<%BANNER%>

Assessing Dimensions of Disruptive Child Behavior with the Eyberg Child Behavior Inventory

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 E20101112_AAAADO INGEST_TIME 2010-11-12T18:44:25Z PACKAGE UFE0020122_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 6861 DFID F20101112_AACGYW ORIGIN DEPOSITOR PATH stern_m_Page_31thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
a3438e7f4fd21951001fa64440433598
SHA-1
bd2572a2c9d427656643ac505bd21422a933f1ca
24325 F20101112_AACGRA stern_m_Page_19.QC.jpg
89fdf7dd8b0df82bebab68f83acf6670
d33eac3afae178db1a8a82e68454e0a2a5785fa0
5019 F20101112_AACGTY stern_m_Page_02.jp2
cdfc0a475d3b1b0d5f913666b75dbff3
0f9feecd046014fa29c7e53a1ffa61bf6d60fede
26002 F20101112_AACGRB stern_m_Page_32.QC.jpg
0c68a295c8b7d2b979a9f209a3aba2bc
0fb3da6084173fa19bc8a67fc0efe5cb4b4ee7b0
26081 F20101112_AACGTZ stern_m_Page_03.jp2
f77d233b8bab5de81aeba836b91c5dbd
87a521b7beb8801c6c35a40e91b6d39d522dfff9
2319 F20101112_AACGYX stern_m_Page_01thm.jpg
fcb349b20082d91a06f4be6715c2b4bd
38fdc373e2877f16ca0de2cb391e43125abb8298
22959 F20101112_AACGRC stern_m_Page_03.jpg
b05007d5ff2a09a48b5dbd4960e6d191
220f5f19dfdf8b72d874fa1ff3965987c10dcab1
7831 F20101112_AACGYY stern_m_Page_05.QC.jpg
21336556cb02d90c00be2a5b5ce89748
db245fc7ae0c7955846525fa98147685decc7632
10832 F20101112_AACGRD stern_m_Page_26.jp2
1e198dec476e8f583d3fb8c3e662ccda
e589a804cada248702cf6da70ddbaf0eac795d4c
1053954 F20101112_AACGWA stern_m_Page_31.tif
212de253e0b949d5fa75dc8a191faa66
33f0eb56d5720650d7471f07c95608f9387c2af0
6743 F20101112_AACGRE stern_m_Page_24.QC.jpg
54a634225330b7c01013ac52bc949334
abaa661dc5fad1d50602ca126ca7039e04d53bbf
F20101112_AACGWB stern_m_Page_33.tif
213c34eabf530f941f61688ac86bfeb6
85db353101ebdfafb5ba36feb19d6b415fb2fcb0
6393 F20101112_AACGYZ stern_m_Page_07thm.jpg
547b7b8c2b17a9f22818e873dc820a5d
4558fc8cd8f36ccb4165d216782b1af03a99f239
1639 F20101112_AACGRF stern_m_Page_37.txt
58994688fec6a9438398f2411dde3f73
2472d4bf75b8dcf9372f87c9f60b9a9b343810da
F20101112_AACGWC stern_m_Page_34.tif
2d862f049c3e11f9f482af3f319deb1e
d112bca05c3128661fe83e6d495fe5f514528e65
3076 F20101112_AACGRG stern_m_Page_02.QC.jpg
cbab08456c0e626953b3b0bcc6cfa70f
1372086427ff7cbe9acd154e9648cca163aafc65
F20101112_AACGWD stern_m_Page_35.tif
7c2d9034515cf5362a92f0dc7187ec80
7ae5806d374c47e90626dc0b8937be8b67de83eb
2092 F20101112_AACGRH stern_m_Page_15.txt
adf0a82235883b6c8b19bfd7f553c6e1
bf03c88a11db3f04ce1d7efc7db55cdfac95592e
F20101112_AACGWE stern_m_Page_38.tif
cd1322926612eee4fb317dac74f08b9f
d2ce7d77c70ce23023faa0afa58dc130f9eaba3a
18286 F20101112_AACGRI stern_m_Page_37.QC.jpg
374eb5ddf7565fbf79b615385e11e26d
77cb2e4331c8181f555d4d973f95b1a83d0c0731
8238 F20101112_AACGWF stern_m_Page_01.pro
e9be11febf62d16d2d55cb7c430944a7
d63dcc57f59778c877faab861444e0fd356bbd80
1966 F20101112_AACGRJ stern_m_Page_18.txt
698111155fff84af04e48ed7acace1ce
2e30246655f294992ed448d36fbe7109b30cceec
884 F20101112_AACGWG stern_m_Page_02.pro
e981bbfca1f338bad37ed1ec3112114d
2e75c0eca5b2fb7c74515f9b68e79b9173c93749
10354 F20101112_AACGWH stern_m_Page_03.pro
28cbcd58b04fba3f206147e19c076184
e98e87e64066a3e957089dce23179d628a7a1a02
115719 F20101112_AACGRK stern_m_Page_09.jp2
850e6ffc31eb17e6d5c21ea1494d91ed
d6d81475cbbabbc93f56d85d96fedd7822a3894a
19016 F20101112_AACGWI stern_m_Page_05.pro
616b3337626736998e6c0300beae85ca
e2438a14c6d4957263634353a3c71eb8bd4a54a8
F20101112_AACGRL stern_m_Page_36.tif
c9d672bdbd505c6088c54e1892f72e85
425fd6ba282a39115bbd5f1edce5742e1ab6939c
5033 F20101112_AACGWJ stern_m_Page_06.pro
7acfcf7c0aba7c103d637b92c58830bd
533dd8875230856ac435c41ca8b9ef19252b471a
F20101112_AACGRM stern_m_Page_37.tif
3d28c3e87eb05d5334f37ef07e080ea0
0cbe6ec4bf8b22bf4dea65de8afedcb57c0ce2ff
47620 F20101112_AACGWK stern_m_Page_07.pro
78c4b44a3093eb87f092a78c36fe18a3
5eabb1047a0755a2a8ee103e5032c021194982a0
F20101112_AACGRN stern_m_Page_28.tif
7d1d3089b3c335de757ee085af1ed2f0
0c068452b02b301b2cf1ff6c1d0d122da2f60427
10276 F20101112_AACGWL stern_m_Page_08.pro
6bc43573ebb064b15e9f8c0b689b1517
cfae128d019d2f82309fa6189ec61354200aa4d6
6948 F20101112_AACGRO stern_m_Page_30thm.jpg
5adf358132b88bc988284bf49b7a3f20
7f27e455c6e7a1f1230a68838a8d8088fa15276b
52933 F20101112_AACGWM stern_m_Page_09.pro
44731fdc2cf7f74083bc990542073c2f
6cb0e2d9b328d95a91ae2c32b40c8113cc808329
40615 F20101112_AACGRP stern_m_Page_25.pro
146b9a66699c69108d57920115c88ece
eb72db67f4ea94df8b16b46a62d7337105866470
54868 F20101112_AACGWN stern_m_Page_10.pro
e0cf7bf6fad2de19847c387b54ff5987
923a7809844317c4aedc78751be2dd1ce2fce20f
2511 F20101112_AACGRQ stern_m_Page_05thm.jpg
c0a586ecd5f0045b36f3f575728114f0
d23518d26b8ad043227742993311029cbf14d186
53591 F20101112_AACGWO stern_m_Page_11.pro
27e0fbb7d532732ee9302fa385a9098a
f3f44b4f9643bb97eea08ef13e74b453a378b135
25271604 F20101112_AACGRR stern_m_Page_05.tif
de5a96c65a2b6d25d1c7cb9eaee774a3
4c6c4fc61211fc52cd717aecdfd9f2d1feb7d6c6
53092 F20101112_AACGWP stern_m_Page_12.pro
01b7ac195b14a15458bb03ac61b5bf44
08a843486bfe98829ac8536d814aa3cbad2d9a08
54074 F20101112_AACGWQ stern_m_Page_13.pro
b95a8ac849ee98b4f6db2a646289ffba
ab238e2dafad29aeba7f1e3defdefa5586877443
6092 F20101112_AACGRS stern_m_Page_24.pro
38ea11f74258b8cabcac42e73cef3262
d6d3a767625e908ca36c37c25afbc62efdd61211
53285 F20101112_AACGWR stern_m_Page_15.pro
