change in the number of residential (new construction or major renovation) permits issued as well as the number and type of capital improvement projects planned or that have occurred in the neighborhood over time. In addition, surveying local residents may identify indicators not mentioned in the literature. Incorporation of these other indicators not measured by the census as well as those identified by residents (and not mentioned in the literature) would further support changes related to other indicators and greatly enhance the effectiveness of the model.

PAGE 71

59 Develop weights and thresholds through survey. Community involvement in determining the weights and acceptable thresholds could greatly strengthen the validity of the model, as the value and thresholds related to community indicators are usually decided upon by members of the community. The weights for each indicator were developed based on the researchers interpretation of information presented in the literature search and the data gathered on each indicator, lending itself to a certain amount of subjectivity some may consider problematic. More accurate weights could be developed by surveying other housing and planning experts as well as area residents through public meetings or written surveys. The range of weights relating to each indicator reported in the surveys could, perhaps, be averaged to determine the actual weight used in the model; therefore, creating a better equation with more accurate results. Run model again in the future to see if results change. As implied by the indexes for each neighborhood in our study, some neighborhoods are further into the process of gentrification than others. As neighborhoods, cities and regions are dynamic entities, the gentrification index as calculated by the model may be different in the future for each neighborhood. One possible extension of this research would be to re-evaluate these neighborhoods at the time of the 2010 census to monitor how they have changed since 2000. Determine a tipping point index and assigning appropriate policies and programs to specific indexes. One of the major goals of our study is to create a monitoring tool for use in policy decision-making. Therefore, determining the index value that describes a neighborhood in the early or moderate states of gentrification as opposed to when the process of gentrification is fully underway and therefore quite

PAGE 72

60 difficult to address would be excellent continuations of our study. Then proper policy and programs to could be related to specific index ranges through testing this model on neighborhoods in other cities to show that neighborhoods with the same index generally display similar attributes. Similarly, neighborhoods could be re-evaluated over time to discover how long it takes neighborhoods to cycle through the gentrification process. Our study successfully accomplishes its goal of developing a model for measuring gentrification and monitoring community change with results that can have meaningful effects on policy and program decisions. It is also a good example of how qualitative information, such as the affinity for architectural style or the desire to be close to the amenities of the central business district, can be combined with quantitative data, such as the percentage of housing built before 1950 and the measured distance of a neighborhood from the central business district, to produce usable information on community change. Although several revisions could possibly improve the model, it provides an excellent foundation for future research into the development of more effective models relating to monitoring gentrification as well as a wide range of other related planning issues.

PAGE 73

APPENDIX A DATA TABLES Regional to Local Comparison Indicators Table A-1: Total population Area 1980 1990 2000 Change (-) St. Petersburg 238647 238629 248232 4.02% Bartlett Park 4827 4269 3912 -18.96% Old Southeast 2625 2775 2538 -3.31% Roser Park 2302 1349 1128 -51.0% Uptown 2250 2207 2034 -9.6% Crescent Lake 3847 3724 3811 -0.94% Table A-2: Housing units Area 1980 1990 2000 Change (-) St. Petersburg 119486 125452 124618 4.3% Bartlett Park 2256 2261 1844 -18.26% Old Southeast 1459 1380 1305 -10.56% Roser Park 1541 591 332 -78.48% Uptown 1414 1259 1062 -24.84% Crescent Lake 2821 2759 2359 -16.38% Table A-3: Professional job employment (as defined by US Census) Area 1980 1990 2000 Change (-) St. Petersburg 23.96% 26.67% 34.05% 10.09% Bartlett Park 9.27% 7.36% 14.47% 5.20% Old Southeast 23.93% 25.03% 33.75% 9.82% Roser Park 9.28% 15.67% 29.10% 19.82% Uptown 23.96% 15.00% 32.32% 8.36% Crescent Lake 15.86% 24.32% 31.95% 16.09% Table A-4: College-educated population (bachelors degrees or higher) Area 1980 1990 2000 Change (-) St. Petersburg 14.57% 18.63% 22.82% 8.25% Bartlett Park 5.13% 6.02% 6.02% 0.89% Old Southeast 17.29% 28.43% 29.59% 12.30% Roser Park 6.08% 6.86% 17.93% 11.85% Uptown 9.22% 13.95% 14.06% 4.84% Crescent Lake 13.85% 16.85% 19.97% 6.12% Table A-5: Age 25 through 34 Area 1980 1990 2000 Change (-) St. Petersburg 13.02% 14.96% 13.76% 0.74% Bartlett Park 19.81% 16.34% 13.62% -6.19% Old Southeast 16.11% 18.27% 12.33% -3.78% 61

PAGE 74

62 Table A-5 Continued Area 1980 1990 2000 Change (-) Roser Park 14.81% 20.24% 9.57% -5.24% Uptown 15.96% 19.80% 16.91% 0.95% Crescent Lake 14.04% 20.62% 17.90% 3.86% Table A-6: Age 55 through 64 Area 1980 1990 2000 Change (-) St. Petersburg 12.15% 10.86% 9.17% -2.98% Bartlett Park 4.56% 7.33% 8.28% 3.72% Old Southeast 9.79% 8.43% 10.17% 0.38% Roser Park 7.91% 6.89% 6.74% -1.17% Uptown 9.42% 7.70% 7.37% -2.05% Crescent Lake 11.62% 7.00% 7.71% -3.91% Table A-7: Area Median income (AMI in dollars) Area 1980 1990 2000 Change (-) St. Petersburg 11798 23577 34597 193% Bartlett Park 8135 13224 19125 135% Old Southeast 10386 25047 31163 200% Roser Park 7584 11505 19531 158% Uptown 8466 16824 22768 169% Crescent Lake 6964 15846 23225 234% Table A-8: Single-family home value (dollars) Area 1980 1990 2000 Change (-) St. Petersburg 35800 62700 81000 126% Bartlett Park 20600 37200 45800 122% Old Southeast 37900 70700 85400 125% Roser Park 19200 45000 68100 255% Uptown 29000 52500 78200 170% Crescent Lake 28700 59300 89200 211% Table A-9: Mean commute time (minutes) Area 1980 1990 2000 Change (-) St. Petersburg 19.5 19.2 20.6 5.64% Bartlett Park 21.5 21.2 21.8 1.39% Old Southeast 17.5 19.4 21.8 24.57% Roser Park 22.2 19.7 23.8 7.12% Uptown 14.8 17.3 20.4 37.84% Crescent Lake 19.4 22.1 20.75 6.96% Table A-10: Housing vacancy Area 1980 1990 2000 Change (-) St. Petersburg 9.76% 15.74% 12.00% 2.24% Bartlett Park 17.02% 28.79% 28.77% 11.75% Old Southeast 15.97% 14.93% 13.56% -2.42% Roser Park 29.46% 36.72% 29.82% 0.36% Uptown 15.91% 21.84% 19.68% 3.77% Crescent Lake 14.82% 31.53% 21.49% 6.67%

PAGE 75

63 Table A-11: Owner-occupied housing Area 1980 1990 2000 Change (-) St. Petersburg 57.04% 53.07% 55.87% -1.17% Bartlett Park 36.92% 29.72% 34.76% -2.16% Old Southeast 51.41% 47.17% 53.95% 2.54% Roser Park 14.15% 14.38% 23.79% 9.64% Uptown 31.90% 28.28% 32.58% 0.68% Crescent Lake 25.81% 21.89% 27.04% 1.23% Table A-12: Rooms (median number for owner-occupied units) Area 2000 St. Petersburg 5.5 Bartlett Park 5.3 Old Southeast 6 Roser Park 7.4 Uptown 5.2 Crescent Lake 5.5 Neighborhood-Specific Indicators Table A-13: Housing pre-1950 Area 2000 Bartlett Park 41.16% Old Southeast 44.08% Roser Park 42.17% Uptown 57.47% Crescent Lake 56.04% Table A-14: Proximity to central business district Area 2000 Bartlett Park 1 Old Southeast 1.5 Roser Park 0 Uptown 0 Crescent Lake 0 Table A-15: Proximity to transportation corridors (interstate highways) Area 2000 Bartlett Park 1 Old Southeast 1.5 Roser Park 0 Uptown 0 Crescent Lake 0 Table A-16: Historical designations Area 2000 Bartlett Park 0 Old Southeast 4 Roser Park 1

PAGE 76

64 Table A-16 Continued Area 2000 Uptown 0 Crescent Lake 2

PAGE 77

APPENDIX B AREA MAPS CRESCENT LAKE DOWNTOWN OLD SOUTHEAST BARTLETT PARK ROSER PARK UPTOWN Figure B-1: Neighborhoods Source: Yahoo! Maps ( http://maps.yahoo.com ) 65

PAGE 78

66 235 234 DOWNTOWN 213 205 204 Figure B-2: Census Tracts: Source: Florida Geographic Data Library (www.fgdl.org)

PAGE 79

APPENDIX C GENTRIFICATION INDEX 235 234 213 205 204 Figure C-1: Gentrification index Index Value Range: Census Tracts: 204 = Old Southeast 205 = Bartlett Park 213 = Roser Park 234 = Uptown 235 = Crescent Lake 67

PAGE 80

LIST OF REFERENCES Anselin, Luc. 1998. GIS Research Infrastructure for Spatial Analysis of Real Estate Markets. Journal of Housing Research 9(1): 113-133. APD, Inc. 2002. The Fifth Avenue/Pleasant Street Neighborhoods: Controlling Gentrification. City of Gainesville, FL Community Redevelopment Agency. Atkinson, Rowland. 2003. Introduction: Misunderstood Saviour or Vengeful Wrecker? The Many Meanings and Problems of Gentrification. Urban Studies 40(12): 2343-2350. Berry, Brian J. L. 1999. Comments on Elvin K. Wyly and Daniel J. Hammels Islands of Decay in Seas of Renewal: Housing Policy and the Resurgence of Gentrification Gentrification Resurgent? Housing Policy Debate 10(4): 783-788. Birkin, Mark and Graham Clark. 1998. GIS, Geodemographics and Spatial Modeling in the U.K. Financial Service Industry. Journal of Housing Research 9(1): pp. 87-111 Can, Aayse. 1998. GIS and Spatial Analysis of Housing and Mortgage Markets. Journal of Housing Research 9(1): 61-86. Chapple, Karen, John V. Thomas, Dena Belzer, and Gerald Autler. 2004. Fueling the Fire: Information Technology and Housing Price Appreciation in the San Francisco Bay Area and the Twin Cities. Housing Policy Debate 15(2): 347-383 Clark, Gordon, Andrew Goudie, and Ceri Peach. 1996. The New Middle Class and the Remaking of the Central City. New York, NY: Oxford University Press Inc. Downs, Anthony. 2002. Have Housing Prices Risen Faster in Portland Then Else-where? Housing Policy Debate 13(1): 7-31 Freeman, Lance, and Frank Braconi. 2004. Gentrification and Displacement: New York City in the 1990s. Journal of the American Planning Association 70(1): 39-52. Galster, George C., Roberto G. Quercia, and Alvaro Cortes. 2000. Identifying Neighborhood Thresholds: An Empirical Exploration. Housing Policy Debate 11(3): 701-732. Goodman, Allen C., and Thomas G. Thibodeau. 1995. Age-Related Heteroske-dasticity in Hedonic House Price Equations. Journal of Housing Research 6(1): 25-42 68

PAGE 81

69 Higgins, Lindley. 2001. Gathering and Presenting Information About Your Neighborhood. Washington, D.C.: Local Initiatives Support Corporation Center for Home Ownership and Knowledge Sharing Initiative. Kennedy, Maureen, and Paul Leonard. 2001a. Dealing With Neighborhood Change: A Primer on Gentrification and Policy Choices. Washington, D.C.: Brookings Institution. Kennedy, Maureen, and Paul Leonard. 2001b. Gentrification: Practice and Politics. Washington, D.C.: Local Initiatives Support Corporation Center for Homeownership and Knowledge Sharing Initiative. Land, Robert E., James W. Hughes, and Karen A. Danielson. 1997. Targeting Suburban Urbanites: Marketing Central-City Housing. Housing Policy Debate 8(2): 437-470. Lang, Michael H. 1982. Gentrification Amid Urban Decline: Strategies for Americas Older Cities. Cambridge, MA: Ballinger Publishing Company. LaPeter, Leonora. July 18, 2004. A Revitalization Riddle. St. Petersburg Times. Section B: pp. 1, 6 Nelson, Kathryn P. 1988. Gentrification and Distressed Cities: An Assessment of Trends in Intrametropolitan Migration. Madison, WS: University of Wisconsin Press. Peng, Ruijue, and William C. Wheaton. 1994. Effects of Restrictive Land Supply on Housing in Hong Kong: An Econometric Analysis. Journal of Housing Research 5(2): 263-291. Phillips, Rhonda. 2003. Community Indicators. Washington, D.C.: American Planning Association Public Advisory Report Number 517. Quercia, Robert G., and George C. Galster. 1997. Threshold Effects and the Expected Benefits of Attracting Middle-Income Households to the Central City. Housing Policy Debate 8(2): 409-435) Redfern, P. A. 2003. What Makes Gentrification Gentrification? Urban Studies 40(12): 2351-2366. Smith, Neil, and Peter Williams. 1986. Gentrification of the City. Boston, MA: Allen & Unwin. Solnit, Rebecca, and Susan Schwartzenberg. 2000. Hollow City: The Seige of San Francisco and the Crisis of American Urbanism. New York, NY: Verso. Turner, Margery Austin. 1997. Achieving a New Urban Diversity: What Have We Learned? Housing Policy Debate 8(2): 295 305

PAGE 82

70 Williams, Brett. 1988. Upscaling Downtown: Stalled Gentrification in Washington, D.C. Ithaca, NY: Cornell University Press. Wyly, Elvin K., and Daniel J. Hammel 1999. Islands of Decay in Seas of Renewal: Housing Policy and the Resurgence of Gentrification. Housing Policy Debate 10(4): 711-771.

PAGE 83

BIOGRAPHICAL SKETCH Ashon Jahi Nesbitt originates from St. Petersburg, FL the area of focus for the study in this paper. He spent his entire childhood there before going on to attend Florida Agricultural and Mechanical University, where he majored in Architecture and participated in the world-renowned Marching 100 as well as gained other campus activities. Ashon graduated from Florida A&M University in the spring of 2002 with a Bachelor of Science in Architectural Studies. Although Ashon sought to pursue a professional degree in architecture, he found his home in Urban and Regional Planning at the University of Florida after a year of unsuccessful attempts to gaining employment in the field of architecture. Ashon chose to concentrate on Housing and Economic Development. Ashon first became interested in this area due to exposure to his moms professional career, who worked many years in real estate and as director of a leading local nonprofit housing agency in the city of St. Petersburg. As a student in the Urban and Regional Planning program at the University of Florida, Ashon has cultivated that interest through coursework, employment as a Graduate Research Assistant with the Center for Building Better Communities, and attendance at such conferences as the Florida Housing Coalition Annual Conference. In addition to his academic pursuits, Ashon actively participated in the Student Planning Association, serving as the President for the 2004-2005 school year. He also 71

PAGE 84

72 served on the Florida Chapter of the American Planning Associations (APA) Student Council for that year, Student Representative on the San Felasco APA Executive Committee and holds memberships with the American Planning Association and Florida Housing Coalition. Beyond the department, Ashon participated in the Black Graduate Student Organization, serving as Vice-President for the 2004-2005 school year, as well as church and other activities throughout the community. Ashon hopes his educational and professional experiences will land him a position with the Department of Housing and Urban Planning, where he hopes to hold the top position one day. Ashon ultimately plans to obtain a Ph.D. in Public Policy, become a developer, focusing on urban infill, affordable housing developments and to teach at the university level upon retirement.