0fd4a0fa586cb833676fd513024aa949
777d9dc413902082a16d8949e0a4bc63b710f514
4387 F20101112_AACGRT stern_m_Page_04thm.jpg
60cbbfa9f6a5e0654b4b259f77e494d2
5a506991c73c7a8b4c768e7f858b263761d3f6cd
51637 F20101112_AACGWS stern_m_Page_17.pro
72c01aa623cd296eb4b672bad7c291c5
6ff40207dcfbe65363fbfbc0b0713947c6e7091f
27271 F20101112_AACGRU stern_m_Page_21.pro
02ec83cc6d0af63fa9a6a31844aada33
6eef51e6c1f76629a3948723a2ea29f05ddd36e2
48891 F20101112_AACGWT stern_m_Page_18.pro
e2544c66cd46a0acc0e25cb1c9a477d6
4f04caa314082f65fea65bead56507a26b26f917
F20101112_AACGRV stern_m_Page_15.tif
4c88ef76cddfdd891d90c829bb716c00
864fa385abd5e56f05dc32d6a9f52d98ca2b08ba
49295 F20101112_AACGWU stern_m_Page_19.pro
eed4b79b6db81bb982cf31b2089751cd
230438a0992edb03c6a5599cd3ea50325dea8c59
127939 F20101112_AACGRW stern_m_Page_06.jp2
9157c9f4256c4853c77f75769c7ffec6
bf3fe1cdcdb9a66aada12028220d92f12892808b
48368 F20101112_AACGWV stern_m_Page_22.pro
ba2485cbac1ce7ffaffb1fa09ae6dce2
b3b463ad7db981261b457ddd089b57bf619adcbc
498 F20101112_AACGRX stern_m_Page_01.txt
e589cbe380896f249a9af2d4c1cb2bbd
23e7e5cdf9c860d5979d8879ad87a6459f03b3c4
35781 F20101112_AACGWW stern_m_Page_23.pro
1718e9210d6f3f3fc9e2b833138faa05
8e22d8c3128ed128919a963ad4bd18a47f2bd206
9832 F20101112_AACGRY stern_m_Page_27.QC.jpg
55eecdbdb06d9f7c6e9207f231df5841
4fec8261ce0aca1761caa5cdefea0ceb93607773
450 F20101112_AACGRZ stern_m_Page_03.txt
ee768df8e982d72cc73e59f05b353e1d
44341c1dd22ff848d570faa4c8e510d84917cb46
3975 F20101112_AACGWX stern_m_Page_26.pro
baf1a50d6667374a43de94ddcd70075c
af2398d4b7ef7c6b919758d848c8238e9f9232d6
1051954 F20101112_AACGUA stern_m_Page_04.jp2
15c3a420b62f62f22c0d2cde40463c57
a393655aef73ba7e971ba1a7922fd5880de2dbbb
23517 F20101112_AACGWY stern_m_Page_27.pro
43685367f0759764752fa455808d7464
1d315ede9bda01650b1d0cc222dbb304672c6bc7
570582 F20101112_AACGUB stern_m_Page_05.jp2
322019ae1ce6d0a3cf9be402bde1ac17
7d091568f7c79dc7ad21b48977bc66af33830501
13621 F20101112_AACGWZ stern_m_Page_28.pro
cd7050e5ab626461ae76581ac81cb5a9
bff69628a883bf44d7c195decacf5fdf5af32b90
104351 F20101112_AACGUC stern_m_Page_07.jp2
085ad3dac8d121152e7afa3957547b36
30f9b3cdc91459edb4a20e65a03044cf3dde1ff9
25350 F20101112_AACGUD stern_m_Page_08.jp2
e3178ac4129fb0b12ef91ac30fca552c
3fa8be5d9513986c4a8a34cc711986b1b92bf47e
2989 F20101112_AACGZA stern_m_Page_28thm.jpg
866a8589cff1c2ea8a173faa139d1321
5aeb80ac00f8704b118e48276cb7c3d7a84bd8cf
120516 F20101112_AACGUE stern_m_Page_10.jp2
c60dcfd8b6218f31810a944e45818f35
5dc05464633c5a1584c9f8ab75b235096df5e0aa
6907 F20101112_AACGZB stern_m_Page_12thm.jpg
31d39cad334485c4240280b3112c46f9
c387d6f82a9550b943c938736bc18339a844f386
114088 F20101112_AACGUF stern_m_Page_11.jp2
27fecc7b05a354c8169870cf42667002
fc576190f470664cd809953ddb1ed745090b15c0
25052 F20101112_AACGZC stern_m_Page_31.QC.jpg
547a7babaf9e5d9ac630fdba615f6e85
a62638ed45d177fa9cbccbd6a9174af7d55a6a52
6463 F20101112_AACGZD stern_m_Page_29thm.jpg
39a0d2aaf15f55fc7d6cd3253a639a3b
7ea0f390b35e049dffb7c715173ad0ac21b8409e
114041 F20101112_AACGUG stern_m_Page_12.jp2
cf00c4cd658debaa4ee218347657236f
1d05a2e9b652e8749a8d6bd2e845e8b1892aa881
5167 F20101112_AACGZE stern_m_Page_23thm.jpg
902d0de0641129d5a49a580388ee8db6
c650fe051283ff253eeb0496fafde7dd96a2c57f
115397 F20101112_AACGUH stern_m_Page_13.jp2
fde267f0e3ae1bf46edb0ee67ec9f35c
330142669d5f45cae0dd9ddb1837bf0fbd9c1c83
25776 F20101112_AACGZF stern_m_Page_30.QC.jpg
ff1e028b6b7877fff2dae3a96b574f3c
687d435e8e516f191cfead54c860f2e7d97c18de
11422 F20101112_AACGUI stern_m_Page_16.jp2
c6496382dbc15f54161c6b592548e878
656b16a36980655c17f9053c03837150b21e9343
1645 F20101112_AACGZG stern_m_Page_16thm.jpg
7f75ee9d4e697e9c6e89fa2ec4d0b1bc
6a636c9f4a98a330df133aa88975044a19a6b8c4
112268 F20101112_AACGUJ stern_m_Page_17.jp2
76d342a13e108f399212b484e67e3e08
4f95a60866c623d8771b22904090711a63ad030b
26699 F20101112_AACGZH stern_m_Page_36.QC.jpg
cae9fce691454260231413ddac8d2c71
5845edb749d3e7423e7747c5fade11ee4491f6f7
109157 F20101112_AACGUK stern_m_Page_18.jp2
d14394d26eea2494ec26d76fa4f42d74
57e1d8ffc1b43afe55715b18352abdb58bad294c
2270 F20101112_AACGZI stern_m_Page_03thm.jpg
070449562cd2fb05ac89fafc192b8db7
562f4ddadeb0ab808db18754ff16d62ca6ad3fa2
109947 F20101112_AACGUL stern_m_Page_19.jp2
b705f364eb1ef6fd4685ab2519cbce3f
4d24cf87d5039c45e832f8524485fd2df2ceab21
24396 F20101112_AACGZJ stern_m_Page_12.QC.jpg
b63a22e88f6a61bf14aa01d4da9ae331
55932a9cde30f4422fea03fa12bdbda4838ecd38
118645 F20101112_AACGUM stern_m_Page_20.jp2
37783e0c0eac1c7e674910b0ed1cb9fb
24209e71aa8cbb8de552f6cda9cde3cb3d42fd9c
60140 F20101112_AACGZK UFE0020122_00001.xml FULL
3831ed8be4484a6c0ad10b225d878aa7
1872232b81114b06d2ab89b99a7117f06f2589b5
63813 F20101112_AACGUN stern_m_Page_21.jp2
d3a436d3b29d18de6b0703d853022e72
b5f7c2fe401d745a1ee6b2dd93fc4a8da9d03c92
7364 F20101112_AACGZL stern_m_Page_01.QC.jpg
0114bdfb6e58bfdfd5e7d0cb52069aed
1cd4211f25fed762129786e68fa144954c90ba04
103619 F20101112_AACGUO stern_m_Page_22.jp2
de2e8ec7564c15bb00b0326d10bbbe44
0a30b57434e48d5ea541b864a56d2ad00590f621
1332 F20101112_AACGZM stern_m_Page_02thm.jpg
b614c566c4e7ac2104753d99febde8a4
729b40aed817c16a040d3ac91899edc5ad161597
19206 F20101112_AACGUP stern_m_Page_24.jp2
0d68d1e47da19cb7019c6ad2386e97f8
98deba67444990d57170ff3dfaa9415583354893
16266 F20101112_AACGZN stern_m_Page_04.QC.jpg
a7763568c4276fb2935ae01d3151f387
9a7dc69581bcf3680df66186f804e973e93aa483
82753 F20101112_AACGUQ stern_m_Page_25.jp2
5eeae290bbb71b32311373d834e654b3
9db71b0cf165f35a8366048375a33bcb08dfc81a
4046 F20101112_AACGZO stern_m_Page_06.QC.jpg
ee02d86d1e9bc167cfb62ea319bdc70f
58ce17fe2f583c40f1b0dd0f2829948b75ab0f7c
37958 F20101112_AACGUR stern_m_Page_27.jp2
099af9e3a735f8d112dece92c08a077a
60d927eb2ae790ed79b7d8d3393afe74fafc8254
22152 F20101112_AACGZP stern_m_Page_07.QC.jpg
2fbaf6c516222ce0134f9adfbd28cd4a
c20fd009c6a5ff6074383c791cf257bedb21858f
37835 F20101112_AACGUS stern_m_Page_28.jp2
e8832578668cb2c6274a766e6d2454b3
843dc343bc850dbd04f0e3c6ae31098f4b5107de
6856 F20101112_AACGZQ stern_m_Page_08.QC.jpg
bd1056c44332544e9243a296269753ea
65fcf100639e800481cb092a329f7bd0bfebae2b
105212 F20101112_AACGUT stern_m_Page_29.jp2
56bd15602de86f6388d599dadcb33bc5
5dd0ea385059deaa55e698286ee43984a875e471
6858 F20101112_AACGZR stern_m_Page_09thm.jpg
510078ca698f71bd6a60370e255f9986
4f8e8dcaebfe922870504629c110615c08923266
115034 F20101112_AACGUU stern_m_Page_31.jp2
01e9bc21538cb0420de192d587b649f2
3f7efe88ec43be3b9427cbf83b14a51ce52b9e37
24824 F20101112_AACGZS stern_m_Page_11.QC.jpg
1f48a3de490ae0dcc2a127f9e65a1325
40b71956f28f7ecdb25fe14133c05003d8aa7c76
6958 F20101112_AACGZT stern_m_Page_11thm.jpg
e4e931b72a949277c4cc09b37f6ea99b
720a36d0c5cf28d075922dc3b40c2c22aced4df5
119435 F20101112_AACGUV stern_m_Page_32.jp2
7cc09a81f9fb46a0e3913a984c995f32
d445d34cc93882f7c27680d0c546c0c6371f415b
26677 F20101112_AACGZU stern_m_Page_14.QC.jpg
9fcccd82f22c9fd68be39d5d3472d509
5f44c663f056e014ed7086bc4eb6fbedabe7749d
110153 F20101112_AACGUW stern_m_Page_33.jp2
7c4fc14fc6f6e7a0857e39e8780b15bb
218a960a21639b72644cd411c766b016a165de72
7140 F20101112_AACGZV stern_m_Page_14thm.jpg
7138daf3a3cdd2c345e0311e9788ab31
415f8d5695b1440b4c76ba415869b15cffbbff41
118675 F20101112_AACGUX stern_m_Page_34.jp2
d2a4c1ab3c2fb977158aaf5bfd75b6a4
6ec65d6f7b62355fb13262d487fdc37be0c5068b
4440 F20101112_AACGZW stern_m_Page_16.QC.jpg
90d6779ebb732396e2ae5aa3117b9a7d
c3a534ca57052e751dabe68ed81ed39be35a0bfc
F20101112_AACGSA stern_m_Page_02.tif
52e876a9c6e7c78e773ab406fa82af56
ea7b73a729661b3e08894842fc2254c922396c02
29308 F20101112_AACGUY stern_m_Page_35.jp2
108a473aaa5e380044617be23feaaa63
fd2ae747a1b0299b7f8ac7049070ea8fe8dce71c
6677 F20101112_AACGZX stern_m_Page_17thm.jpg
595693ca51f67f0127249017fb5db5d5
3c808a2f6eaebb930db1aa1e4eef5a1faa358f65
576 F20101112_AACGSB stern_m_Page_28.txt
d9fb0e1da95d12c1e75e8301312be074
a3903a9691638b93d98bbf374a29b2503f112620
138976 F20101112_AACGUZ stern_m_Page_36.jp2
ee788027a50a11f0724b0c61c56dd613
0b2b53e74757eca6e7bec351bde9c019db187d1f
23251 F20101112_AACGZY stern_m_Page_18.QC.jpg
91a1cec35366956fd0364ebe77ac68b8
73e08f014bd90ee1d04901ca409f67f2f7ccda1e
48828 F20101112_AACGSC stern_m_Page_29.pro
9ebbd7ed40d348135795da816e6c3020
95a11d3e080a3fc2220202f3c6d4c6d424fc5aac
55099 F20101112_AACGXA stern_m_Page_30.pro
9e574a8e4203a5f4ed6ac918ff86e568
cb3a57a821b83b261b960db99cab2a21266a2c5e
6595 F20101112_AACGZZ stern_m_Page_18thm.jpg
15002891afc40d738297cc265d302d7d
da5de5f4b443e5fb8f82f4c7f07c1234975032ce
78678 F20101112_AACGSD stern_m_Page_30.jpg
97ef7b4c63d1602bd8000ef8b0c4c1e7
89011a7fd38de3cb4d04d1ac4b8881db1922e1fe
53526 F20101112_AACGXB stern_m_Page_31.pro
3c0c7a74127050240ba77ea45c1d0394
2a79db55fe0eac768a01ff2ed16b50ea7781bc45
2147 F20101112_AACGSE stern_m_Page_17.txt
e50ae3fb057b0a1635b71766b1152fea
476701f41012fa427f03bb4c134a369cab4f7f6f
114857 F20101112_AACGSF stern_m_Page_15.jp2
23dee73661c0fef558be181e218286dd
bbbfdc0e2ed57e5fe403c1712dc13184ac60c7ca
55327 F20101112_AACGXC stern_m_Page_32.pro
da342d6a2cc7d1f0e817abc91563d417
d9df9ab2b9d938ed7298ef5f74311130b38a3704
70561 F20101112_AACGSG stern_m_Page_29.jpg
ba395521cb67f95f7491023a61f3846e
7aee6a00015b3f45c443ff414e81e4204a03bb3b
51041 F20101112_AACGXD stern_m_Page_33.pro
19f5e8c0dc420f5bb8b4aa7bdc707404
18eae7cafee4d5db0ff167ccbc0ea410e3cb5d3f
26404 F20101112_AACGSH stern_m_Page_10.QC.jpg
80536169defdc97b58cd2eb80c632b71
56133d6c9f81508eae370a0ff6004b784d8d350f
12122 F20101112_AACGXE stern_m_Page_35.pro
0b75b15cf7767391aa9e97f5f3d262d2
663ea240942d7bf942f727758e8a39e494ecb302
81845 F20101112_AACGSI stern_m_Page_23.jp2
2526e38d7cde7b2e370a8f1d8e90c9c7
67cd839e20c6c4aac783008acb3a77247c0a4932
63952 F20101112_AACGXF stern_m_Page_36.pro
ce465c6b61143c81f3d3aa5ceb6fe944
c24b93974078323030d20c0539abc7b973d5b379
2255 F20101112_AACGSJ stern_m_Page_14.txt
0183af05811935b373f88a9d0081fda1
d845253cfcc4a17a1b63f84f186af98437f6e27e
40157 F20101112_AACGXG stern_m_Page_37.pro
2d09822806212dde216fbbd44cde9532
339400a1bda2b459f8e1e0aba17b2fde1ba6ecf7
118727 F20101112_AACGSK stern_m_Page_30.jp2
d7d7cc9084b70ab672becb69c967ca11
941c4373eda64932879a1ed9860e20a78094ab3d
31758 F20101112_AACGXH stern_m_Page_38.pro
fe9a75e8e871cef80fb18316dccb8d36
1cc499b745347f5c932fbb4116f03429c0639e04
77221 F20101112_AACGSL stern_m_Page_09.jpg
8d4d9516ded956d98a86b246db6dbaf1
5386ee8308993342e50053e7d8548f30ece33bbf
2729 F20101112_AACGXI stern_m_Page_04.txt
91690b208c4358e86523078e2eca7a60
3228f2ac9e20e00758f7e2255f16d7916535dc58
1747 F20101112_AACGSM stern_m_Page_06thm.jpg
6629d840d82be458414bcb1b57ffb1ff
d5a1a959f9f9a5d28e54dc6b974487bed4d2dea0