Permanent Link: http://ufdc.ufl.edu/UFE0010582/00001

Material Information

Title: A Model of Gentrification: Monitoring Community Change in Selected Neighborhoods of St. Petersburg, Florida Using the Analytic Hierarchy Process
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010582:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010582/00001

Material Information

Title: A Model of Gentrification: Monitoring Community Change in Selected Neighborhoods of St. Petersburg, Florida Using the Analytic Hierarchy Process
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010582:00001


This item has the following downloads:


Full Text












A MODEL OF GENTRIFICATION: MONITORING COMMUNITY CHANGE IN
SELECTED NEIGHBORHOODS OF ST. PETERSBURG, FLORIDA USING THE
ANALYTIC HIERARCHY PROCESS















By

ASHON J. NESBITT


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN URBAN AND REGIONAL PLANNING

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Ashon J. Nesbitt















To all housing and economic development planners















ACKNOWLEDGMENTS

First, I acknowledge the Most High for extending His grace, and giving me the

knowledge, passion and strength to complete this study. Secondly, I acknowledge my

committee members. I thank Dr. Kristin Larsen for all of her advice, assistance and

encouragement throughout the completion of this study. Her passion for housing has truly

affected me, and caused me to take even greater interest in this area of planning. Also,

her skills and talents as an author, mentor, professor, planner, researcher and scholar are

admirable and have benefited me immensely in this process. I thank Dr. Paul Zwick for

his expertise and his belief in me as a scholar. I thank Dr. Rhonda Phillips for providing

me with exclusive access to her publications and for her financial support through the

Center for Building Better Communities. I thank Dr. Marc Smith for balancing my views

on housing and pointing me to other sources of information and research I wouldn't have

otherwise thought to look in. I thank them all for your input and support.

In addition, I would like to thank Karen Freggens, database manager with the City

of St. Petersburg Building Department, Connie Clark, Secretary with the City of St.

Petersburg Budget Office, and the Data Processing department of the Pinellas County

Property Appraiser. I thank them for working with me and providing me with important

data for this study.

Finally, I acknowledge my family, close friends, and all of my fellow students who

provided input or gave an encouraging word. I thank them for their kindness and love. It

did not go unnoticed.









This thesis is more than a demonstration and testament of my knowledge of an

Urban Planning issue and my ability to conduct scholarly research. It is also a reflection

of all who played a role--big or small--in the process of completing this study. I thank

them again, and I hope they benefit from it, as much as I have benefited from them.
















TABLE OF CONTENTS

page

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

LIST OF TABLES ......................................... ......... ............ ........... .. viii

LIST OF FIGURES ................................. ...... ... ................. .x

ABSTRACT ........ .............. ............. ...... .......... .......... xi

CHAPTER

1 INTRODUCTION .............. .....................................................1..

2 LITER A TU R E R EV IEW ................................................................ ........................ 3

Gentrification ................. .............. .......... ........................... 3
O rigin an d Introdu action .............................................................. .....................3
L location and Scale ....................................................... 4
W ho and W hy ......................................... ......... ......... .. ........ .. ..
D isplacem ent .................................................................. ........................... . 9
R researcher's D definition .................................................................. ............... 10
Indicators .............................. ................. ........... ................. ............. 11
D definition and A applications ......................................................... ............... .. 11
G entrification Indicators.............................................. ............................ 15
Thresholds ........................................................... ... ........ 16
Geographic Information Systems (GIS) ...................................... .............18
In tro d u ctio n ..................................................................... 1 8
D e fin itio n ...................................................... ................ 1 8
Functions and A applications ....................................................... .... ........... 18
S u m m a ry ............................................................................................................... 2 1

3 S T U D Y A R E A ........................................ ........................................... ................ .. 2 3

4 M E T H O D O L O G Y ............................................................................ ......................25

E explanation of M odel ............................................................................ ...... 25
Identifying the Indicators ............................................... ............................ 26
D ata C o llectio n ...............................................................................................2 8
Developing the Equation .............................................. 32










W eighted Suitability M odel ............................................................................34

5 FIN D IN G S A N D R E SU L T S ........................................ ...........................................37

R regional to Local Com prisons ........................................ ........................... 37
Professional Job G row th.......................................................... ............... 37
Change in Population ........................ .......... ............... ..... ... ....38
C change in H housing U nits................................................ .......... ............... 38
Change in College-Educated Population.................................. ............... 39
Change in Age 25 through 34 Population ................................. ............... 40
Change in Age 55 through 64 Population ................................. ............... 40
Change in A rea M edian Incom e................................... .................................... 41
Change in Median Single-family Unit Value....................................................42
Change in H housing V acancy ........................................ .......................... 43
Change in Owner-Occupancy.................... ....... .......................... 43
U n it S iz e ............................................................4 4
Change in Commute Times ...... .............................................. .. ............... 45
Neighborhood-Specific Indicators......... ........... ........................45
Percentage of Housing Constructed before 1950 ........................ .......... 45
Proximity to the Central Business District and Interstate Highways ..................46
H historic D esignations .................................. ................. ..... ....... 46
M aj or R relationships .................. ...................................... .. ............ 47
R e su lts ...........................................................................................4 9
W eights ...................................................................................................... 49
V a lu e s ......................................................................................................5 0

6 CONCLUSION ............... ................. ................................... ... 55

U universal A pplicability............... ................................ .. .. ...... .. .. ..... .....55
P olicy Im plications ..................... .. ............................................ .... .... . ...........56
Recommendations for Future Research .......................................... ...............57

APPENDIX

A D A TA TA BLE S .......................... ..................... .. .. .. .. ...... ........... 61

Regional to Local Com prison Indicators.............................................................. 61
Neighborhood-Specific Indicators......... ..........................................................63

B AREA MAPS ..................... ....... ... ...........................65

C GEN TRIFICA TION INDEX ................................................ ............................ 67

L IST O F R E FE R E N C E S ....................................................................... ... ................... 68

B IO G R A PH IC A L SK E T C H ..................................................................... ..................71
















LIST OF TABLES

Table page

4-1: Regional to neighborhood comparison indicators.................................................26

4-2: N eighborhood-specific indicators ........................................ ......................... 28

4-3: Sources for regional to neighborhood comparison indicators............... ................ 29

4-4: Sources for Neighborhood-specific indicators .......................................................29

4-5: Pairw ise com prison m atrix ............................................... ............................ 33

4-6: Pairwise comparison matrix value pattern ...................................... ............... 34

5-1: Regional to neighborhood comparison indicators.....................................................50

5-2: N eighborhood-specific indicators ........................................ ......................... 50

A -i: T otal population ............ .... ............................................................ ................... 26 1

A-2: Housing units .................................................................... ........... 61

A -3: Professional job em ploym ent ....................................................... ............... 61

A -4: College-educated population ......................................................... ............... 61

A -5: A ge 25 through 34 ...................................................... ......... .. ............ 61

A -6: A ge 55 through 64 ...................................................... ......... .. ............ 62

A -7 : A rea m edian income e ................................................................................ ...... ...62

A -8: Single-fam ily hom e value............................................................... .....................62

A -9 : M ean com m ute tim e ................................................................................ ...... ...62

A -10 : H ou sing v acancy ........... ............................................................................ .. ....... .. 62

A 1 : O w ner-occupied housing ........................................................................... .... ... 62

A -12 : R o o m s ................................................................................. 6 3



viii









A-13: H housing pre-1950 ........................................... ............ ............ 63

A-14: Proxim ity to central business district ...................................... ........ ............... 63

A-15: Proximity to transportation corridors ........................................... ............... 63

A -16 : H historic designation s ....................................................................... ..................63
















LIST OF FIGURES

Figure page

B-l N neighborhoods ......................................................... .. ........ .... 65

B-2 Census Tracts ....... ...................... .. ............ ........ ... .. .. ............ 66

C-1 Gentrification Index ................................................... .. ... .. ........ 66















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts in Urban and Regional Planning

A MODEL OF GENTRIFICATION: MONITORING COMMUNITY CHANGE IN
SELECTED NEIGHBORHOODS OF ST. PETERSBURG, FLORIDA USING THE
ANALYTIC HIERARCHY PROCESS

By

Ashon J. Nesbitt

May 2005

Chair: Kristin Larsen
Cochair: Rhonda Phillips
Major Department: Urban and Regional Planning

Gentrification has emerged as a major issue in urban and regional planning,

particularly in the central cities of large metropolitan areas. As more middle-class and

upper-class residents begin to choose city life and reject suburban living, many older

neighborhoods, once occupied exclusively by very-low income and low-income

residents, are being re-inhabited by more affluent residents. Research on this topic is

extensive, and several researchers have come to the same conclusions on the indicators of

gentrification and the characteristics of the gentrifyer. However, there have been few

attempts to develop methods to identify neighborhoods more likely to gentrify and

monitor change in neighborhoods toward gentrification, which would allow planners and

policy-makers to be proactive in their approach to preventing many of the negative

affects of gentrification.









In our study, we developed a model for monitoring gentrification based upon the

indicators of gentrification identified in previous studies on the subject. The model uses

St. Petersburg, FL as its base region and identifies four neighborhoods as potential areas

of gentrification. The model uses statistics derived mostly from census data and converts

them into spatial data using geographic information systems, and calculates a

gentrification index based upon the indicators it identifies as most important to

identifying gentrification.

We found that two of the neighborhoods are indeed more likely to gentrify, and

perhaps the process has already begun. Two neighborhoods may be likely to gentrify in

the near future; while one may be likely in the distant future. The results of the analysis

and gentrification index suggest policy changes and program implementation. Moreover,

our study demonstrates that indicators, statistical analysis and the spatial analysis

capabilities of geographic information systems can be used to identify complex planning

issues and monitor community change related to those issues so that appropriate policy

responses can be established.














CHAPTER 1
INTRODUCTION

As urban development returns to formerly distressed neighborhoods, gentrification

emerges as a significant planning issue. Much of the discussion and research on this issue

relates to affordable housing in general, and the plight of very-low, low and moderate-

income families in terms of housing options for these income groups. As more middle-

and upper-class households choose urban instead of traditional suburban living, how can

cities maintain affordability for lower-income households that do not possess the

financial resources to allow them to choose where to live? Planners and researchers

continue to struggle with solutions to this problem.

While dealing directly with the affordable housing issue and striving to solve such

a complex problem, planners and researchers have learned much about gentrification.

They know much about the profiles of these middle and upper-income households that

would potentially choose urban, or central city, living over suburban living. They also

know the attributes these households look for in urban neighborhoods. In addition,

research on gentrification identifies the major indicators of gentrification and establishes

a basic understanding of each indicator in determining gentrification. However, with all

of this knowledge, very few studies have sought to create a method of synthesizing

quantifiable data related to these indicators in order to identify neighborhoods likely to

gentrify and to monitor community during, and even prior to, the gentrification process.

Our aim was to develop such a method by applying community indicators, the analytic

hierarchy process and weighted suitability modeling. Thus, proper steps can be taken by









planners and policymakers to mitigate the negative effects of gentrification before the

process occurs.

Developing the model involved several steps. First, we reviewed the current body

of literature on gentrification to determine its major indicators. We examined information

on community indicators and their application to planning as well as methods of spatial

analysis and deterministic modeling currently available, yet typically unused in the field

of housing planning. Second, we examined background information on St. Petersburg,

Florida, the test city, and the five neighborhoods in St. Petersburg to justify the use of this

area and to demonstrate implementation of the model. Finally, we discussed the findings

related to each indicator; outcomes of the model; overall applicability of the model and

recommendations for improvements and future research.

Our study focused on identifying gentrification specifically. We also intended to

demonstrate a useful application of spatial analysis and generate discussion and further

research into its use to create a more proactive culture in the field of urban and regional

planning as opposed to the reactive means of operation that presently characterizes much

of professional practice. Geographic modeling can be a powerful tool in planning and

policymaking. Our study demonstrated its particular usefulness in housing planning, and

how indicators and spatial analysis can be applied to a real planning issue.














CHAPTER 2
LITERATURE REVIEW

Our study assessed three planning issues often considered separately. Specifically

associated with housing, these issues are that of gentrification, community indicators and

applications of the spatial analysis capabilities of geographic information systems (GIS).

Much has been written in planning journals and other related publications about all three

subjects. Researchers and practitioners continue to disagree on the true meaning of

gentrification. Several articles and books have been written on the effectiveness of

indicators in determining a community's economic direction. The application of GIS to

community, housing planning research and practice is still in its infancy; however,

researchers and practitioners are beginning to look for ways to use this powerful software

to examine such planning activities.

Gentrification

Origin and Introduction

According to Atkinson (2003), Ruth Glass originated the term gentrification in the

United Kingdom in 1964. The word is derived from "gentry", referring to the middle and

upper class households that are "seen to displace local working-class groups". According

to Glass, this displacement causes a change in the area. This change is the action referred

to in the term "gentrification", or the process of becoming a place for the gentry. This

urban phenomenon has been studied and analyzed for forty years, since the inception of

the term. Many definitions and ideas as to the causes of gentrification have been

presented and debated over time. In this section, these definitions and ideas will be









explored and discussed. Throughout the discussion, recurring themes, as well as key

points most relevant to our study will be highlighted. This section will conclude with a

definition of gentrification framed by the researcher.

Location and Scale

Perhaps a good place to begin a discussion of gentrification would be to define

where it occurs and at what scale. According to the literature, gentrification is defined as

an urban phenomenon,1 occurring in large metropolitan areas. Most of the studies on

gentrification have been done in large cities, and the process was first observed in

London. In the United States, studies have been done on such cities as New York,

Boston, Washington, DC, San Francisco, Atlanta, and Cleveland, Ohio. Further,

gentrification is typically attributed to central cities. However, there are cases in which

older suburbs in large metropolitan areas are experiencing change often associated with

gentrification. Examples of this are Vallejo and East Palo Alto, California (Kennedy and

Leonard 2001a). In addition, questions have arisen as to whether gentrification is truly

limited to large metropolitan areas. Could gentrification also occur in smaller cities? A

study done by the City of Gainesville, Florida Community Redevelopment Agency looks

at the possibility of gentrification occurring there in an economically distressed

community west of central business district known as the Pleasant Street neighborhood.