806 F20101112_AACGXJ stern_m_Page_05.txt
0ac6ef2274014e338a79948251080bd9
088a57c9099c795610a04c9fe2a7ee458c729650
90656 F20101112_AACGSN stern_m_Page_37.jp2
523db37470ee5a4bdb96a1e36de2266e
0a5c1b940af9d0103893fc3e38383cdaf1fe883b
231 F20101112_AACGXK stern_m_Page_06.txt
4975a37035ed066ffc36d18b06e0b192
3c1d87e5f6953bc9f412632a48aa9cf64dd5872c
123972 F20101112_AACGSO stern_m_Page_14.jp2
345d3d26e347e0e4ff8db7cadbcb156b
99a94429b2e7d6a45d5ac148297d4e0502a5751b
2079 F20101112_AACGXL stern_m_Page_07.txt
d140ab8df7e6d9aaf19f13b2220816da
a3bb6a7e6c867e055210806c474076177da632a6
79462 F20101112_AACGSP stern_m_Page_20.jpg
4d47a9049e3167651f76488e7e4cac07
8f3dd2639109783738eca7a3f5764000bd7b51e7
411 F20101112_AACGXM stern_m_Page_08.txt
0c7caebd0c3f5cacaa9ae82e5599122a
b5be024aaace714d9ee90c415756c393fcf19283
2171 F20101112_AACGSQ stern_m_Page_10.txt
f238b7dc2b87355d3abf1bfec40e2cdd
688855e7349d5f0522bfaaa896549b3f798823aa
2174 F20101112_AACGXN stern_m_Page_09.txt
5a7e1ee332b4466ff5d00fa83bdd1cae
e178d0347beb261f490e49448556e3fdd5073321
46776 F20101112_AACGSR UFE0020122_00001.mets
466a40c085ded38d6ca168a60065c7ed
82a55fef4ab86b7ff11ba6eadf760516c605135e
2112 F20101112_AACGXO stern_m_Page_11.txt
37d19de69cbf9ca8d914f00b0112d683
020bd5796af1f4053fdd95c45919e7cbce73a48e
2084 F20101112_AACGXP stern_m_Page_12.txt
f8251043f85f54799691db499bf08c0e
63abeb7f21bc36d5b55bda4cafc42ed5e36d21a7
2129 F20101112_AACGXQ stern_m_Page_13.txt
2d7b6b985fc57d217971e401bb4df78e
ff2d1786a0d2697a7dbeb42b57bcb60588031828
148 F20101112_AACGXR stern_m_Page_16.txt
558fef01e346acaed7228347595d8a75
43652cbd59e7d01a3e149dd64902ea03a918d984
24233 F20101112_AACGSU stern_m_Page_01.jpg
50b30b951109b20c083e63cadf87dff3
28c4b6ab67a82ee7cf64359d626a9bfe92487d72
2011 F20101112_AACGXS stern_m_Page_19.txt
9b04f2cae26e96fafb5dedce4b4f81d8
a7ae67a505c2d133c75a6fcda839e31d7a8a683c
9885 F20101112_AACGSV stern_m_Page_02.jpg
6abebdf29c706e1971fa57a870392481
4be28b43c0d0340979d636c603f76967ced3999b
1089 F20101112_AACGXT stern_m_Page_21.txt
18b6a9fa911bd8c30a38a9dcc61fcc09
270434f7a9b75d198a455d812e027ea6b3d67320
55839 F20101112_AACGSW stern_m_Page_04.jpg
16b16010fbb25d691266c43f20a8275c
9a69077abdbc6bab9ec2aeaffa67ab92ad29daa3
1468 F20101112_AACGXU stern_m_Page_23.txt
8be7bedea2d9a4b3e742f8009076e8c9
9ff8fefcd7c4e0390f35ada78f899f9d68d748db
13285 F20101112_AACGSX stern_m_Page_06.jpg
70182e1cb3d642bdb47ff96503e4cf9f
c1d50b351fc42386b5c534931d18debaebc3f8fe
395 F20101112_AACGXV stern_m_Page_24.txt
3d6ad5db836148f08037ca91fd0bafcb
8fc642e4c2f7f0fde8cd19998beecd54869764ae
71706 F20101112_AACGSY stern_m_Page_07.jpg
9da936dd02c6ce06dd4ae2cabfa6ff86
da254534a0589d54a73406186e9eea238cab4a40
1731 F20101112_AACGXW stern_m_Page_25.txt
746aa8301b8d4dc234f3ca590a937dc1
e927e8da007deec90e8fa4851c1acf5afa7143f0
21076 F20101112_AACGSZ stern_m_Page_08.jpg
25b4ee89c08243e30117eb04743b292e
428e3ef60c2f78b2e14fed7e8abda98980d16545
181 F20101112_AACGXX stern_m_Page_26.txt
68e7443f7fda3a45b1026cb878023a51
17be749a7407b8819d08732e084a9f2e110e7a8d
73209 F20101112_AACGVA stern_m_Page_38.jp2
0b82e154b13de2e84bc56e94b11b7a34
36a4fae0858ceb7c3148b8374bec4338c2fc3191
1295 F20101112_AACGXY stern_m_Page_27.txt
9acd2874f444dc6ae7740b0c61026418
c135c44c2758e9a30c73224b4f2e15c5e45bc94d
F20101112_AACGVB stern_m_Page_01.tif
544267ddaef45fbb31870429bfceebe5
38ad0b8ca1e5adcfd5dbd7ec137e26a554ede7cb
2003 F20101112_AACGXZ stern_m_Page_29.txt
0ca3ae30856708bdb3c4419ddd56decf
82843a30df7029676651dceb737083b75251d11c
F20101112_AACGVC stern_m_Page_03.tif
f5e49224797465624f1ecca4c7261c80
9a67f8ff12882c23d9b14bb0720c8797fddefd3c
57677 F20101112_AACGQF stern_m_Page_14.pro
4aa503cc6f890b00ab26dc3a17a44b99
f8cf852419238d0784caa38537a3ed16b91770df
F20101112_AACGVD stern_m_Page_04.tif
50cb3374e3febd98589af50b22443700
67d2c1107ef3820f097ff832730a171e7b5f2060
F20101112_AACGQG stern_m_Page_32.tif
d354478db9dfb8bf7f353d849fd021de
442c510520070ecfae8aa0cc574ec6c97ee11888
F20101112_AACGVE stern_m_Page_06.tif
cb7b6927e0eddb187110df66eb5670f2
87f2c2283180a70487efdc7f38ef1f3ea55fe66a
63711 F20101112_AACGQH stern_m_Page_04.pro
a1e65db9304b66af0de5c0404dd86691
c25cce66fdb1227132bff33795c2ced96f3a9537
F20101112_AACGVF stern_m_Page_07.tif
82d8f1086fafa878f5fda128cdf822a5
21f96996638a9fb60651fdf39346f1ef90798b0a
26996 F20101112_AACGQI stern_m_Page_05.jpg
597b10971e375e32119213d1f64d9a9d
f80fa89821d6ef725b9909ec4f31cf56dbcfedc1
F20101112_AACGVG stern_m_Page_08.tif
233e1a9dc077b403b1ced79fda103bbc
3d5b62495c939499692baa5767535085e2ff4bc1
F20101112_AACGQJ stern_m_Page_09.tif
6876260d4c132aa1c709eee361846500
4a3ac673f0cec9823fec98e3f98f95a6efb2e1c9
F20101112_AACGVH stern_m_Page_10.tif
431fb3c9a5121df11c549647c6d98dad
0f456cdbd2084afa059f6af95fe6b92be9d6a3d5
4768 F20101112_AACGQK stern_m_Page_38thm.jpg
4c1ef5f49b5f19ed8d8ed4ed830bba31
26c08e035e017944a696db404e86e77cc013acb2
F20101112_AACGVI stern_m_Page_11.tif
190d9ce54cc7d208fcfbe9a7ff88f8a6
fe36151b32fae36ede5133d94b0b8e575e05575b
92 F20101112_AACGQL stern_m_Page_02.txt
523bb78ad4c732d1d7f5adb0026aff23
a738845271ca849a6e188e6a84f471567cc6aa1f
F20101112_AACGVJ stern_m_Page_12.tif
a5c0ecc57882eefe7d444d3512088685
5c7ec80bc6172392f03ee4b92097ef88db3876a3
55402 F20101112_AACGQM stern_m_Page_34.pro
52e64c5337185268703212737fbd81d5
b90e9ac5ea06a634063c696aa6ea866ee078365b
F20101112_AACGVK stern_m_Page_13.tif
94c6d1d51d0f8adf84392edf7541614f
3f2ee3796abe4d5b7a9bea83b92a0f4dd94f17dc
6991 F20101112_AACGQN stern_m_Page_15thm.jpg
ce9b83dc0834f73664711f85a237535b
2dfac2dd4e46f17bc727ca645dc18f350e3a32f1
F20101112_AACGVL stern_m_Page_14.tif
c39e8a057fe953dadee573ead075d8f0
8c3db739bc97e0aa65da799b044cd27461377950
2164 F20101112_AACGQO stern_m_Page_30.txt
dc10372f1f0dc294b46df6a5d6c7464f
49c6bdac2fb6cba900099dfecf8ce60b10392a35
F20101112_AACGVM stern_m_Page_16.tif
23b1d9dd24a3a1ab5f3a5722c6f82e3a
6cae4ab2b91f0d807c4cc0809da35f8c7332c3e9
2026 F20101112_AACGQP stern_m_Page_22.txt
be6f403b8853ad2d17fc1d1724a78905
7e52fe520b84ebb5160b75f0a2be13c69c6ff4da
F20101112_AACGVN stern_m_Page_17.tif
c576513dd316d28f2674a6945fd5f30d
8596ca67986bcdd1026d80b9082e957a587a7780
4857 F20101112_AACGQQ stern_m_Page_37thm.jpg
b5f679acf3bdc1656d1f513c4aca7c03
c076dec5f96ceb038cfc4cc402a4dfd30e3b078a
F20101112_AACGVO stern_m_Page_18.tif
94bc1566e4b62c512afa2f7e4faef841
cee0c8610a30a3d20679adefa26f9a6718732ccd
F20101112_AACGVP stern_m_Page_19.tif
dd19e90ac89913900db43271fc9c943d
c948ec33027f7a6f765edbdab30a9b4308627c8e
18754 F20101112_AACGQR stern_m_Page_24.jpg
5fadb6e0148323f856d05a52129ec39b
34c8ea4c8241aba54d6c1a636b20e662c152a5e2
F20101112_AACGVQ stern_m_Page_20.tif
622e99ca1943456112b5130752a48cde
6831684129f5003189c52f488669c96784b2539a
24978 F20101112_AACGQS stern_m_Page_13.QC.jpg
d1371b44c8cd7295dee88348a5d146e7
b62c520f02361437e26be343fdf2e2fdd8505a6f
F20101112_AACGVR stern_m_Page_21.tif
373a6886e21e7e7098da5411d7e8ada2
8030b43ec9b248a1f70f941037c48d98553dc053
74811 F20101112_AACGQT stern_m_Page_17.jpg
3af71b611682833d8d277c8a4386a54e
85df6c826c14ac2d972dab964123744a8ff17f8f
F20101112_AACGVS stern_m_Page_22.tif
f4ecbabb8d3aef34da9fe243bc4ec647
a358c9f7b891e1e68dec5bebf4349013776dc7b2
55786 F20101112_AACGQU stern_m_Page_20.pro
82fa5283354513ce22faa2e03740b086
332ec7a94c617520fb212992530c918cfad00729
F20101112_AACGVT stern_m_Page_23.tif
0f786e56207b1ed3a3d9f0b9742f9bfb
f5f0dc0bfbcc3f3b39ad6ac2788b1b6c0dc2ec9d
2187 F20101112_AACGQV stern_m_Page_20.txt
bea3535d1d78adcbf32be76f36fb0953
6dccb56cf2d0682f1ca803bd889cbcfc95d8b4f5
F20101112_AACGVU stern_m_Page_24.tif
088fa0ac7f6767ce4b5aa51d824a14c4
38939dfc2f176870a676c11a4d7d6511b85395ec
79019 F20101112_AACGQW stern_m_Page_32.jpg
2ae75efa5cefb1a6b4e83fd98f06ddfa
2f740eb6f47f1b2c47b24a2fc17efae52c6ee067
F20101112_AACGVV stern_m_Page_25.tif
02a2b6b47b3e0759b5bfe8c43e4e0270
55a1d1c526d6b686b8b84b98a6fd51c41699d204
24814 F20101112_AACGQX stern_m_Page_17.QC.jpg
e8d0963895f5cb67634bd8623cfe64dd
557455fa93d23630ed0354eb9300880c8918a3ad
3627 F20101112_AACGQY stern_m_Page_16.pro
4d3daab5560d3563edaa8281b3fa2a63
fe6d2aa6197a52d525a884691e34b5fee4ca338b
F20101112_AACGVW stern_m_Page_26.tif
7fce0ef1339aa604c1a60030ba1021f2
38ace4ed0a530b2064c807cf5b83270ca1d5e978
F20101112_AACGQZ stern_m_Page_33.txt
9313bb08e13ad7fc25c4dbdc1f38e841
9617b342f5a8bbd89e58c3e3c2070d40731db744
F20101112_AACGVX stern_m_Page_27.tif
30325aa023804098219615c9b0dcb6c5
276c51550588009237562e9bd32904923ad23bc0
F20101112_AACGVY stern_m_Page_29.tif
6dd9ccac28efe1fcae485bdf07859052
840c09f9fae60a32a1faa45def943e264a5f3430
79712 F20101112_AACGTA stern_m_Page_10.jpg
8978340a321b2cd7068780a25ab2b339
ea7d40d9b8f950043a146215ecc1d3002c67c770
F20101112_AACGVZ stern_m_Page_30.tif
b50bddb3f9551a4f0b464c8672501e02
ac72c51d01ec2450e10edad5feb6acd2bb1aa94f
76181 F20101112_AACGTB stern_m_Page_11.jpg
04ca51d4b07d917f84b5fd309037394b
f92429f4a793a8a6e423bae2bd6309c7e1e6675e
75560 F20101112_AACGTC stern_m_Page_12.jpg
b9e0c94edd671f2572c41b05637688e4
ba95e962112c31115dc0c6faec0e2edb79246f23
2107 F20101112_AACGYA stern_m_Page_31.txt
dc7ec88002e480e5030bc1e269865658
7c5c26e424ca34acc13af96378b256d53df07cd3
6774 F20101112_AACHAA stern_m_Page_19thm.jpg
e642ca049082c2e42c99b33ce3400d1a
661e2e41797d64a78d1bdf4311589ae4da19a66a
75960 F20101112_AACGTD stern_m_Page_13.jpg
970f5ef05ad6dc1856e7dbe5ded365b7
4b5c216040c4c4b38ff0badcbe85b4ff2c67eef6
F20101112_AACGYB stern_m_Page_32.txt
dd902ec646aca9d61c3dab1fcb5f5b98
c5810ff24309dac84eaeaee87a5fd2a467ee349d
26191 F20101112_AACHAB stern_m_Page_20.QC.jpg
7fd253938b907b3e5f7b4d9318baec74
cf19249c553696a2c814aa4db35e85b4acf5efa5
81501 F20101112_AACGTE stern_m_Page_14.jpg
f1a022d0924c4d369330957c45dc0e17
aee9b67cd3de04048c812d08af27a2636809f603
2180 F20101112_AACGYC stern_m_Page_34.txt
c1aa47825218026b4c2ccf836106fb77
a1f04ea7a405f77cc76a488a478cde9bbfbe638f
4189 F20101112_AACHAC stern_m_Page_21thm.jpg
d74c3919ebe0a2a4bdad1b9115dacef5
1349768b7e28b8fff4d3917d713fdcd4ff6132fa
77202 F20101112_AACGTF stern_m_Page_15.jpg
043f9afd135f6822cfc37d986b7b0099
0e9f64a41109c560935a71a3b4854967fc58113d
482 F20101112_AACGYD stern_m_Page_35.txt
118e0557bd9c24b4f93119f4ff363f4e
3a2fd11ee41b655d47cdf8f9f68d92bc428f5995
23068 F20101112_AACHAD stern_m_Page_22.QC.jpg
b5e94caae3af1173f045baa76d204724
9431cc3e9a97bed63248ade5eb68b0bd16f8510c
12977 F20101112_AACGTG stern_m_Page_16.jpg
82a2df1b95795780313aae3960735e38
f87f12f841f3138a40c572aac8669b0649e3beec
2568 F20101112_AACGYE stern_m_Page_36.txt
0591b952b7d6d9a7690c3a144797c386
b2d6e25c9d7b04b20717cc88d97d1add1227d609
18169 F20101112_AACHAE stern_m_Page_23.QC.jpg
357be434685e725310b6437a21847f44
1020c67b897c0141524b5cb428e1f3b9c51164a2
71028 F20101112_AACGTH stern_m_Page_18.jpg