These examples challenge the notion that gentrification is only a central-city issue and

perhaps speaks to the future of gentrification studies (ADP, Inc. 2002).2



1 Some recent studies show that gentrification is also taking place in small towns and rural communities

2 As the poor are being pushed out of central cities to "inner ring" suburbs, these older suburban areas are now seen as
a possible location for future waves of gentrification.









Gentrification is a process denoted by the middle and upper class reinvesting into

the housing stock of poor inner city neighborhoods with high levels of abandonment,

disinvestment and vacancy. Although gentrification only occurs in neighborhoods with

specific attributes within cities (Gordon, Goudie and Peach 1996; Lang 1982), it is a

significant phenomenon that is happening in an ever-increasing number of cities (Wyly

and Hammel 1999; Wyly and Hammel 1999). For instance, in the United States the

"return to the city" trend, which started in the larger, older metropolitan areas,

particularly in the Northeast and Midwest, has begun to filter down to more recently

urbanized areas in the South and West. Even with gentrification occurring in more and

more locations, these neighborhoods have yet to outstrip the suburbs as the primary

residential area for the middle and upper classes. One reason for the process of

gentrification lagging behind suburban expansion is that, in most cases, neighborhoods

prone to gentrification are not large enough to meet all the housing needs of a

metropolitan area's growing middle and upper classes. Also, these areas tend to be

attractive to a certain subset of the middle and upper class population based on such

attributes as neighborhood location, urban character and architectural style.

Who and Why

Now that we know where gentrification occurs, what are the characteristics of

"gentrifyers" and why do they choose to live where they do? Although there appears to

be a general consensus on what the characteristics of gentrifyers are, there are some

differing ideas on why gentrification occurs, and why in these areas of urban decay.

Following a description of gentrifyers, this section will broadly discuss why

gentrification occurs and specifically why in these urban neighborhoods.









Since gentrification generally occurs in cities, in order to understand gentrifyers,

one should understand why people like city living. City dwellers "like the privacy...

specialization, and the hundreds of one-of-a-kind shops...the excitement...the

heterogeneity, the contrasts, the mixture of odd people." (Land, Hughes, Danielsen 1997,

p. 437). However, most people identify city dwellers as less affluent or poor. In

opposition to that perception, gentrifyers, also part of this city-dwelling population, are

generally moderate to upper income households normally associated with suburban

communities. In "What Makes Gentrification 'Gentrification'?", Redfern describes the

gentrifyer as being 'other' to the suburbanizing middle class." (Redfern 2003, p.2355)

What makes the gentrifyer different from their suburban as well as their urban

counterparts?

First, unlike other city dwellers, such as the inhabitants of public housing

complexes and working class households who cannot afford a house in the suburbs,

gentrifyers can choose where they live. Second, gentrifyers are often highly educated

professionals. Third, gentrifyers tend to be untraditional households. Gordon, Goudie and

Peach (1996) identify gentrifyers as often being young, unmarried and childless as

opposed to the typical two-parent, two-child household found in the suburbs or working

class neighborhoods for that matter. Another population of gentrifyers includes empty

nesters, those older couples or individuals who no longer have children living in the

house with them.

Other groups associated with gentrification are artists and gay and lesbian

households. Often called "urban pioneers", these are usually the first groups to move into

a deteriorating area, rehabilitate the housing, and make the area attractive again (Solnit









and Schwartzenberg 2000; Wyly and Hammel 1999). What's interesting is that these

groups often become the victims of what is called a "second gentrification" where these

"urban pioneers" having proven the worth of a neighborhood, are subsequently displaced

by investors and more affluent households. (Solnit and Schwartzenberg 2000; Wyly and

Hammel 1999)

Land, Hughes and Danielsen (1997) describe potential city dwellers, referred to in

our study as gentrifyers, in the context of the environments from which they originate.

They describe two different types of gentrifyers: "suburban urbanites" and "urban

suburbanites". These descriptions provide more insight into what gentrifyers seek in a

neighborhood based on the urban context of the metropolitan area as a whole, and will

thus help determine a neighborhood's potential for gentrification.

The "suburban urbanite" is defined as a suburban resident with a similar lifestyle to

a central-city resident. Suburban urbanites are found in the inner suburbs of Northeastern

and Midwestern cities. Cities in these regions tend to be smaller in land area, denser, and

surrounded by high-density suburbs that have "central-city-type neighborhoods." (Land,

Hughes, Danielsen 1997, p.441). Because they already live in neighborhoods that have

similar characteristics of central city neighborhoods, they are more likely to choose

central city living.

In contrast, "urban suburbanite" would most likely be found in the suburbs of

Sunbelt cities. These cities tend to be larger in land area with less dense urban cores as

well as suburban-style subdivisions within the central city. These individuals are looking

for areas that offer all of the advantages of urban living with all of the comforts of the









suburbs. Therefore, in different urban contexts, gentrifyers seek different characteristics.

The presence of these characteristics in a neighborhood affects its gentrification potential.

The distinction between "suburban urbanite" and "urban suburbanite" is an

interesting and significant one that bears importance in this particular study. The

neighborhoods in our study are located in St. Petersburg, Florida. Although it is not an

extremely expansive city geographically, its development pattern fits the Sunbelt City

mode, with its less dense urban core and suburban-style subdivisions within its city

limits. Therefore, gentrifyers in St. Petersburg would probably have the qualities of the

"urban suburbanite".

In addition to the socioeconomic status of the gentrifyer, another, perhaps more

controversial attribute of the gentrifyer is addressed in the literature race. Suburban

expansion is associated with the term "white flight", which refers to the exit of the white

population from the central city to surrounding suburban communities. Gentrification

counters this trend, with white residents returning to the city, sometimes going right back

to the same communities they fled decades past. Still, gentrifyers are not necessarily

white. For example, in certain areas of Atlanta affluent blacks are returning to the city

(Kennedy and Leonard 2001a). Therefore, although "gentrifyer" usually has a clearly

white racial identity, sometimes the term includes members of minority races.

Gentrification occurs in regions where the housing market is tight (Kennedy and

Leonard 2001a, 2001b; Lang 1982). When new housing demand outpaces the production

of new housing, the price of housing will escalate. Thus, investment in the existing

housing stock becomes an option considered by those with means (Nelson 1988, p. 15).

Typically, areas chosen for investment have the greatest opportunity for reinvestment due









to high levels of abandonment, disinvestment and vacancy. However, these attributes

don't always guarantee a high potential for gentrification. Gentrifyers also choose areas

characterized by their architectural style and high historic value of the homes as well as

location near cultural amenities and/or the traditional central business district

employment center (Lang 1982; Nelson 1988; Redfern 2001).

Because these neighborhoods are so undesirable at the time of initial investment,

the housing is cheap. In fact, Nelson (1988) argues that cheaper housing and the

perceived profitability is more important than being fashionable. The reality of the

situation most likely involves affordability, architectural style and profit.

Thus, a gentrifyer is a middle or upper class, nontraditional household that prefers

urban living. Gentrifyers are usually affluent whites, although this is not always the case.

Further, gentrification is the result of a tightening housing market, making cheap inner

city housing appear more desirable due to its affordability, profitability, location and

style.

Displacement

One major issue of debate regarding what defines gentrification involves the issue

of displacement. As more is invested in an area and property values rise, the poor and

working class households that comprise the original residential population of a

neighborhood will no longer be able to afford to stay there, resulting in displacement.

While such displacement may be of economic benefit to cities overall as the rising

property values increase the tax base (Kennedy and Leonard 2001a, 2001b), many view it









as an unavoidable, socially detrimental consequence that overburdens the original

residents, particularly renters in the neighborhood(Lang 1982, LaPeter 2004).3

Many definitions and studies of gentrification require displacement to occur in

order for an area to be declared gentrified (Kennedy and Leonard 2001a, 2001b).

However, Wyly and Hammel (1999) speak of "urban pioneers", the initial investors, as

possibly displacing the original residents and oftentimes displaced by a second group of

gentrifyers. Lang (1982) also uses the word often to describe displacement in the

gentrification process (Lang 1982, p.6). Freeman and Braconi's (2003) study of New

York found that significant displacement does not have to occur for gentrification to take

place. For instance, if the abandonment and vacancy rate is extremely high, then the

likelihood of displacement is very low. Similarly, a study done by the City of Gainesville,

Florida for its Pleasant Street neighborhood found that abandonment and vacancy were

high enough for reinvestment to occur without large numbers of residents being displaced

(ADP, Inc. 2002).

Researcher's Definition

Based upon the various characterizations of gentrification explored in previous

studies and their applicability to our study, we offer the following definition for

gentrification:

Gentrification is the process by which the socioeconomic status of a neighborhood
populated mostly by lower-income households is substantially elevated by renewed
interests and investments by higher-income households, including homebuyers,
renters and commercial interests from outside the neighborhood so as to change the
overall character of the neighborhood, and usually results in widespread

Gentrification changes the character of a neighborhood. The new middle and upper income residents not only
upgrade the housing stock, they also bring with them new consumer demands, which affect area amenities, such as
public spaces and retail offerings. Sometimes businesses are displaced as well as residents. However, this study has a
residential focus.









displacement of the lower-income residents already living in the neighborhood as
well as the businesses they support.

This definition includes the social as well as economic implications of

gentrification. It also addresses both the residential and commercial aspects of

gentrification. Although our study and previous studies on the subject tend to focus on

the residential, the commercial component of gentrification is worth mentioning in any

definition or discussion.

Indicators

Often used in community planning and economic development planning,

community indicators evaluate social and economic change in an area. Different types of

indicators function on different scales. Gentrification definitely has economic

ramifications, thus certain types of indicators are typically present when it is occurring or

likely to occur in a given area. This section defines indicators and outlines those relevant

to gentrification. These specific indicators will become the basis of the gentrification

model.

Definition and Applications

Phillips (2003) defines indicators as "measurements that provide information about

past and current trends to assist planners and community leaders in making decisions that

effect outcomes" (p.1). These measurements quantify the social, environmental and

economic factors that work together to create change in a community or region. She

describes them as "gauges" that document how much progress is being made toward

reaching a certain goal or to show what a community or region is likely to become

according to data gathered on the indicators. According to Hart (2003) and Oleari (2000),

combining several indicators together to create a "measuring system", or model, can









"provide (useful) information about past trends, current realities and future direction in

order to aid decision making" (quoted in Phillips 2003, p.2).

Two basic types of indicators are defined in the literature. They are system

(descriptive) indicators and performance indicators. System indicators condense

individual measurements that describe multiple characteristics of a specific system in

order to communicate the most pertinent information to decision-makers (Phillips, 2003;

Hardi et al. 1997). System indicators work best with painting a picture of the current state

of a system and are used to guide policy writing. Performance indicators are similar to

system indicators in that they are both descriptive. However, performance indicators are

also "prescriptive". This type of indicator has a goal, reference value or target attached to

it and measures how much progress is being made toward reaching that goal or target.

Performance indicators are good for policy or program evaluation; therefore, these

indicators can guide policy or program changes. Our study accurately describes the

current situation in a neighborhood and assesses where the neighborhood is headed if the

current trends continue, which will guide decision-making and policy writing. Therefore,

performance indicators are most appropriate for our study.

Indicator studies comprise three basic categories: economic, environmental and

social. Indicators are most often employed in economic studies, which is what our study

is. Of course, environmental studies assess ecosystems. An example of a social indicator

study is the School Readiness Pilot Study for a Social Infrastructure Network completed

by the Hillsborough County Planning Commission in 2003. This study measured several

indicators derived from research in the field of education, and formulated a model that

determines the likelihood of school readiness in neighborhoods throughout Hillsborough









County, Florida. Although it is a social study, it provides a helpful example of how to use

indicators in building a model for monitoring a community.

Another important aspect of indicators is their scale. Phillips (2003) defines four

levels of indicators in her publication. They are national and multinational, regional,

local, and neighborhood indicators. National and multinational indicators measure trends

on a national or international level. Regional indicators may exist on many different

levels, as regions are defined in different ways. A region could be one state or a large

section of a state, encompassing many different cities, towns and metropolitan areas. It

could be a group of states, or it could be just one metropolitan area. Therefore, the scope

of regional indicators is defined based on how the region is defined. Local indicators deal

with specific municipalities. However, they assess the municipality holistically. Just like

regional indicators, local indicators have varying scopes. They could be for one small

town, a large city or an entire county. Neighborhood indicators look at the conditions in

individual neighborhoods within cities or towns. For our study, regional to local

comparisons as well as neighborhood-specific indicators will be used to develop the

model.4

In order to build a model that produces meaningful results, the proper indicators

must be used. Phillips (2003) lists several criteria for the successful selection of

indicators. Those criteria are: validity, relevance, consistency and reliability,

measurability, clarity, comprehensiveness, cost-effectiveness, comparability and

attractiveness to the media. Validity involves insuring the indicator is based on accurate

data. Relevance is making sure the indicator relates directly to the issue at hand.

4More specifics on the indicators and their justifications will be given in the Methodology chapter of this thesis.









Consistency and reliability relate to the ability to collect the same quality of data over a

period of time. Measurability addresses the ability of the indicator data to be collected

directly from the neighborhood, locality, region or nations) being studied.5 Clarity

concerns how well the indicator is understood. Comprehensiveness measures the ability

of one indicator to cover a wide range of issues yet retain the focus of the overall model.

Cost-effectiveness reflects how much money (or time) must be put into collecting the

data. Comparability involves how effectively the indicators can be used in different

communities. Attractiveness to the media deals with how well the indicators and model

are accepted by the press.6

Although the aforementioned criteria are important in selecting indicators for

monitoring community change, Phillips (2003) states that the true test of the success of

an indicator or a model is whether or not the data collected in relation to that indicator or

the results of the model prompt government officials to take action. However, out of all

the criteria previously discussed, perhaps the most emphasis should be placed on the

validity or accuracy of the data. In order for proper action to take place, the data

associated with the indicators must be accurate. Indicators and models can then produce

meaningful information that decision-makers can work with to affect proper change.

Producing results that support proper shifts in policies and programs is the aim of our

study.




Lindley Higgins' "Gathering and Presenting Information About Your Neighborhood" published in 2001 by the Local
Initiatives Support Corporation provides useful advice on collecting data (how and where).


6 In this case, the "press" would be journals and other respected publications.









The use of indicators has a strong foundation in economic development planning

and research. Most applications have targeted sustainable development, which is defined

as development that seeks to meet the needs of the present without compromising the

needs of the future. Most indicator projects evaluate community progress. However,

indicators research presents very little on how individual indicators can be evaluated

together to monitor community change. Our study creates a model for monitoring

gentrification that involves the use of several indicators evaluated together.

Gentrification Indicators

The literature describes several indicators of the likelihood of gentrification. Some

are regional; others are local or relevant at the neighborhood level. Further, gentrification

is "notoriously difficult to measure and the results (of the model) are sensitive to the

indicators chosen", the time periods over which the indicators are measured and how

neighborhoods are defined (Wyly and Hammel 1999, p. 726).

Kennedy and Leonard (2001 a) identify rapid job creation, a regional indicator, as

the most significant indicator of potential gentrification. Rapid job creation provides

more opportunity for those already living in the region as well as attracts new residents.

Second on the list comes the supply of housing units in relation to demand. As more

residents move to an area and current residents earn higher incomes, the demand for

housing increases. If the current supply of housing cannot meet the demand, then housing

prices will increase to curb demand. Thus, cheaper inner city housing becomes a viable

alternative to more expensive, suburban housing. Other regional or local indicators

include increased commute times, growth in certain population groups and nontraditional

households and public investments. At the neighborhood level, the historic value of the

housing stock, level of abandonment and percentage of owner-occupied housing are all









indicators. For our study, these indicators and several others were chosen based on the

literature. They will be identified and explained in the Methodology chapter of this paper.

Thresholds

Galster, Quercia and Cortes (2003) define "threshold" as the critical value of an

indicator that triggers more rapid change. Another way to view a threshold is the point

when change is completely apparent and cannot be easily stopped or reversed.

Knowledge of the correct indicators is important to monitoring community change. Just

as important is knowledge of the threshold related to each indicator. Thresholds are not

arbitrary values. Accuracy in determining the threshold value plays a huge role in

determining the success or failure of a model for monitoring change.

Quercia and Galster (1997) describe four aspects of thresholds: geographic scale,

absolute or relative impacts, time of impacts and pattern of relationship. Geographic scale

is the area over which each variable is measured, and the corresponding threshold applies

at that geographic scale. For instance, the threshold for a regional indicator should apply

in the same manner throughout the region; whereas, the threshold for a local indicator

will only apply to that specific locality. Absolute or relative impacts reflect, respectively,

thresholds measured by absolute numbers or by percentages. For example, does the

growth in the number of people from the ages of twenty-five through thirty-four have to

increase by ten thousand in order to indicate change, or does it have to increase by ten

percent? Time of impact addresses whether change has to continue for a certain period of

time before rapid change occurs. For instance, does job growth have to continue for a

certain number of years before there is a surge of interest in companies wanting to add

jobs to an area? Finally, observing a pattern of relationship helps determine how the

threshold of each indicator relates to those of other variables. For instance, how does job









growth relate to population growth? Do job growth and population growth increase at the

same rate all the time? Or, is there some point when jobs are increasing at such a rate as

to cause an exponential increase in population from in migration? Is this job growth rate

related to a rapid decrease in housing vacancy in the same manner as it relates to

population growth? Data on each indicator should be tested against all other variables to

determine the best value for each threshold.

Several articles have been written on thresholds that relate to the study of

gentrification Quercia and Galster (1997) determine that there is a threshold of middle-

class households that must be reached before significant benefits, such as increased

property values and retail demand. Downs (2002), Peng and Wheaton (1994) study the

effects of restrictive land supply on housing prices, finding the point at which the amount

of developable land available begins to effect housing price; however, housing output

remains fairly constant. Chapple et al. (2004) study the effects of job growth on housing

prices, finding that rapid job growth (particularly in certain industries) begins to effect

housing prices over a certain period of time in certain locations depending on the

structure of the metropolitan area.7 The last example of threshold-related literature is

Goodman and Thibodeau (1995) who found that the relationship between the age of

housing units and price is a nonlinear relationship. All of these examples demonstrate that

thresholds exist, they are very specific, they vary by indicator, and they possibly vary by

location. Therefore, gentrification can be measured by the value of each indicator in

relation to its threshold.


Growth in industries with the potential for rapid expansion, such as technology-based industries, could indicate the
potential for a high rate of job creation over a short period of time in a region, creating new wealth and drawing new
residents at a rapid pace. This results in a tightening housing market, leading to gentrification.









Geographic Information Systems (GIS)

Introduction

Due to its spatial applications and analysis capabilities, a geographical information

system (GIS) is a critical component of our study. The following paragraphs define what

GIS is, examine the functions of GIS, and review how GIS has and can be used in real

estate research. Some of this information is similar to the material presented on

indicators. These overlaps will also be highlighted.

Definition

Luc Anselin (1998) defines GIS as "a powerful set of tools for collecting, storing

retrieving at will, transforming and displaying spatial data from the real world for a

particular set of purposes" (p. 116). Most people associate GIS with specific software

packages. Generally, GIS synthesizes value information with locational and topological

information into a searchable database. Value information, or attributes, include the price

or size of a housing unit. Locational and topological information include the address or

census block where the unit is located.

Functions and Applications

Anselin (1998) also outlines the four major functions of GIS: input, storage, output

and analysis. Of the four functions, analysis, or spatial analysis, is the focus of our study.

Spatial analysis has four sub-functions. They are selection, manipulation, exploration and

confirmation. Selection involves obtaining information relating to certain variables

specific to a certain location from a spatial database. Data manipulation involves the

creation of spatial data and is done through attribute values (averaging, summation),

spatial information (coordinates) and data integration (combination of attribute values

and spatial information).









The next two capacities of spatial analysis are exploration and confirmation. These

two are considered the heart of spatial analysis. Exploration, or exploratory spatial data

analysis (ESDA) is described as being a body of techniques used to "describe and

visualize spatial distributions", find patterns of association (spatial clustering), identify

extremely unique observations (outliers) and "suggest different spatial regimes or other

forms of spatial instability (nonstationarity)" (Anselin, 1998 p. 120). ESDA identifies two

classifications of indicators of spatial association. They are global and local. Most of the

recent research and literature has focused on the use of local indicators of spatial

association (LISA). These indicators can detect patterns of association as well as test a

specific pattern's uniformity. LISAs are well suited for map visualization, and overlaying

LISA maps of different variables is very helpful in deciding variables that should be used

in models. For these reasons, our study focuses on LISAs how they illustrate patterns

and are used to build models.

Confirmation, or confirmatory spatial data analysis is described as "model-driven."

It involves four steps: model specification, estimation, diagnostics and prediction. These

four steps imply an iterative process in which models are tested until the best one is

found. As mentioned in the previous section on indicators, studies such as this one should

result in recommendations for government action based on the results. Therefore, it is

important to find the best model for studying and producing the most meaningful results

for the issue at hand. Also, in the discussion on confirmatory spatial analysis, Anselin

(1998) addresses regression models and their usefulness in predicting values. One

previous study uses a regression model to predict rental rates in several markets and

geographically illustrate their results for Atlanta and Boston. This model incorporates the









physical attributes of apartments and their relation to price based upon previous research.

The model illustrates geographically how rents are likely to vary in relation to the

average rent based on location and demonstrates how variables, or indicators, can be

analyzed using GIS to graphically display a neighborhood reality.

The aforementioned study testifies to the effectiveness of regression analysis,

demonstrating how the interaction of variables can be assessed to accurately display and

monitor an issue. Our study uses a deterministic model involving the pairwise

comparison method to determine the weight of each variable associated with

gentrification. This method, developed by Saaty in 1980, involves comparing each

variable to the other variables individually, creating a ratio matrix that outputs the relative

weights of each variable. This method was chosen based on the knowledge of the general

effects of each indicator on the likelihood of gentrification expressed in the literature as

well as research as well as its compatibility with the spatial analysis functions of GIS.

The application of GIS, and its spatial analysis capabilities, to housing research has

been very minimal. According to Can (1998) this lack of research is due to ignorance of

available tools; difficulty in obtaining the updated, detailed and accurate information

required for GIS-based analysis; and the relatively recent availability of "special

processing requirements" for housing research. These reasons are valid, particularly the

availability of data to make using GIS worthwhile and meaningful. Most of the specific

data collected on housing is done through the census. Some data is estimated on a yearly

basis, but these estimations are generally not done at the census block level (Can 1998, p.

69). However, some information not available in its most recent version may be available

through other non-traditional sources such as the local Property Appraiser or Chamber of









Commerce. In fact, it is possible to get more specific information from a source such as

the Property Appraiser down to the parcel as opposed to census data, which only

measures down to census tract for certain types of data. One important issue to consider

when gathering information from a variety of sources is consistency. While accuracy is

very important, ensuring that all data for all variables relates to the same year and is

measured at the same geographic level is equally important when using GIS to conduct

research and build models.

Despite the challenges, GIS is an appropriate tool for housing research. The

visualization capacity of GIS allows researchers to see patterns and trends that might not

be evident just by examining tables and graphs (Ghose and Huxold, 2004, p. 19). Also, its

analysis capabilities allow for the examination of several forces and indicators at one

time to determine their effect and guide policy action.

Summary

The goal of this review of the literature was to establish a working definition of

gentrification and examine indicator studies and GIS tools to show their application to the

study of gentrification and the creation of a model for monitoring gentrification. The

review discussed the major issues and debates in the study of gentrification, resulting in a

definition of gentrification for use in our study. Next a discussion of indicators outlined

how they have been used (particularly in economic development planning) and how they

can be applied to the study of housing and model building. Finally, an overview of GIS

and its application to housing research continued to build on themes offered in the

discussion on indicators as well as demonstrated the practicality of GIS in relation to

housing research and community monitoring. In all of these discussions, important points

were highlighted and analyzed in their relation to our study. The next two chapters







22


describe the specific geographic area used for our study and the specific details of our

model.














CHAPTER 3
STUDY AREA

Our study focuses on St. Petersburg, Florida as the test region due to its growing

population, rapid job growth, geographic constraints, dwindling availability of large

developable parcels, and growing affluence. With a population of nearly 250,000

residents, St. Petersburg ranks as the fourth largest city in the state of Florida, and

functions as one of the urban centers in the Tampa Bay metropolitan area the state's

second largest metropolitan statistical area and one of its fastest growing. St. Petersburg

is located in Pinellas County, a densely populated, nearly built-out county along the west

coast of Florida. The county itself is a large peninsula, surrounded on three sides by

water. St. Petersburg, at the southern end of the county, is also surrounded by water on

three sides. Also like the county, St. Petersburg is nearing build-out in terms of

undeveloped land. Due to its geography, no outward expansion can take place, including

typical large-scale, suburban-style developments that characterize current development in

much of the rest of Florida. Moreover, the city is experiencing significant job growth,

particularly in high-paying financial services and technological-oriented jobs, attracting

thousands of new residents in recent years. Therefore, as these trends continue, we

contend some St. Petersburg neighborhoods are bound to experience gentrification.

Our study identifies four neighborhoods as probable targets for gentrification:

Bartlett Park, Old Southeast, Roser Park, and Crescent Lake. Although each

neighborhood is unique, they all share aspects that attract gentrifyers. All are located

immediately adjacent or within 1.5 miles from the central business district. All are among









the oldest neighborhoods in the city. Roser Park, Old Southeast, and a portion of Crescent

Lake called Round Lake are designated historic districts on the national level, local level

or both.

One neighborhood, Uptown, has been identified as the control neighborhood. This

neighborhood features many of the same characteristics of the four neighborhoods

identified as gentrification targets. It is a historic district and sits directly adjacent to St.

Petersburg's central business district. However, it does not receive the same attention

from officials, planners, residents and the press as the other neighborhoods in terms of the

characteristics of and potential for gentrification. Therefore, our study asserts that change

occurring in Uptown will most accurately reflect the overall change taking place in the

city of St. Petersburg.

The national trend of central city redevelopment has not missed St. Petersburg. In

fact, St. Petersburg's central business district has been recognized several times as an

example of successful downtown redevelopment. As the central business district

generates more activity, we hypothesize that the identified four surrounding

neighborhoods will begin to feel the effects of eminent gentrification. The model

developed for our study will prove or disprove the correctness of that hypothesis.














CHAPTER 4
METHODOLOGY

Gentrification literature describes the various measurable indicators of

gentrification. It also describes the difficulty in reversing the negative effects of

gentrification, most notably the displacement of residents. Since the indicators are

known, gentrification must be measurable. However, no attempts to quantify these

indicators and relate all of them empirically to some index of the likelihood of

gentrification occurring in a neighborhood have been found in previous studies. This

chapter describes the method created for monitoring gentrification in our study,

determines specific indicators outlined in the gentrification literature using common

statistical methods and GIS technology, and tests the model on the five neighborhoods

described in the previous section.

Explanation of Model

Building the model for monitoring gentrification involved four basic steps, each of

which contained smaller steps. The first basic step was the identification of the indicators

of gentrification to be used in the model. The second basic step involved collecting the

appropriate data for those indicators, converting that data into usable statistics, and

mapping those statistics for each indicator using GIS independently. The third step

involved determining relationships between the indicators and the threshold values for

each indicator. The fourth and final step established an equation for a gentrification index

based on the statistics and thresholds to determine the likelihood of gentrification

occurring in the study area and mapped the results of the equation using GIS.









Identifying the Indicators

This first step in developing the model identified the appropriate indicators.

Perhaps the most important step in the process, choosing the right indicators to use,

greatly determined the effectiveness of the model. Our study considers sixteen indicators

based upon gentrification literature and the researcher's definition of gentrification. The

majority of the indicators chosen use census data and other data readily available to

researchers, demonstrating the accessibility of the model for practicing planners.

We divided the indicators into two groups: regional to neighborhood comparisons

and neighborhood-specific indicators. Regional to neighborhood comparisons describe

conditions that exist or changes in regional demographics that should reflect on all areas

of the metropolitan region. For instance, if area median income (AMI) increased by a

large percentage for the region, one expects to find a large increase in the AMI of each

neighborhood in the region. Neighborhood-specific indicators describe conditions and

qualities specific to a particular neighborhood. A neighborhood's location would classify

as a neighborhood-specific indicator. We chose twelve regional to neighborhood

comparison indicators and four neighborhood-specific indicators (Tables 4-1 and 4-2).

Table 4-1: Regional to neighborhood comparison indicators
Name Description Justification
Change in Professional The change in the number These tend to be higher-
Employment of people working jobs wage jobs. An increase in
requiring post-secondary the number of higher-paid
education (AA, AS, BA, workers increases area
BS, MA, MS, Ph. D., median income (AMI),
technical certificate) as a driving up housing costs.
percentage of overall
employment









Table 4-1 Continued
Name Description Justification
Change in Population The change in the total A rapid population increase
population usually relates to a growing
job market, one of the
leading indicators of
gentrification.
Change in Housing Units The change in the total A slow growth in the
number of housing units number of housing units
with respect to population
and job growth leads to
rising housing costs.
Change in college-educated The change in the One of the characteristics of
population percentage of the a likely gentrifyer; tend to
population that is college- have higher incomes and
educated affinity for city amenities.
Change in Age Cohort 25- The change in the This cohort relates to one of
34 percentage of the the characteristics of a
population in this age range likely gentrifyer (high-
wage, young, single or
married w/ no children).
Change in Age Cohort 55- The change in the This cohort relates to one of
65 percentage of the the characteristics of a
population in this age range likely gentrifyer (empty-
nester; active lifestyle).
Change in area median The percentage change in Growing AMI usually
income (AMI) AMI relates to a growing job
base, increased educational
level of residents, and
relates to an increase in
housing costs.
Change in Median Owner- The percentage change in Rising housing costs
Occupied Unit Value the value of owner- signifies increase demand
occupied single-family for housing, a leading
residential units attached as indicator of gentrification.
well as detached.
Change in Average The number of minutes One main reason residents
Commute Times commute times have are choosing to move back
increased/decreased over to central cities relates to
time increased commute times.
% Housing units occupied The change in the Higher occupancy in
percentage of housing units combination with high
that are occupied by either demand raises housing
renters or their owners prices.