f664ec8858a5ba453a5687f8b5515a93
8b7d7e2665d10b63c895dc6f8c01f2971d5dae13
1301 F20101112_AACGYF stern_m_Page_38.txt
e4cd60321512fd99d76b07fbb6e96033
1dd5ac1b3b3d3bbfdb382614812f0101386af91d
18567 F20101112_AACHAF stern_m_Page_25.QC.jpg
5e67de4dbcb770d1acdea67ba2e90438
2d8af7b5d886ee52d2b58dc9887a18473d2ba7e5
72804 F20101112_AACGTI stern_m_Page_19.jpg
362f8b6d282b9cd9e26985b173738749
41759ddfa8a4e1c446943c9ee91776af6f55b4bf
91855 F20101112_AACGYG stern_m.pdf
833eb21efcdee3b1f3dc08a76b24153e
a75bec8e6d5aa2af4eae2cb4c477608ee91a9f85
5217 F20101112_AACHAG stern_m_Page_25thm.jpg
3b460cedbe00d2b55ea356778c414428
1f27bf6717e0f5079a63a1e83967216adbc7713a
45217 F20101112_AACGTJ stern_m_Page_21.jpg
6526b682886f05cf3119ffe3579a4a8c
b8473ef0b8111de928ddddede65f7071cc5802dd
6441 F20101112_AACGYH stern_m_Page_22thm.jpg
7bf84fafd3af9e48d8ec0c5672eba803
1fbd4a5039daaeef13193921ba552c96eef0a406
1680 F20101112_AACHAH stern_m_Page_26thm.jpg
466f757ba97e10583de4e2add922cf18
e66a5de19108a0741037fc3172eeeac04cd9c01b
70043 F20101112_AACGTK stern_m_Page_22.jpg
b00522ff399b59023cd62ac1c9c8216f
c48e13b0d25caeeeecca3ed54fe6122b59e31871
7046 F20101112_AACGYI stern_m_Page_20thm.jpg
9de1fb6f477c3e3d1331fc27a5f11f03
dc9029edee44ac49492c0762e4ed5a70330840fc
9456 F20101112_AACHAI stern_m_Page_28.QC.jpg
933180351c795ca1ebf80e6dd526233e
bffeb8dc75c55350e2d7217210c4b902e33d2e8f
55464 F20101112_AACGTL stern_m_Page_23.jpg
8f81b88e463a229c3ad59a27732ab4c0
88c15f544378464ab1e4b93fd654ce219f885af8
3202 F20101112_AACGYJ stern_m_Page_27thm.jpg
bf86576f80f694f0a348bbe2d0f53c4c
8f04cd41c52332095d0b6e5af6f67737f92acf59
6935 F20101112_AACHAJ stern_m_Page_32thm.jpg
33ead7f2dde6d87be0dbce2d0d812a5f
866c3b25683e97ca0a5d58dab44e2f53f61ce107
60459 F20101112_AACGTM stern_m_Page_25.jpg
bdb09ec9212cae1c956434ac35abd5e9
1b76d6a56177f1e0fdfb35084d3cba78824cc031
4508 F20101112_AACGYK stern_m_Page_26.QC.jpg
9f1efcf81c2bb860d90c2805cbe192de
4de6a0e3ed8e16c517151f0612c09fc3f04f2024
23993 F20101112_AACHAK stern_m_Page_33.QC.jpg
9841fd043f07c481ceaf0d367497b4a6
4cfda9bfd61fbcad0b96efb2729b43f5e013d24b
13942 F20101112_AACGTN stern_m_Page_26.jpg
4ef0876de264eb1fd961790c45263690
961488fe8eee5faee09db90f10475a5b4bfeb988
25251 F20101112_AACGYL stern_m_Page_15.QC.jpg
d6986629c43f3a1c026764a25919d843
3775e9807bd38846d5e9744be64568dd677f39ef
25719 F20101112_AACHAL stern_m_Page_34.QC.jpg
3c9d37acce4b41c22cbdf65d5b5bd482
d1ba12fc6006635ee5c1e68b25452d027eaaacfa
30718 F20101112_AACGTO stern_m_Page_27.jpg
4472a5beb14576fe0a14196601a43abf
92529845cd9f2fc9ec5d7d8b25ff195a50c1ed7e
2310 F20101112_AACGYM stern_m_Page_24thm.jpg
21fc648d7b2370c35a932a371e008d60
5cd87a143b7ee2c31b79bf22e419d3319972eb09
6741 F20101112_AACHAM stern_m_Page_34thm.jpg
97ff66475dfe2d2268145ed9717201d1
59d06fbacccaeefac3b3fc64496ebefcba3131d6
30056 F20101112_AACGTP stern_m_Page_28.jpg
52ddfd30471ee991376448081ccf82bc
d0d230403a112cd9360e174c0708efdd333e96b6
7529 F20101112_AACGYN stern_m_Page_35.QC.jpg
cd70c0a0da7ad9c19e07c69bd347cf44
76f1cc2f43b4fc64701bfb8b9d9ef17e83aa035c
2464 F20101112_AACHAN stern_m_Page_35thm.jpg
bab9aaac75c30f1fb29448bd8a5c7884
11a49328fbd5c498e5c56129147b71a02f6d4b0f
76774 F20101112_AACGTQ stern_m_Page_31.jpg
a5ebcd20bcc79943808df97413db6d80
f8316be7a201cd88b3489a9db693dcbade9340fa
2360 F20101112_AACGYO stern_m_Page_08thm.jpg
bbeef4ba36e7c1a3fb7dba373dc5fe84
d84545fa4f265ded4194ccda951f256c5e6eb155
7061 F20101112_AACHAO stern_m_Page_36thm.jpg
0fd2962fb508aa7ea85ae39b52aafdbc
e3aba9908a8b5e1c16bb6e1397f554605995d481
73417 F20101112_AACGTR stern_m_Page_33.jpg
f161c2655941fc528b64a98ea2c87b02
44dfc418a6c679900e8ed697c7bb8a210b831e92
6675 F20101112_AACGYP stern_m_Page_33thm.jpg
2b979ea445e7298226a667c25ee8731d
6c3f13119ed8523bed0fa4ba284bf6065c3f3fa6
16894 F20101112_AACHAP stern_m_Page_38.QC.jpg
fd31ea4bd10409ab30af08c44db45e33
b8ec9905bbd57fc352c63bc3c11d8a08eab9ebc4
78392 F20101112_AACGTS stern_m_Page_34.jpg
527fd919da408c582e55e3188b7938db
a0a874eee34b29dbf271bc5a8bf6d9898d94f88b
22468 F20101112_AACGYQ stern_m_Page_29.QC.jpg
26b4318658ee687a5a3729708ee32e5d
40e06d30216de7e305c3456f873edaa5f37fc276
23314 F20101112_AACGTT stern_m_Page_35.jpg
a16bf9fc08bf114629cafee8d936c5b7
c394392e1c359c556ae1d6817b7847ad12a61409
6940 F20101112_AACGYR stern_m_Page_10thm.jpg
c57bdcf7fa2c2aca078031de1d000aba
c5972a470c0047a56cda595f3f6446dd674ea975
6714 F20101112_AACGYS stern_m_Page_13thm.jpg
0fc2cfed6a6bf9df36b74b60c7ea1b4a
bc387002b7e9bff50739576c2f47c88de34545ff
93855 F20101112_AACGTU stern_m_Page_36.jpg
6340db2e09cff41830d93e0fa14d6003
8a87df6876d3cb93534fb80140e103319f53ea7a
25356 F20101112_AACGYT stern_m_Page_09.QC.jpg
fd3fa11346dd9e5f49373064680713af
f69d3cf4896984527d4a6a8ce45e51ba4dca6989
59495 F20101112_AACGTV stern_m_Page_37.jpg
e62f36f6acdedad2693363801214d95d
0af8a6201d86be6548a6a3e49e599d3e9bd2e1de
14720 F20101112_AACGYU stern_m_Page_21.QC.jpg
b080d24900a52c19b97ca3f5d7d822c9
d2df19ae3f6868aba4a5eae6f990167c99cd8356
51232 F20101112_AACGTW stern_m_Page_38.jpg
096a1c620e300ef5b0ed3c1858fcc52f
3048b639cf5c177f6c10e8ebff33e364536aae81
7308 F20101112_AACGYV stern_m_Page_03.QC.jpg
99dda2dd9bfc63c100973f315953209e
d2f55c9debc290b4e261084d89e58ab78b8991d7
25060 F20101112_AACGTX stern_m_Page_01.jp2
098e68ad4b2baee426277bd33f6ab634
f3d6e00884244a5383f7cfd23a4048828b348f48