Table 4-1 Continued.
Name Description Justification
% Owner-occupied units The change in the Rising homeownership
percentage of housing units tends to reflect a greater
actually occupied by their amount of income within
owners households as well as
growing neighborhood
stability an attractive
quality.
Unit Size The number of rooms in a Larger homes tend to attract
housing unit higher-incomes. Therefore
larger homes in older areas
are likely to attract
gentrifyers.


Table 4-2: Neighborhood-specific indicators
Name Description Justification
% Housing Built Pre-1950 The percentage of all the The historical value of the
housing units built prior to houses is part of the allure
1950 of inner-city neighborhoods
to gentrifyers.
Proximity to Central The number of miles the Part of the attraction is the
Business District (CBD) census tract is from those closeness to CBD, where
tracts making up the CBD jobs, culture and
entertainment are located.
Proximity to Major If interstates run through Easy access to corridors
Transportation Corridors city, the number of miles to leading to CBD as well as
(Interstate Highways) the nearest interchange; if suburban markets one of the
not, the number of miles to important factors to
the nearest major corridor gentrifyers.
Historical Designations Number of historic Designations curtail
structures or if entire tract is demolition, encouraging
within historic district renovation; historic value
attractive to gentrifyers.


Data Collection

Most of the data collected comes from the United States Bureau of the Census


(Census). However, some data was collected from other sources.









Table 4-3: Sources for regional to neighborhood comparison indicators
Name Units Source
Change in Professional Percentage Census
Employment
Change in Population Percentage Census
Change in Housing Units Percentage Census
Change in college-educated Percentage Census
population
Change in Age Cohort 25- Percentage Census
34
Change in Age Cohorts 55- Percentage Census
65
Change in AMI (area Percentage Census
median income)
Change in Owner-Occupied Percentage Census
Unit Value
Change in Average Percentage Census
Commute Times
% Housing units occupied Percentage Census
% Owner-occupied units Percentage Census
Unit Size Number Census


Table 4-4: Sources for Neighborhood-specific indicators
Name Units Source
% Housing Built Pre-1950 Percentage Census
Proximity to Central Number Scaled street map of city
Business District (CBD)
Proximity to Major Number Scaled street map of city
Transportation Corridors
Historical Designations Percentage City Government, National
Register of Historic Places


In order to gauge change and show a clear trend, data collection encompassed a 20-

year period (three decennial censuses) for each indicator whose source is the Census

(2000, 1990 and 1980). Data gathered on other indicators also spanned the same twenty-

year timeframe where available. If data was available only over a shorter time period,

data collection began with the earliest year available. Collecting data in this manner kept

the intervals the same to establish trends over the same number of years as the indicators









based on the Census. In addition to consistency in time intervals, the values must also be

geographically consistent. Thus, data not available from the Census was appropriately

scaled or proportioned to match the census tracts used for the neighborhoods analyzed in

our study.

We defined the "region" as the city where the neighborhoods are located St.

Petersburg, Florida. The "neighborhood" refers to each of the five neighborhoods

analyzed in our study area separately. The boundaries of each neighborhood matched up

almost perfectly with the boundaries of their respective census tracts (Figures B-l and

B-2).

Census data generally comes as a simple count (integer) or where appropriate, as a

dollar amount. However, in this research, percentage change bears more relevance. For

instance, the median income in the city could increase by more absolute dollars than a

neighborhood, but the neighborhood could show a higher percentage increase, reflecting

a greater rate of change. Therefore, the counts for each regional to neighborhood

comparison indicator were transformed into a percentage change value using the

following formula:

Percent Change = [(X Y)/Y] 100

where
X = Value from 2000 Census or most recent available, and
Y = Value from 1980 Census


For neighborhood-specific indicators, no rate of change was measured between

1980 and 2000, as they reflect neighborhood characteristics at their present state based on

the 2000 census, demonstrating potential based on current conditions.










Most of the indicators are dynamic and measured by percentage change. However

two indicators describe static conditions and carry number measurements -- distance to

central business district and distance to major transportation corridors. It is quite possible

for distance to major transportation corridors to change due to construction of new

corridors.1 Yet, we determined that no new transportation corridors affecting these

neighborhoods were constructed during the study period. Also, the locations of the

traditional central business district (downtown) and the location of each neighborhood

remain stationary. For these reasons, a number value is the appropriate measure for these

indicators.

Each indicator is then mapped using ArcGIS2 according to the percentage or integer

value associated with each. First, the GIS shape files for the appropriate city boundary

and the census tracts are downloaded from the Florida Geographic Data Library3 into GIS

creating the base map. Then the attribute table for the census tract layer was edited to

include the fields for the values relating to each indicator. Next, the values in each of

these fields were converted from "vector" attributes to "raster" attributes.4 These values


1If new major transportation corridors are constructed, then the distance from a study area to a major transportation
corridor may change; thus making this a dynamic variable that may be more appropriately measured by percentage
change.


2 ArcGIS is a GIS software package from ESRI most often used by planners, developers and researchers


The Florida Geographic Data Library is an electronic resource providing free access to GIS shape files for all
counties in the State of Florida and their corresponding attribute tables and metadata files.


4
Vector data associate attributes with each feature point, line, and polygon; whereas raster data represents surfaces
as grids of equally sized cells that contain attribute values and location coordinates. With raster data, groups of cells
that share the same value represent the same type of geographic feature. For instance, all census tracts would be
represented with the same color regardless of their associated rate of population increase when displayed as vector data;
whereas, with raster data, only tracts with the same rate of increase in population would share the same color on the
map.









are then reclassified using the binary system of 0 and 1 according to their value in

relation to the regional percentages.5 The reclassification assigned a value of 0 to all

values less than the regional percentage, and assigned a value of 1 to all values greater

than the regional percentage in most cases. In a few instances, the reclassification was

based on the opposite relationship. For example, a reclassification value of 1 was

assigned to tracts with a change in vacancy rates less than the regional rate. The

reclassified values were converted to individual shape files and added to the base map as

separate layers. The purpose of doing this was to spatially and visually reinforce the

change occurring in the study area in relation to each indicator.

Developing the Equation

The equation used to analyze the five neighborhoods utilizes deterministic

neighborhood value analysis in combination with weighted suitability analysis to

determine a gentrification index. The following sections outline this process

Deterministic Neighborhood Value Analysis

Since monitoring gentrification engages several indicators, the study used

deterministic neighborhood value analysis to weight the values of several variables to get

one final index for gentrification. Deterministic neighborhood value analysis uses the

following equation:

I C1X1 + C2X2 + C3X3 + ... + CnXn

where
I = index
Ci = weight of the first indicator X1
C2 = weight of the second indicator X2

Since the current body of literature establishes no generic thresholds for these gentrification indicators, the most
appropriate measures of change are the regional percentages.










C3 = weight of the third indicator X3
Cn = weight of the nth indicator Xn

The weights for each value were determined using the pairwise comparison method

established by Saaty in 1980 described in the literature review.6 This method determines

the weight of variables in decision-making using the comparison matrix (Table 4-5),

testing each variable against all other variables individually:

Table 4-5: Pairwise comparison matrix
Variable X1 Variable X2 Variable X3 ... Variable Xn
Variable X1 1 X2:X1 X3:X1 ... Xn:Xi
Variable X2 XI:X2 1 X3:X2 ... Xn:X2
Variable X3 XI:X3 X2:X3 1 ... Xn:X3
1


Variable Xn XI:Xn X2:Xn X3:Xn ..:Xn 1


Comparisons were done on a scale of 1 to 9 using the following descriptions:

1 = equally important
2 = slightly more important
3 = somewhat more important
4 = moderately more important
5 = more important
6 = much more important
7 = significantly more important
8 = very much more important
9 = extremely more important


When comparing variables to themselves, the value always equals one. If the

comparison of variable X2 to X1 yields one value, then the comparison of Xi to X2 yields




6 An alternative to the researcher developing the weights would be to survey local professional planners with housing
expertise as well as area residents using the same criteria and develop the weights through a method of consensus
building an iterative process by which all those involved would come to an agreement on the value of each indicator
to the whole equation.









the reciprocal value. For example, if variable X2 is significantly more important than Xi

(value =7), then variable X1 is significantly less important than X2 (value = 1/7).

Table 4-6: Pairwise comparison matrix value pattern
Variable X1 Variable X2 Variable X3 ... Variable Xn
Variable X1 1 1/XI:X2 1/X3:X1 ... 1/XI:Xn
Variable X2 XI:X2 1 1/X2:X3 ... 1/X2:Xn
Variable X3 XI:X3 X2:X3 1 ... 1/X3:Xn
1


Variable Xn Xi:Xn X2:Xn X3:Xn ..:Xn 1


These comparison values were then normalized by the following equation:

Normalized Value = Comparison Value (1/ Total of all values in column).

Then these normalized values were summed up by column. This total became the weight,

or coefficient C, assigned to each indicator.

After establishing the C values for each indicator, the deterministic neighborhood

value analysis equation uses reclassified values for each indicator described in the

previous section as X values to measure their total effect. For each neighborhood, the

study analyzed the regional to neighborhood comparisons and neighborhood-specific

indicators separately, providing a total for both to be used later in the weighted suitability

analysis. Although the study analyzed the five neighborhoods separately, it used the same

equations for each, employing the same C values. Using the same equation demonstrates

the regional applicability of this analysis. The uniqueness of the totals for a neighborhood

would come from its X values.

Weighted Suitability Model

The weighted suitability model is a method of spatial analysis often used in real

estate development to determine the suitability of a site for a specific type of









development targeting a specific demographic. It assigns weights to multiple groups of

variables in the same manner that multivariate regression applies weights to individual

variables. Since our study uses two categories of indicators, the weighted suitability

model effectively illustrates the relationship between the two sets of indicators and their

effect on the overall decision-making of potential gentrifyers.

The weighted suitability model is used to establish the equation for the final index

of the likelihood of gentrification, G. For our study, regional to neighborhood comparison

indicators YRegional carried a coefficient of 0.8, accounting for 80% of the result, and

neighborhood-specific indicators YNeighborhood carried a coefficient of 0.20, accounting for

20% of the result. We derived these proportions from the gentrification literature that

identifies the major indicators for gentrification as increasing commute times, rapid job

and population growth, and changes in demographics of age and income, all issues

accounted for in the regional to neighborhood comparisons. Neighborhood-specific

attributes, such as proximity to the central business district and architectural character,

also bear much significance. However, according to the gentrification literature, these

characteristics carry less importance than the regional to neighborhood comparisons. For

this reason, the 80% to 20% ratio applied well to the model, giving the regional to

neighborhood comparison indicators the majority of the weight without marginalizing the

effects of the neighborhood-specific indicators.

Using the weighted suitability model, the data accurately produces a gentrification

index (G) for each neighborhood in the study area with the following equation:

G = 0.8YRegional + 0.2YNeighborhood

Where









YRegional = deterministic neighborhood value analysis of regional to local comparison
indicators, and
Neighborhood = deterministic neighborhood value analysis of neighborhood-specific
indicators.


The Raster Calculator in the Spatial Analyst menu of ArcGIS calculated the G

values for each neighborhood and added their graphic representation to the base map as a

separate layer. The G values were measured on a scale of 0 to 1, with 0 equal to 0%

likelihood of gentrification and 1 equal to 100% likelihood of gentrification.

This process outlines a method for empirically measuring and graphically

displaying the potential for gentrification. It provides a means to quantify physical and

social attributes of an area and relate them mathematically to describe neighborhood

change.














CHAPTER 5
FINDINGS AND RESULTS

This thesis focuses on the use of census and other relevant data to reveal long-term

patterns of change and use them to monitor gentrification in a neighborhood. The

following chapter will report the findings for each indicator separately, looking at overall

trends from 1980 to 2000 as well the differences between the rate of change in the 1980s

and the rate of change in the 1990s. Although our model does not use the rates of change

from 1990 to 2000, the trends they reveal are worth discussing.

Regional to Local Comparisons

In many cases, indicators in the local areas (neighborhoods) were consistent with

the general trend in the region. However, in some cases, the local areas and region

registered opposite trends. Overall, the findings for these indicators revealed that

although these neighborhoods share common characteristics, such as their geographic

locations, they are each unique; therefore, lending themselves to a range of possibilities

in their likelihood for gentrification.

Professional Job Growth

Between 1980 and 2000, the city of St. Petersburg experienced a 10.09% increase

in the number of residents with professional jobs. Further analysis reveals that the

majority of that increase occurred between 1990 and 2000, a 7.38% increase.

From 1980 to 2000, all five neighborhoods in the study area register an increase in

the number of residents with professional jobs. Two neighborhoods, Roser Park and

Crescent Lake, show an increase much higher than the city. With a 19.82% increase in









professional jobs, Roser Parks' rate of increase is nearly twice that of the city. Crescent

Lake's 16.38% increase is also significantly higher. This shows the strong appeal of these

neighborhoods to professionals. Bartlett Park, Old Southeast and Uptown also showed

increases of 5.2%, 9.82% and 8.36% respectively, perhaps implying a growing interest,

but not yet on the level of the other two neighborhoods.

Change in Population

The census reports that the population of the city of St. Petersburg increased from

238,547 in 1980 to 248,232 in 2000, a 4.02% increase in population. Further examination

shows that the majority of this population increase occurred between 1990 and 2000, as

the census reports a population of 238,629 in 1990.

The trend of increasing population for the city of St. Petersburg as a whole does not

hold true in any of the neighborhoods in the study area. In fact, some neighborhoods

experienced a sharp decline in population. The Crescent Lake neighborhood, represented

by Census Tract 235, had the smallest change, with a 0.94% decrease in population from

1980 to 2000. In ascending order, Old Southeast (Tract 204) shows a 3.31% decrease,

Uptown (Tract 234) shows a 9.6% decrease, Bartlett Park (Tract 205) shows a 18.26%

decrease, and Roser Park (Tract 213) shows a 51.0% decrease.