PAGE 1

1 ASSESSING DIMENSIONS OF DISRUPTIVE CHILD BEHAVIOR WITH THE EYBERG CHILD BEHAVIOR INVENTORY By MELISSA K. STERN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Melissa K. Stern

PAGE 3

3 ACKNOWLEDGMENTS I thank my supervisory committee chair, Dr. James H. Johnson, for his invaluable guidance and mentoring. I would also like to thank th e members of my committee for their suggestions and comments. I thank the stud ents and research assistants of the ADHD study lab and members of the child area in the Clini cal and Health Psychology Departme nt for their help and support.

PAGE 4

4 TABLE OF CONTENTS Page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........5 FIGURE......................................................................................................................... ..................6 ABSTRACT....................................................................................................................... ..............7 CHAPTER 1 INTRODUCTION................................................................................................................... .9 ADHD and the Disruptive Behavior Disorders........................................................................9 The Eyberg Child Behavior Inventory (ECBI).......................................................................10 2 METHODS........................................................................................................................ .....17 Data Collection................................................................................................................ .......17 Measures....................................................................................................................... ..........18 The ECBI....................................................................................................................... ..18 The CPRS and BASC......................................................................................................18 Participants................................................................................................................... ..........19 Statistical Analyses........................................................................................................... ......19 Factor Analyses...............................................................................................................19 Validity Analyses............................................................................................................21 3 RESULTS........................................................................................................................ .......22 Factor Analytic Findings....................................................................................................... .22 Validity Analyses.............................................................................................................. ......23 4 SUMMARY AND CONCLUSIONS.....................................................................................29 LIST OF REFERENCES............................................................................................................. ..36 BIOGRAPHICAL SKETCH.........................................................................................................38

PAGE 5

5 LIST OF TABLES Table Page 3-1 Factor loadings for principa l axis factoring via oblique ro tation three-factor analysis of the ECBI.................................................................................................................... ....25 3-2 ECBI factor correlations................................................................................................... .26 3-3 Items not loading at .4 or higher in the thre e-factor oblique rotation factor analysis of the ECBI....................................................................................................................... ......27 3-4 Pearson correlations between ECBI factors and CPRS and BASC subscales...................28

PAGE 6

6 FIGURE Figure Page 3-1 Scree plot of the ECBI in dividual frequency item data.....................................................24

PAGE 7

7 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science ASSESSING DIMENSIONS OF DISRUPTIVE CHILD BEHAVIOR WITH THE EYBERG CHILD BEHAVIOR INVENTORY By Melissa K. Stern May 2007 Chair: James H. Johnson Major: Psychology The Eyberg Child Behavior Inventory (ECBI) is a well-validated parent rating scale that assesses behavior problems in both children and adol escents. While originally conceptualized as a one-dimensional measure of disruptive behavior there are discrepancies in the literature regarding its factor structure, w ith some studies finding a single factor and others finding support for a multidimensional structure. While previous research samples have been diverse, additional research is needed on other clinical samples. The primary goal of our study was to further assess the factor structure of this measure using data ob tained from a clinical sample displaying a wide range of disruptive behaviors such as childre n referred for evaluation of Attention Deficit Hyperactivity Disorder (ADHD) and associated conditions. Data from parents of 181 children referre d to an ADHD Program were used for the analyses. Upon initial assessment in the ADHD clinic, parents completed a number of assessment measures that included the ECBI as well as measures used in validity analyses. A factor analysis using principal axis factor ing and promax rotation was performed on the individual item data of the ECBI. Analysis of eigenvalues greater than one, item distribution breakdown, and inspection of the scree plot reveal ed that the ECBI was best accounted for by a three-factor structure with factors representing oppositional defiant behavior, attention problems,

PAGE 8

8 and conduct problems. Convergent and discrimina nt validity analyses using measures of relevant constructs from psychometrically s ound measures demonstrated good validity for all three factors. Current research continues to demonstrate that th e ECBI is a valid tool with a multidimensional structure, which could increase the measures utility in both clinical and research contexts.

PAGE 9

9 CHAPTER 1 INTRODUCTION ADHD and the Disruptive Behavior Disorders Attention Deficit Hyperactivity Disorder ( ADHD) is one of the most commonly diagnosed childhood behavioral disorders o ccurring in approximately 3 to 5% of the general child population (American Psychiatric Association (APA), 1994). Barkley (1998) suggests that children with this disorder also comprise approxi mately 40% of referrals to outpatient clinics. The guidelines defined in the Diagnostic and Sta tistical Manual of Mental Disorders, Fourth Edition (DSM-IV) are generally used to make di agnostic decisions regard ing the presence or absence of ADHD. Current criteri a include clinically significant levels of symptoms in the domains of attention, hyperactiv ity and impulsivity; symptom ons et before 7 years of age; symptoms that are present across situations; that cause impairment in social, academic, or occupational functioning; and symptoms that have occurred apart from other mental disorders. Subtypes of the disorder are based on thes e dimensions and include the predominantly inattentive type, the predomin antly hyperactive/impulsive type, and the combined type. ADHD also has high rates of comorbidity w ith Oppositional Defiant Disorder (ODD) and Conduct Disorder (CD). ODD is defined as a pa ttern of hostile, defiant, and negativistic behavior towards adults (APA, 1994). Symptoms include argui ng with adults, actively defying rules or requests, as well as of ten being angry and vindictive. To meet diagnostic criteria for ODD, a child must have exhibited the pattern of defiant behavior for at least 6 months, must show evidence of impairment, and the symptoms mu st not be due to another mental disorder. Conduct Disorder is characterized by a pattern of be havior that violates th e basic rights of others or age-appropriate societal rules (APA). Symp toms are grouped into the categories of aggression towards people or animals, destruction of propert y, deceitfulness or theft, and serious violations

PAGE 10

10 of rules such as running away. Subtypes for conduct Disorder are based on age of onset and include childhood-onset type and ad olescent-onset type. All three of the described disorders are viewed as disruptive behavior or externalizing disorders and are highly comorbid with one another (APA). Indeed, the high degree of co morbidity between ODD and CD has prompted some researchers to suggest that they are not, in fact, separate disorders, although a variety of studies clearly support distinc tions between these two conditi ons (Loeber, Lahey, & Thomas, 1991). While several measures exist that assess for symptoms of each of these disorders (e.g. structured interviews, Child Be havior Checklist, Diagnostic Inte rview Schedule for Children), no widely used brief questionnaire or screener assesses all of thes e domains. A short measure that accomplishes such a task has the potential fo r great clinical and research utility. The Eyberg Child Behavior Inventory (ECBI) The Eyberg Child Behavior Inventory (ECBI) is a parent rating scale which assesses disruptive behavior problems in both children and adolescents (Eyberg & Robinson, 1983). The ECBI was designed to differentiate normal ch ildren and adolescents from those with conduct problems and as a follow-up instrument for asse ssing treatment effects for a broad age range (Eyberg & Ross, 1978). While demonstrating g ood psychometric characteristics in previous research (Boggs, Eyberg, & Reynolds, 1990; Funderburk, Eyberg, Rich & Behar, 2003; Robinson, Eyberg, & Ross, 1980), there are discrepa ncies regarding the factor structure of the ECBI. Eyberg designed the ECBI as a one-dim ensional assessment scale of general conduct problems and studies have found support for such a structure (Colvin, Eyberg, & Adams, 1999; Eyberg, 1992). Other researchers, however, have found evidence of a multidimensional structure of the ECBI. In an examination of the crosscultural validity of the ECBI in a sample of Australian preschoolers, Werba (2003) found evid ence for a two-factor st ructure of the ECBI.