Considering the increase in city population, these neighborhood-level decreases are

unexpected. On face value, these decreases in population could represent disinterest and

disinvestment. However, this population decrease may be explained by trends relating to

other indicators.

Change in Housing Units

Between 1980 and 2000, the number of housing units in the city of St. Petersburg

increased 4.3%. However, over both censuses, all five neighborhoods report a decreasing









number of housing units. Still, Roser Park shows a strikingly high decrease in housing

units, reporting a 78.46% decrease. The second-highest decrease occurred in Uptown,

reporting a 24.84% decrease. Bartlett Park ranks third, with an 18.26% decrease,

followed by Crescent Lake and Old Southeast, with 16.35% and 10.56% decreases

respectively.

These decreases in housing units may be explained by conversion of housing units

to office space. For instance, due to its location near a large hospital district and

university campus, some housing units in the Roser Park neighborhood may have been

purchased by those institutions for future expansion or by businesses wishing to be close

to them. Another explanation could be the conversion of large structures back to single-

family uses that were formerly rented as multiple units.

Change in College-Educated Population

From 1980 to 2000, the number of persons with Bachelors, Graduate and

Professional degrees in the city of St. Petersburg has increased 8.25%, from 14.57% in

1980 to 22.82% in 2000. This increase appears to be steady, with 4.19% occurring

between 1990 and 2000.

All five neighborhoods also report an increase in the number of residents with four-

year degrees or higher. Three neighborhoods show a rate of increase higher than that of

the city. They are Old Southeast, Roser Park and Crescent Lake, with 19.82%, 8.36% and

16.09% increases respectively. These larger increases imply that these are clearly

neighborhoods of interest for college-educated persons. Bartlett Park and Uptown report

increase of 5.2% and 6.12% respectively. Although these represent a gain in college-

educated residents, the smaller values indicate these neighborhoods aren't as popular as

the other three.









Change in Age 25 through 34 Population

From 1980 to 2000, St. Petersburg shows a slight increase in the number of

residents from the age of 25 through 34 with an overall increase of 0.74% from 13.02%

of the population in 1980 to 13.76% of the population in 2000. There was a larger

increase from 1980 to 1990, going from 13.02% to 14.96%, then decreasing in 2000 to

13.76%.

The population in this cohort increased during the twenty-year period in two of the

neighborhoods and decreased in the other three. Uptown's increase of 0.95% is slightly

above the city's rate of increase. Crescent Lake experienced a more significant 3.86%

increase. However, Bartlett Park, Old Southeast and Roser Park all experienced decreases

- 6.19%, 3.78% and 5.24% respectively. Although the rate of increase appears slow for

Uptown and Crescent Lake, both are gaining residents of this age faster than the city,

indicating an attractiveness of these neighborhoods to younger adults. The decreases in

Bartlett Park, Old Southeast and Roser Park imply an unattractiveness of these

neighborhoods to younger adults.

Change in Age 55 through 64 Population

The population aged 55 through 64 has decreased in St. Petersburg from 12.15% in

1980 to 9.17% in 2000, a 2.98% decrease. The majority of this decrease occurred

between 1990 and 2000 when the 55 to 64 population decreased 1.69% from 10.86% to

9.17%.

Two neighborhoods registered an increase in this age group, whereas the

population in this age group declined in three of the neighborhoods. Bartlett Park

experienced an increase of 3.72% from 1980 to 2000, the majority occurring between

1980 and 1990 (2.87%). This slowing increase may imply a developing disinterest in the









area from this age group. Old Southeast reports an overall increase of 0.38%. Although

the population in this age group decreased between 1980 and 1990 from 9.79% to 8.43%

of the total population, it increased again between 1990 and 2000 to 10.17%. This

indicates that the Old Southeast may be developing into a neighborhood of interest for

this age group. Roser Park, Uptown and Crescent Lake report decreases of 1.17%, 2.05%

and 3.91% respectively. In all three cases, the majority of decrease occurred between

1980 and 1990. This slowing decrease may also indicate increasing interest in these three

neighborhoods for this age group.

Change in Area Median Income

The area median income has increased dramatically in St. Petersburg, going from

$11,798 in 1980 to $34,597 in 2000, a 193% increase, or nearly tripling in twenty years.

The majority of that increase took place between 1980 and 1990, when median income

experienced a 146.26% increase from $11,798 to $23,577. This significant increase in

median income could be explained by an increasing number of two-wage earner

households and the greater upward mobility of women during this time period.

All five neighborhoods experienced significant increases in median income.

Crescent Lake experienced the largest increase (234%), going from $6,964 in 1980 to

$23,225 in 2000. Not far behind with a 200% increase is Old Southeast, rising from

$10,386 in 1980 to $31,163 in 2000. Uptown experienced a 169% increase from $8,466

in 1980 to $22,768 in 2000. The smallest increases were in Bartlett Park and Roser Park,

reporting 135% and 158% increases respectively. Bartlett Park increased from $8,135 to

$19,125, while Roser Park increased from $7,584 to $19,531. Just as with the city, all

five neighborhoods experienced their greatest gains between 1980 and 1990.









Although all five neighborhoods have gained significantly, their median incomes

still lag behind that of the city of St. Petersburg as a whole. However, with gains of 200%

and 234%, incomes in Old Southeast and Crescent Lake are growing at a faster rate than

the city's rate of increase, indicating interest in these areas from higher-income

households. Moreover, of the five neighborhoods, Roser Park is the only neighborhood in

which a higher rate of increase in income occurred from 1990 to 2000 than the city's rate

during that same period an increase of 69.76% for the neighborhood compared to

46.74% for the city, implying that Roser Park has caught the attention of higher-income

households. Yet the overall numbers from 1980 to 2000 reveal that there still remains a

large presence of low-income households in the neighborhood.

Change in Median Single-family Unit Value

From 1980 to 2000, single-family homes in the city of St. Petersburg increased in

value by 126%, going from $35,800 in 1980 to $81,000 in 2000. This increase mostly

took place during the 1980s, when values increased by 96.81%, or nearly doubled. Both

Bartlett Park and Old Southeast experienced similar rates of increase 122% and 125%

respectively. Values in Bartlett Park grew from $20,600 in 1980 to $45,800 in 2000;

whereas values in Old Southeast grew from $37,900 in 1980 to $85,400 in 2000.

The three other neighborhoods saw values rise at a higher rate than the city. Roser

Park and Crescent Lake experienced the greatest increase in single-family home values.

In Roser Park, values rose an impressive 255%, more than tripling from $19,200 in 1980

to $68,100 in 2000. Likewise, Crescent Lake values grew by 211%, also more than

tripling from $28,700 in 1980 to $89,200 in 2000. Although not as high, Uptown values

rose 170% from $29,000 in 1980 to $78,200 in 2000. In addition, all three neighborhoods

had higher rates of increase between 1990 and 2000 than the 29.19% rate of the city, with









Roser Park reporting a 51.33% increase, Crescent Lake reporting a 50.42% increase and

Uptown reporting a 48.95% increase. Of these three neighborhoods, values in two -

Roser Park and Uptown still lag behind the regional median value. Still, the rising

values generally relate to rising demand, implying specific interest of homebuyers in

these three neighborhoods.

Change in Housing Vacancy

Interestingly, from 1980 to 2000 the city reports an overall increase in vacancy of

2.24% from 1980 to 2000. However, the vacancy rate decreased by 3.74% between 1990

and 2000, indicating increased absorption of housing units in the city overall.

Four of the five neighborhoods followed similar patterns. Bartlett Park experienced

the highest increase in vacancy, rising from 17.02% in 1980 to 28.77% in 2000. Vacancy

in Crescent Lake rose 6.67% over the same time period. In Uptown, the rate grew 3.77%.

Roser Park reported the smallest increase with 0.36%. However, all four experienced

decreases in their vacancy rates in the 1990s. Crescent Lake reports a 10.04% decrease

during that decade. Roser Park had the second-highest decrease of 6.9%. Uptown and

Bartlett Park experienced decreases of 2.16% and 0.02% respectively. Old Southeast is

the only neighborhood to experience an overall decrease in vacancy from 1980 to 2000.

Vacancy decreased by 2.41%, going from 15.97% in 1980 to 13.56% in 2000. Still, all

five neighborhoods continue to have higher rates of vacancy than the city as a whole.

However, with vacancy rates decreasing at a faster rate than the city between 1990 and

2000, both Roser Park and Crescent Lake appear to be neighborhoods of interest.

Change in Owner-Occupancy

Surprisingly, owner-occupancy decreased over the twenty-year period by 1.17% in

the city of St. Petersburg from 57.04% in 1980 to 55.87% in 2000. However, the rate of









owner-occupancy increased by 2.8% between 1990 and 2000. Only one other

neighborhood followed a similar pattern Bartlett Park. Here, owner-occupancy

decreased by 2.16% between 1980 and 2000, but it increased by 5.04% between 1990

and 2000.

The other four neighborhoods experienced growing owner-occupancy over both

time periods. Ownership in Roser Park grew 9.64% from 1980 to 2000, with 95% of that

growth taking place in the 1990s. Old Southeast, Uptown and Crescent Lake also

experienced an increase in ownership from 1980 to 2000, with increases of 2.54%, 0.68%

and 1.23% respectively. However, these neighborhoods saw greater rates of increase in

the 1990s than over the twenty-year span of 1980 to 2000. Old Southeast reports an

increase of 9.78% during the 1990s. Uptown and Crescent Lake saw increases of 4.3%

and 5.15% respectively.

With the exception of Bartlett Park, owner-occupancy increased faster in the

neighborhoods than in the city overall from 1980 to 2000. However, ownership increased

faster in Bartlett Park than the city overall from 1990 to 2000. Both trends imply a

growing number of homeowners, associated with a stabilizing neighborhood. Moreover,

these rates indicate the growing appeal of these neighborhoods to homebuyers.

Unit Size

The median number of rooms in owner-occupied units in 2000 was 5.5 rooms for

the city. Of the five neighborhoods, Old Southeast and Roser Park had a higher median

number of rooms, with 6 and 7.4 rooms respectively. Bartlett Park homes tend to be

smaller than that of the city, with a median of 5.3 rooms. The same applies to Uptown,

with a median of 5.2 rooms. Crescent Lake reflects the citywide median of 5.5 rooms.









The larger homes of Old Southeast and Roser Park lend themselves to greater

attractiveness; whereas, the smaller homes of Bartlett Park and Uptown may not be as

attractive. As the homes of Crescent Lake tend mirror the city as a whole, other indicators

would have a greater effect on the likelihood of gentrification taking place there.

Change in Commute Times

Over the twenty-year period the average commute times increased in all instances.

The city average commute time increased 5.64% from 19.5 minutes in 1980 to 20.6

minutes in 2000. Uptown reports the greatest increase in commute times, rising 37.84%

from 14.8 minutes in 1980 to 20.4 minutes in 2000. The second-largest increase

happened in Old Southeast, with a 24.57% increase from 17.5 minutes in 1980 to 21.8

minutes in 2000. Crescent Lake, Roser Park and Bartlett Park experienced increases of

6.96%, 7.21% and 1.39% respectively. If gentrification is happening in these areas, then

these commute times are still low enough to attract new residents. An alternative

explanation may be that a change in commute times is not a significant indicator of

gentrification.

Neighborhood-Specific Indicators

Percentage of Housing Constructed before 1950

All neighborhoods have relatively high percentages of housing units built prior to

1950. Two neighborhoods, Uptown and Crescent Lake, have maintained the majority of

their older residential units, reporting that 57.47% and 56.04% of their units were built

prior to 1950. However, the three of the four neighborhoods believed to be targets of

gentrification reported the lowest percentages of old homes. Bartlett Park reports in 2000

that 41.16% of its units were constructed before 1950. The percentages for Old Southeast

and Roser Park were 44.08% and 42.17% respectively. It appears that Uptown and









Crescent Lake did a better job of preserving historic character over the years than has

Bartlett Park, Old Southeast and Roser Park. If these three neighborhoods are gentrifying,

this data may counter the hypothesis that gentrifyers are generally attracted to the

architecture of older neighborhoods.

Proximity to the Central Business District and Interstate Highways

Roser Park, Uptown and Crescent Lake are directly adjacent to the business

district, and are all bordered on at least one side by an interstate highway. In all cases, the

bordering interstate highway is the divider between the neighborhood and the central

business district. Bartlett Park and Old Southeast are located further away one mile and

1.5 miles respectively. However, they are both within a five minute drive of the central

business district. Their proximity to the central business district and the interstate

highways, which provide access to suburban job markets, make these neighborhoods

attractive to gentrifyers looking for shorter commutes to the central business district or

who don't mind the "reverse" commute to the suburbs in exchange for easy access to the

cultural and entertainment amenities of the central business district.

Historic Designations

Old Southeast contains the greatest number of historic designations with a local

historic district designation and three individual historic structure designations, two

national and one local. Crescent Lake follows with a portion of the area designated as the

Round Lake national historic district and one historic structure. Lastly, Roser Park is

designated a national historic district. Both Bartlett Park and Uptown have no historic

designations.

According to previous studies (Redfern, 2001; Nelson, 1988; Lang, 1982),

maintenance of historic character makes an area more attractive to gentrifyers. Historic









designations in a neighborhood or the designation of an entire neighborhood as a historic

district attest to the neighborhood's commitment to maintain that character. Therefore,

two of the four neighborhoods believed to be targets for gentrification Old Southeast,

Roser Park are likely to succeed; whereas, Bartlett Park and Uptown may not attract as

many gentrifyers as they are not designated like the other two.

Major Relationships

Examination of these statistics revealed some relationships between indicators.

There were some expected correlations, such as that between population and housing

units. However, some relationships didn't follow usual patters, such as that between

housing vacancy, number of units and value. The following paragraphs will discuss

relationships found between these indicators.

Overall, the number of housing units in the city increased at the same rate as

population increase, indicating that housing production in the city has generally kept pace

with population increase. However, although population has decreased in the

neighborhoods, the number of housing units has decreased at a much higher rate in all

cases except Bartlett Park. Although the city's growing population may be redistributing

itself in other areas, there still remains interest in these neighborhoods in 2000, perhaps

by larger households than had previously occupied them in 1980. This theory runs

counter to how gentrification research identifies a gentrifyer -- described as a

nontraditional household (young, single persons or unrelated individuals), or a married

couple with no children living in the house (younger couple or older yet active, empty-

nest couple). The theory of growing household size is further supported by the overall

decrease in population of the age cohorts generally associated with these two

demographics ages 25 through 34 and ages 55 through 64. An increasing household









size may also indicate that gentrification does not necessarily relate to growth in those

demographics, but could possibly relate to growth in families with upwardly mobile

householders; thus, adding another dynamic to ideas of how gentrification manifests itself

in different cities.

Likewise, as the number of residents with bachelor's degrees or higher increases,

the number of residents with professional jobs increases. In most cases, the number of

professional workers has increased at a higher rate than the number of college-educated

residents. This, perhaps, indicates an increasingly competitive job market that continues

to attract new, highly-educated residents. In addition to possibly reflecting an increasing

number of two-income households, the increase in area median income in all geographic

areas also relates to the growing number of highly-educated professional workers as

demonstrated by the statistics gathered for this research. This increase in income and

percentage of college-educated residents supports the hypothesis that these

neighborhoods are targets for gentrification, as previous studies on the subject indicate

that job growth, particularly professional job growth, is the major indicator of the

potential for gentrification.

Finally, interesting relationships exist among the statistics relating directly to the

housing units. As the number of units decreases, one expects the vacancy rate to also

decrease. Conversely, as the number of units decreased, the vacancy rate increased in

nearly all instances. Despite an increasing vacancy rate, the value of single-family units

continued to rise. This increase in value probably relates to the general increase in owner-

occupancy, which also supports previous gentrification research that points to increasing

home-ownership as a sign of gentrification. In addition, the two neighborhoods with the









largest homes, Old Southeast and Roser Park experienced the highest rates of increase in

homeownership. Roser Park, with the largest homes, experienced the highest rate of

increase in home value, while Uptown and Crescent Lake, with the largest collection of

homes constructed before 1950, experienced the second and third-largest increases in

home value. Moreover, these three neighborhoods immediately adjacent to the central

business district Roser Park, Uptown and Crescent Lake experienced the highest rates

of home value increase. This supports gentrification research on the attractiveness of

large, older homes close to the central business district to gentrifyers.

Results

Using the model described in the previous chapter the results strongly support the

hypothesis in one neighborhood. In other neighborhoods, the results counter the

hypothesis. The following paragraphs will describe the application of the statistics

developed from the census data, the relationships discovered among the statistics related

to each indicator in the model, and the resulting gentrification index.

Weights

The weights for each indicator were calculated using the pairwise comparison

described in the methodology chapter. Each indicator was compared to the other

indicators individually based in part on their ranking of importance as expressed in the

literature on gentrification and in part on their specific relevance to gentrification in St.

Petersburg. For instance, the change in commute time is a major indicator of

gentrification according to the gentrification literature, as neighborhoods experiencing

gentrification should register decreasing commute times. However, four of the five

neighborhoods report commute times increasing at a higher rate than the region (the city

of St. Petersburg). Therefore, in fitting with the hypothesis, change in commute times









carries a smaller weight with neighborhoods in St. Petersburg. Tables 5-1 and 5-2

display the weights calculated for each indicator:

Table 5-1: Regional to neighborhood comparison indicators
Name Weight Percent of Total Weight
% Change in Population 0.0864 8.64%
% Change in Housing Units 0.1684 16.84%
% Change in Professional 0.1875 18.75%
Jobs
% Change in College 0.0712 7.12%
Educated Population
% Change in Age Cohort 0.0362 3.62%
25-34
% Change in Age Cohort 0.0439 4.39%
55-64
% Change in Area Median 0.0630 6.30%
Income
% Change in Single-Family 0.1062 10.62%
Unit Value
% Change in Commute 0.0379 3.79%
Time
% Change in Housing 0.1141 11.41%
Vacancy
% Change in Owner- 0.0419 4.19%
Occupancy
Unit Size 0.0380 3.8%


Table 5-2: Neighborhood-specific indicators
Name Weight Percent of Total Weight
% Housing Pre-1950 0.43175 43.17%
Proximity to Central 0.26025 26.03%
Business District
Proximity to Major 0.2076 20.76%
Transportation Corridors
(Interstate Highways)
Historic Designations 0.3478 34.67%


Values

For use in the equation, the model reclassified the statistics for each indicator using

the binary system values of 0 and 1. The regional (city) values were used as the









thresholds to determine how indicator value was reclassified. Since gentrification

literature gives neither universal thresholds nor any direction on how to stratify the

reclassification of values based on preset thresholds, reclassification based on the city

values using the binary system was the most appropriate and effective means of

evaluating each indicator. The reclassification for each indicator is as follows:

% Change in Population % Change in Housing Units
1 = Tract > 4.02% 1 = Tract < 4.3%
0 = Tract < 4.02% 0 = Tract > 4.3%

% Change in Professional Employment % Change in College-Educated Pop.
1 = Tract > 10.09% 1 = Tract > 8.25%
0 = Tract < 10.09% 2 = Tract < 8/25%

% Change in Age 25-34 Population % Change in Age 55-64 Population
1 = Tract > 0.74% 1 = Tract > -2.98%
0 = Tract < 0.74% 0 = Tract < -2.98%

% Change in AMI % Change in Single-Family Home Value
1 = Tract > 193% 1 = Tract > 126%
0 = Tract < 193% 0 = Tract < 126%

% Change in Commute Times % Change in Housing Vacancy
1 = Tract < 5.64% 1 = Tract < 2.24%
0 = Tract > 5.64% 0 = Tract > 2.24%

% Change in Owner-Occupancy Unit Size
1 = Tract > -1.17% 1 = Tract > 5.5 Rooms
0 = Tract < -1.17% 0 = Tract < 5.5 Rooms

% Housing Pre-1950 Proximity to Central Business District
1 = Tract > 0% 1 = Tract = 0 miles (directly adjacent)
0 = Tract = 0% 0 = Tract > 0 miles

Proximity to Transportation Corridor Historic Designations
1 = Tract = 0 miles (directly adjacent) 1 = Historic designations present
0 = Tract > 0 miles 0 = No historic designations present


This reclassification was done using the "reclass" function in the Spatial Analyst menu of

ArcGIS. The resulting equation for the gentrification index (G) was









G = 0.8 [(0.0864 A in population) + (0.1684 A in units) + (0.1875 A in
professional jobs) + 0.0712 A in college-educated) + (0.0362 A in age 25-34) +
(0.0439 A in age 55-64) + (0.0630 A in AMI) + (0.1062 A in single-family value) +
(0.0379 A in commute time) + (0.1141 A in housing vacancy) + (0.0419 A in owner-
occupancy) + (0.0380 unit size)] + 0.2 [(0.43175 housing pre-1950) + (0.26025 *
proximity to CBD) + (0.2076 proximity to transportation corridors) + (0.3478 historic
designations)]

This equation used the reclassified values for each indicator to calculate the

gentrification index G. We used the trends from 1980 to 2000 to establish the values for

each indicator in the gentrification index calculation. This equation was inputted into the

Raster Calculator in the Spatial Analyst menu of ArcGIS, which inputted the reclassified

values into the equation and yielded gentrification indices with the following values:

Bartlett Park = 0.1559
Old Southeast = 0.4577
Roser Park = 0.7358
Uptown = 0.4072
Crescent Lake = 0.6277

Multiplying those values by 100 more clearly communicates the relative likelihood of

gentrification:

Bartlett Park = 15.59%
Old Southeast = 45.77%
Roser Park = 73.58%
Uptown = 40.72%
Crescent Lake = 62.77%

Both Roser Park and Crescent Lake show the greatest likelihood for gentrification

with gentrification indexes (probabilities) of 73.58% and 62.77% respectively. Old

Southeast and Uptown have lower likelihood of gentrification, with indexes of 45.77%

and 40.72%. Bartlett Park's index comes in substantially lower than Uptown at 15.59%.

These indexes strongly support the hypothesis with Roser Park and Crescent Lake,

moderately support the hypothesis with Old Southeast, and disprove the hypothesis for









Bartlett Park. With a likelihood of 40.72%, Uptown proves not to be representative of the

city of St. Petersburg and should be re-evaluated in its role as the control neighborhood.

Clearly, Roser Park and Crescent Lake are experiencing the most rapid change, and

likely would gentrify before the other neighborhoods in the study area. Perhaps, the

process has already begun in these two neighborhoods. What differentiates these two

neighborhoods from the others that explain this higher likelihood? Geographically

speaking, Roser Park, Crescent and Uptown are adjacent to the central business district.

However, Roser Park and Crescent Lake are closest to the core of the central business

district where most of the activity takes place. Both neighborhoods showed great

increases in the percentage of residents in professional employment, the only two with

higher rates of increase than the city. Uptown and Crescent Lake both have high

percentages of older housing, Uptown with the highest of all neighborhoods in the study

area. However Crescent Lake homes are larger, equal to the city average. Similarly, Old

Southeast has a slightly larger collection of older homes; however, single-family homes

are significantly larger in Roser Park than in Old Southeast. Neither Bartlett Park nor Old

Southeast are directly adjacent to the central business district. However, Bartlett Park has

shown the smallest increase in professional employment and college-educated residents;

its average home size is smaller than the city average, and it has the smallest collection of

older homes of all the neighborhoods in the study area. While these explanations do not

address every indicator, they begin to explain why Roser Park and Crescent Lake exhibit

high potential for gentrification and Bartlett Park trails so far behind. Perhaps, the process

has already begun in those neighborhoods, with Old Southeast and Uptown poised to






54


follow them in a second wave of gentrification and Bartlett Park in the distant future if

ever at all.














CHAPTER 6
CONCLUSION

In our study, we identified several indicators of gentrification according to previous

research on the subject and used them to develop a model that monitors community

change and assesses the likelihood of gentrification with a deterministic statistical

analysis method and a weighted suitability analysis that uses the spatial analyst

capabilities of geographic information systems. Our hypothesis defines four

neighborhoods as targets of gentrification (Bartlett Park, Old Southeast, Roser Park and

Crescent Lake) and one control neighborhood (Uptown). The results are mixed. Our

model proves our hypothesis correct for Roser Park, Crescent Lake, and arguably Old

Southeast. Our hypothesis is proved wrong for Bartlett Park, found not to be a target of

gentrification (yet) and Uptown, found to be more of a target than expected. However,

our study demonstrates the capabilities of statistical analysis and geographic information

systems to address housing issues in a proactive manner by anticipating the likelihood of

gentrification.

Universal Applicability

Since gentrification manifests itself in accordance with the unique dynamics of a

local housing market, it is impossible to develop an equation with coefficients that can be

used for analyzing any neighborhood in any city. However, the indicators of

gentrification are generally the same everywhere. Therefore, in order to apply our model

to other cities, the coefficient values associated with each indicator should be adjusted to

reflect how they interact in that specific market.









Policy Implications

Any model for monitoring a planning issue should produce meaningful results for

use in the development of policies and programs. Our deterministic model of

gentrification allows planners to accurately identify those neighborhoods more likely to

gentrify and use that information a basis for changes to or the creation of new policies,

programs and planning initiatives.

Planning, overall, has developed into a reactionary practice. More proactive

planning needs to take place. However, in order for planners to work proactively, they

must be equipped with the tools necessary to provide solid analysis on which to base their

recommendations. Our model provides an excellent example of how common planning

tools and resources can be used for analysis of a complex planning issue gentrification.

The results of the model can be used to guide the implementation of specific programs,

such as tax credit and grant programs for rehabilitation or new construction to encourage

a mix of incomes and discourage the displacement of low-income residents that often

occurs with gentrification. Implementing such programs before gentrification begins in

earnest will increase the effectiveness of the programs by intervening before any negative

effects can occur.

For St. Petersburg specifically, efforts should focus affordable construction and

rehabilitation dollars in neighborhoods such as Roser Park and Crescent Lake

immediately, as developers and speculators will surely start to purchase properties, if they

have not already. The same should be done in Old Southeast and Uptown as they both

will likely follow the same path of gentrification as Roser Park and Crescent Lake. As for

Bartlett Park, perhaps the city may want to encourage the development of more middle-

income housing to strengthen the neighborhood. However, realizing Bartlett Park shares









many things in common with gentrifying areas, policies should be written to prevent the

neighborhood from falling victim to its own success. For instance, amendments to the

housing and future land use elements of the city of St. Petersburg's Comprehensive Plan

could be written to specifically address the possibility of gentrification in Bartlett Park

and similar neighborhoods. In addition to policy changes, programs such as a community

land trust, municipal purchase of residential properties or tax increment financing for

affordable housing could be implemented to insure that low and moderate-income

households will continue to have housing opportunities in the neighborhood.

Recommendations for Future Research

Overall, our model appears to be effective in calculating a gentrification index and

establishing a model for monitoring community change based on trends over long time

periods. However, specific aspects of the model could be adjusted to increase its

effectiveness, calling for additional research:

Studying the change in the same indicators over a shorter period of time. In

several cases, the statistics revealed different trends between 1980 and 2000, and 1990

and 2000. Although comparing changes in values and statistics associated with the

indicators over a longer period of time gives a broader base of knowledge, examining the

short term trends may help to balance the perspective in assessing the likelihood of

gentrification. Since real estate markets can be very volatile, it may prove beneficial to

run this deterministic gentrification model based on ten year intervals. For instance, in

addition to obtaining the index with a base year of 1980, the gentrification index could be

calculated using 1990 as the base year instead. Based on the data collected, the results

would probably be somewhat different.









Projecting beyond the census. Reliance on census data lends itself to inaccuracy

as years pass. For example, the 2000 census could describe 2001 and 2002 demographics

fairly accurately. However, the 2000 census would not reflect 2005 demographics

accurately. The overall effectiveness of the model depends upon the accuracy of the

statistics inputted. Therefore, one may consider calculating projections of the census data,

such as those done by the Bureau of Economic and Business Research at the University

of Florida, for each indicator to more accurately relate the current situation to that of the

base year.

Use of other indicators in addition to those measured by the census. Previous

research on gentrification identifies several other potential indicators that are not used in

this model. However, some data was collected on these indicators. One major indicator of

gentrification is increased sales activity. According to the Pinellas County Property

Appraiser, Bartlett Park had 33 sales in 2000 as opposed to 10 in 1980. Crescent Lake

had 125 sales in 2000 as opposed to only 13 in 1980. Comparison of these rates of

increase to the rate of change in the city's sales activity would strengthen the model

more. Other indicators include the change in the number of residential (new construction

or major renovation) permits issued as well as the number and type of capital

improvement projects planned or that have occurred in the neighborhood over time. In

addition, surveying local residents may identify indicators not mentioned in the literature.

Incorporation of these other indicators not measured by the census as well as those

identified by residents (and not mentioned in the literature) would further support

changes related to other indicators and greatly enhance the effectiveness of the model.









Develop weights and thresholds through survey. Community involvement in

determining the weights and acceptable thresholds could greatly strengthen the validity of

the model, as the value and thresholds related to community indicators are usually

decided upon by members of the community. The weights for each indicator were

developed based on the researcher's interpretation of information presented in the

literature search and the data gathered on each indicator, lending itself to a certain

amount of subjectivity some may consider problematic. More accurate weights could be

developed by surveying other housing and planning experts as well as area residents

through public meetings or written surveys. The range of weights relating to each

indicator reported in the surveys could, perhaps, be averaged to determine the actual

weight used in the model; therefore, creating a better equation with more accurate results.

Run model again in the future to see if results change. As implied by the

indexes for each neighborhood in our study, some neighborhoods are further into the

process of gentrification than others. As neighborhoods, cities and regions are dynamic

entities, the gentrification index as calculated by the model may be different in the future

for each neighborhood. One possible extension of this research would be to re-evaluate

these neighborhoods at the time of the 2010 census to monitor how they have changed

since 2000.

Determine a "tipping point" index and assigning appropriate policies and

programs to specific indexes. One of the major goals of our study is to create a

monitoring tool for use in policy decision-making. Therefore, determining the index

value that describes a neighborhood in the early or moderate states of gentrification as

opposed to when the process of gentrification is fully underway and therefore quite









difficult to address would be excellent continuations of our study. Then proper policy and

programs to could be related to specific index ranges through testing this model on

neighborhoods in other cities to show that neighborhoods with the same index generally

display similar attributes. Similarly, neighborhoods could be re-evaluated over time to

discover how long it takes neighborhoods to cycle through the gentrification process.

Our study successfully accomplishes its goal of developing a model for measuring

gentrification and monitoring community change with results that can have meaningful

effects on policy and program decisions. It is also a good example of how qualitative

information, such as the affinity for architectural style or the desire to be close to the

amenities of the central business district, can be combined with quantitative data, such as

the percentage of housing built before 1950 and the measured distance of a neighborhood

from the central business district, to produce usable information on community change.

Although several revisions could possibly improve the model, it provides an excellent

foundation for future research into the development of more effective models relating to

monitoring gentrification as well as a wide range of other related planning issues.
















APPENDIX A
DATA TABLES

Regional to Local Comparison Indicators


Table A-1: Total population
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 238647 238629 248232 4.02%
Bartlett Park 4827 4269 3912 -18.96%
Old Southeast 2625 2775 2538 -3.31%
Roser Park 2302 1349 1128 -51.0%
Uptown 2250 2207 2034 -9.6%
Crescent Lake 3847 3724 3811 -0.94%

Table A-2: Housing units
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 119486 125452 124618 4.3%
Bartlett Park 2256 2261 1844 -18.26%
Old Southeast 1459 1380 1305 -10.56%
Roser Park 1541 591 332 -78.48%
Uptown 1414 1259 1062 -24.84%
Crescent Lake 2821 2759 2359 -16.38%

Table A-3: Professional job employment (as defined by US Census)
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 23.96% 26.67% 34.05% 10.09%
Bartlett Park 9.27% 7.36% 14.47% 5.20%
Old Southeast 23.93% 25.03% 33.75% 9.82%
Roser Park 9.28% 15.67% 29.10% 19.82%
Uptown 23.96% 15.00% 32.32% 8.36%
Crescent Lake 15.86% 24.32% 31.95% 16.09%

Table A-4: College-educated population (bachelor's degrees or higher)
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 14.57% 18.63% 22.82% 8.25%
Bartlett Park 5.13% 6.02% 6.02% 0.89%
Old Southeast 17.29% 28.43% 29.59% 12.30%
Roser Park 6.08% 6.86% 17.93% 11.85%
Uptown 9.22% 13.95% 14.06% 4.84%
Crescent Lake 13.85% 16.85% 19.97% 6.12%

Table A-5: Age 25 through 34
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 13.02% 14.96% 13.76% 0.74%
Bartlett Park 19.81% 16.34% 13.62% -6.19%
Old Southeast 16.11% 18.27% 12.33% -3.78%










Table A-5 Continued
Area 1980 1990 2000 Change ('80-'00)
Roser Park 14.81% 20.24% 9.57% -5.24%
Uptown 15.96% 19.80% 16.91% 0.95%
Crescent Lake 14.04% 20.62% 17.90% 3.86%

Table A-6: Age 55 through 64
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 12.15% 10.86% 9.17% -2.98%
Bartlett Park 4.56% 7.33% 8.28% 3.72%
Old Southeast 9.79% 8.43% 10.17% 0.38%
Roser Park 7.91% 6.89% 6.74% -1.17%
Uptown 9.42% 7.70% 7.37% -2.05%
Crescent Lake 11.62% 7.00% 7.71% -3.91%

Table A-7: Area Median income (AMI in dollars)
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 11798 23577 34597 193%
Bartlett Park 8135 13224 19125 135%
Old Southeast 10386 25047 31163 200%
Roser Park 7584 11505 19531 158%
Uptown 8466 16824 22768 169%
Crescent Lake 6964 15846 23225 234%

Table A-8: Single-family home value (dollars)
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 35800 62700 81000 126%
Bartlett Park 20600 37200 45800 122%
Old Southeast 37900 70700 85400 125%
Roser Park 19200 45000 68100 255%
Uptown 29000 52500 78200 170%
Crescent Lake 28700 59300 89200 211%

Table A-9: Mean commute time (minutes)
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 19.5 19.2 20.6 5.64%
Bartlett Park 21.5 21.2 21.8 1.39%
Old Southeast 17.5 19.4 21.8 24.57%
Roser Park 22.2 19.7 23.8 7.12%
Uptown 14.8 17.3 20.4 37.84%
Crescent Lake 19.4 22.1 20.75 6.96%

Table A-10: Housing vacancy
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 9.76% 15.74% 12.00% 2.24%
Bartlett Park 17.02% 28.79% 28.77% 11.75%
Old Southeast 15.97% 14.93% 13.56% -2.42%
Roser Park 29.46% 36.72% 29.82% 0.36%
Uptown 15.91% 21.84% 19.68% 3.77%
Crescent Lake 14.82% 31.53% 21.49% 6.67%













Table A- 1: Owner-occupied housing
Area 1980 1990 2000 Change ('80-'00)
St. Petersburg 57.04% 53.07% 55.87% -1.17%
Bartlett Park 36.92% 29.72% 34.76% -2.16%
Old Southeast 51.41% 47.17% 53.95% 2.54%
Roser Park 14.15% 14.38% 23.79% 9.64%
Uptown 31.90% 28.28% 32.58% 0.68%
Crescent Lake 25.81% 21.89% 27.04% 1.23%


Table A-12: Rooms (median number for owner-occupied units)


Area


2000


St. Petersburg 5.5
Bartlett Park 5.3
Old Southeast 6
Roser Park 7.4
Uptown 5.2
Crescent Lake 5.5


Neighborhood-Specific Indicators


Table A-13: Housing pre-1950
Area 2000
Bartlett Park 41.16%
Old Southeast 44.08%
Roser Park 42.17%
Uptown 57.47%
Crescent Lake 56.04%


Table A-14: Proximity to central business district
Area 2000
Bartlett Park 1
Old Southeast 1.5
Roser Park 0
Uptown 0
Crescent Lake 0


Table A-15: Proximity to transport


Area


2000


Bartlett Park 1
Old Southeast 1.5
Roser Park 0
Uptown 0
Crescent Lake 0

Table A-16: Historical designations
Area 2000
Bartlett Park 0
Old Southeast 4
Roser Park 1


ition corridors (interstate highways)










Table A-16 Continued
Area 2000
Uptown 0
Crescent Lake 2

















APPENDIX B
AREA MAPS


:". N N ---
.: : : .
i r-r.b- | "
C : RES- :
hWE 'NT'*
A.-'rrri,'.rJ UPTOWN! .
'=..-r- h -: .+l^ ... .. .

.. ... .E_. .. A t.i 9 ii *^
S : I- : .




"* ROSER PARMK A.nt




PA7fZ i hESH
S 2. nd .1 Lak.-i. Av


.A.. : *b ..I .t-"N I-,D
Figure B- Neighborhoods
Irf(l $ P ;* '" **S



y -.. ---. '- :.:. .. F,*.
.. .. -. n.






Figure B-l Neighborh goods


Source: Yahoo! Maps i/m,' /maps.yahoo.com)


SE
*,'"

































Figure B-2: Census Tracts:


Source: Florida Geographic Data Library (www.fgdl.org)











APPENDIX C
GENTRIFICATION INDEX


H


1~4'ii.


Figure C-1: Gentrification index

Index Value Ran2e:


High : 1.0

Low : 0,0


Census Tracts:

204 = Old Southeast
205 = Bartlett Park
213 = Roser Park
234 = Uptown
235 = Crescent Lake















LIST OF REFERENCES


Anselin, Luc. 1998. GIS Research Infrastructure for Spatial Analysis of Real Estate
Markets. Journal of Housing Research 9(1): 113-133.

APD, Inc. 2002. The Fifth Avenue/Pleasant Street Neighborhoods: Controlling
Gentrification. City of Gainesville, FL Community Redevelopment Agency.

Atkinson, Rowland. 2003. Introduction: Misunderstood Saviour or Vengeful Wrecker?
The Many Meanings and Problems of Gentrification. Urban Studies 40(12): 2343-
2350.

Berry, Brian J. L. 1999. Comments on Elvin K. Wyly and Daniel J. Hammel's "Islands of
Decay in Seas of Renewal: Housing Policy and the Resurgence of Gentrification" -
Gentrification Resurgent? Housing Policy Debate 10(4): 783-788.

Birkin, Mark and Graham Clark. 1998. GIS, Geodemographics, and Spatial Modeling
in the U.K. Financial Service Industry. Journal of Housing Research 9(1): pp. 87-
111

Can, Aayse. 1998. GIS and Spatial Analysis of Housing and Mortgage Markets. Journal
of Housing Research 9(1): 61-86.

Chapple, Karen, John V. Thomas, Dena Belzer, and Gerald Autler. 2004. Fueling the
Fire: Information Technology and Housing Price Appreciation in the San Francisco
Bay Area and the Twin Cities. Housing Policy Debate 15(2): 347-383

Clark, Gordon, Andrew Goudie, and Ceri Peach. 1996. The New Middle Class and the
Remaking of the Central City. New York, NY: Oxford University Press Inc.

Downs, Anthony. 2002. Have Housing Prices Risen Faster in Portland Then Else-
where? Housing Policy Debate 13(1): 7-31

Freeman, Lance, and Frank Braconi. 2004. Gentrification and Displacement: New York
City in the 1990s. Journal of the American Planning Association 70(1): 39-52.

Galster, George C., Roberto G. Quercia, and Alvaro Cortes. 2000. Identifying
Neighborhood Thresholds: An Empirical Exploration. Housing Policy Debate
11(3): 701-732.

Goodman, Allen C., and Thomas G. Thibodeau. 1995. Age-Related Heteroske-dasticity
in Hedonic House Price Equations. Journal of Housing Research 6(1): 25-42









Higgins, Lindley. 2001. Gathering and Presenting Information About Your
Neighborhood. Washington, D.C.: Local Initiatives Support Corporation Center for
Home Ownership and Knowledge Sharing Initiative.

Kennedy, Maureen, and Paul Leonard. 200 la. Dealing With Neighborhood Change: A
Primer on Gentrification and Policy Choices. Washington, D.C.: Brookings
Institution.

Kennedy, Maureen, and Paul Leonard. 2001b. Gentrification: Practice and Politics.
Washington, D.C.: Local Initiatives Support Corporation Center for
Homeownership and Knowledge Sharing Initiative.

Land, Robert E., James W. Hughes, and Karen A. Danielson. 1997. Targeting Suburban
Urbanites: Marketing Central-City Housing. Housing Policy Debate 8(2): 437-470.

Lang, Michael H. 1982. Gentrification Amid Urban Decline: Strategies for America's
Older Cities. Cambridge, MA: Ballinger Publishing Company.

LaPeter, Leonora. July 18, 2004. A Revitalization Riddle. St. Petersburg Times. Section
B: pp. 1, 6

Nelson, Kathryn P. 1988. Gentrification and Distressed Cities: An Assessment of Trends
in Intrametropolitan Migration. Madison, WS: University of Wisconsin Press.

Peng, Ruijue, and William C. Wheaton. 1994. Effects of Restrictive Land Supply on
Housing in Hong Kong: An Econometric Analysis. Journal of Housing Research
5(2): 263-291.

Phillips, Rhonda. 2003. Community Indicators. Washington, D.C.: American Planning
Association Public Advisory Report Number 517.

Quercia, Robert G., and George C. Galster. 1997. Threshold Effects and the Expected
Benefits of Attracting Middle-Income Households to the Central City. Housing
Policy Debate 8(2): 409-435)

Redfern, P. A. 2003. What Makes Gentrification 'Gentrification'? Urban Studies 40(12):
2351-2366.

Smith, Neil, and Peter Williams. 1986. Gentrification of the City. Boston, MA: Allen &
Unwin.

Solnit, Rebecca, and Susan Schwartzenberg. 2000. Hollow City: The Seige of San
Francisco and the Crisis of American Urbanism. New York, NY: Verso.

Turner, Margery Austin. 1997. Achieving a New Urban Diversity: What Have We
Learned? Housing Policy Debate 8(2): 295 305






70


Williams, Brett. 1988. Upscaling Downtown: Stalled Gentrification in Washington, D.C.
Ithaca, NY: Cornell University Press.

Wyly, Elvin K., and Daniel J. Hammel 1999. Islands of Decay in Seas of Renewal:
Housing Policy and the Resurgence of Gentrification. Housing Policy Debate
10(4): 711-771.














BIOGRAPHICAL SKETCH

Ashon Jahi Nesbitt originates from St. Petersburg, FL the area of focus for the

study in this paper. He spent his entire childhood there before going on to attend Florida

Agricultural and Mechanical University, where he majored in Architecture and

participated in the world-renowned "Marching 100" as well as gained other campus

activities.

Ashon graduated from Florida A&M University in the spring of 2002 with a

Bachelor of Science in Architectural Studies. Although Ashon sought to pursue a

professional degree in architecture, he found his home in Urban and Regional Planning at

the University of Florida after a year of unsuccessful attempts to gaining employment in

the field of architecture.

Ashon chose to concentrate on Housing and Economic Development. Ashon first

became interested in this area due to exposure to his mom's professional career, who

worked many years in real estate and as director of a leading local nonprofit housing

agency in the city of St. Petersburg. As a student in the Urban and Regional Planning

program at the University of Florida, Ashon has cultivated that interest through

coursework, employment as a Graduate Research Assistant with the Center for Building

Better Communities, and attendance at such conferences as the Florida Housing Coalition

Annual Conference.

In addition to his academic pursuits, Ashon actively participated in the Student

Planning Association, serving as the President for the 2004-2005 school year. He also









served on the Florida Chapter of the American Planning Association's (APA) Student

Council for that year, Student Representative on the San Felasco APA Executive

Committee and holds memberships with the American Planning Association and Florida

Housing Coalition. Beyond the department, Ashon participated in the Black Graduate

Student Organization, serving as Vice-President for the 2004-2005 school year, as well as

church and other activities throughout the community.

Ashon hopes his educational and professional experiences will land him a position

with the Department of Housing and Urban Planning, where he hopes to hold the top

position one day. Ashon ultimately plans to obtain a Ph.D. in Public Policy, become a

developer, focusing on urban infill, affordable housing developments and to teach at the

university level upon retirement.