PAGE 11

11 The author found similar results when repeating th e exploratory factor an alysis on a sample of parents of preschoolers in the United States. Re sults from both analys es, while not identical, indicated the factors represented the constructs of ODD and ADHD. Taking a closer look at individual items from th e ECBI, they seem representative of three specific categories and similar to specific DS M-IV (APA, 1994) criteria for ODD, ADHD, and CD. For example, the ECBI item Has short atte ntion span is similar to the DSM-IV criterion Often has difficulty sustaining attention in task s or play activities. Similarly, the ECBI item Acts defiant when told to do something is similar to the DSM-IV ODD criterion Often actively defies or refuses to comply with adult s requests or rules. A nother ECBI item, Teases or provokes other children, appear s at face-value to be simila r to the DSM-IV CD criterion Often bullies, threatens, or intimidates others. When continuing to compare the ECBI item by item to DSM-IV criteria for the three disorders, many other examples ar e revealed. As ODD, CD, and ADHD comprise the primary childhood di sruptive behavior disorders, it is not surprising that a measure such as the ECBI may have a factor structure representing aspects of these three areas. Some researchers have, in fact, found evidence for a three-factor stru cture of the ECBI. Burns and Patterson (1991) collected data from pa rents of 1,526 children in outpatient clinics in Montana, Washington, Idaho, and Oregon. In their exploratory factor anal ysis they utilized principal components analysis with varimax ro tation on the frequency scores of the ECBI. Results indicated a three-factor structure with factors represen ting oppositional-defiant behavior, attentional difficulties, and violatio ns of others rights via aggression. Burns and Patterson (1991) attempted to replic ate their three-factor findings of the ECBI using a random sample of children from an urba n school district in Wa shington. Again using

PAGE 12

12 principal components analysis and varimax rotation, Burns and Patterson performed an exploratory factor analysis, though limiting the resulting model to three factors as they had an a priori hypothesis based on results from their first study. Findings supp orted a three-factor solution representing the categories they had iden tified in previous work. Burns and Patterson (2000) later combined their urba n and original sample and divi ded them into two subsequent random samples in order to further evaluate the f actor structure of the E CBI. Using exploratory factor analysis with maximum likelihood extraction and oblique rotation, they examined twoto seven-factor solutions as results indicated seven eigenvalues greater than one. Here, Burns and Patterson found a four-factor solution to be mo st relevant. The first three factors again represented symptomatology of OD D, attention problems, and CD, but the fourth factor did not represent a discernible dimension. The researcher s explained this fourth factor as a possible response bias effect which resulted in a meani ngless factor as can happen when items with low but significant loadings on the firs t factor separate and form their own separate factor. Burns and Patterson chose the four-factor solution because it resulted in the three primary factors becoming more conceptually clear. Burns and Patterson (2000) next took thei r second random sample and performed a confirmatory factor analysis on the ECBI using the first three factors of the four-factor model (the fourth factor was not used as it was judge d conceptually meaningles s). Burns and Patterson also compared the three-factor solution to a one-f actor solution and a two -factor solution (which combined the ODD and CD factors). Results from the initial three-factor confirmatory analysis indicated adequate to go od fit, and a significantly better fit compared to the oneand two-factor models. Burns and Patterson also identified 90th percentile cutoff scores from their community

PAGE 13

13 sample to identify children with significant levels of oppositional behavior, attention difficulties, and conduct problems. Still, other researchers con tinue to find support for a one-f actor structure of the ECBI (Colvin et al., 1999). Using data collected for the restandardization of the ECBI, researchers found, via four different methods of factor analysis, that the ECBI was best accounted for using a one-factor model. These research ers utilized principal component s analysis with both varimax and oblimin rotations, and principal axis fact oring with varimax and oblimin rotations. A conceptually meaningful one-f actor model was found with all four methods of analysis. In 2005, Weis, Lovejoy, & Lundahl performed a confirmatory factor analysis using individual ECBI item data to compare the pr eviously described Bu rns and Patterson (2000) three-factor model to a oneand two-factor m odel using a large general clinical sample of young children. In this analysis, the researchers used da ta from different studies of stress, affect, and parenting which included ECBIs completed by a pr edominantly Caucasian sample of parents of children ages two to six. The study demonstrated that Burns and Patterson s three-factor model was an adequate fit to the combined data set. A one-factor model demonstr ated a poor fit to the data. The two-factor model (collapsing the ODD and CD categories) did demonstrate adequate fit. It can be noted, however, th at the three-factor model was a significantly better fit to the data than either the oneor two-factor model. In addition to their confirmatory factor anal ysis, Weis et al., (2005) completed a second study assessing the discriminative validity of the three-factor model of the ECBI. In this investigation, the researchers r ecruited caregivers of 115 children referred to an outpatient psychology clinic. In order to as sess the discriminative validity of the Burns and Patterson cutoff scores described earlier, Weis et al. had caregivers complete an ECBI. The primary aim of this

PAGE 14

14 study was to assess the ability of cutoff scores to categorize children with and without significant disruptive behavior problems. The second aim of the study was to discrimi nate between children who do or do not display symptoms of ADHD, ODD and CD as determined by the Disruptive Behavior Disorder Rating Scale (DBDRS; Barkle y, 1997). Results demonstrated that the three factors had adequate negative predictive power and poor positive predictive power. Sensitivity for the three factors was poor (.63 to .77) but results demonstrated good specificity (.91 to .94). Results of a MANOVA using the factor scores of the ECBI as the multivariate dependent variable and the symptom groups from the DBDRS as the independent variables suggested that the Attention Difficulties factor did not appear to discriminate children with attention problems from children with ODD or CD; the same wa s true of the oppositional factor. The Conduct Problems factor did appear to adequately differen tiate children with aggressive behaviors from those with attention problems or ODD. While these researchers did extend the work of Burns and Patterson, the limited age range and nonspecifi c general clinical sample may limit results. The second sample used in the study was also not large and individual diagnostic groups were not well represented. Thus, while the study adds to the literature, furthe r research is needed. In these previous studies, samples have in cluded participants recr uited through pediatric clinics and schools, general outpatient psycholog y clinics, childcare centers, family service agencies, and newspaper advertisements. While re searchers have previously attempted to assess the fit of a three-factor model on a general clinical sample (Wei s et al., 2005), additional research is needed on other clinical samp les. In addition, it is important to assess the validity of factors resulting from such research a nd, perhaps later, the value of cutoff scores for well validated factors. Research along these lines might provide the basis fo r using the ECBI as a brief, preliminary screener for specific problem areas, rath er than just as an ov erall index of disruptive

PAGE 15

15 behavior. Additionally, it can be noted that th e ECBI was also designed to assess change in conduct problem behavior over the course of treat ment (Eyberg & Ross, 1978). Here, the utility of the ECBI could potentially be increased thr eefold in terms of providing change data in multiple areas that may or may not be direc tly targeted by treatment (e.g., ODD, CD, and ADHD related symptoms). This, in turn, might d ecrease time spent in the assessment process by reducing the need for multiple pre-screeners and follow-up assessments. Clinicians, researchers, and patients will save time if such a measure is available. While the ECBI has already been validated as a brief screening instrument for both children and adolescents with conduct problems, it may be able to provide even more information. The current study aimed to add to the literature by further asse ssing the factor structure of the ECBI and conducting validity analyses on the factors of th e model. While previous investigations have studied dive rse samples, more research is needed on clinically relevant samples. One such sample would be one that could be expected to display symptomatology for all three of the disorders purporte d to be assessed by the ECBI. As ADHD is highly comorbid with both ODD and CD, a sample of children re ferred to an ADHD clinic would likely display the symptomatology of all three disorders. Specif ically, it was expected that multiple factors of the ECBI would be identified usi ng such a sample and that these factors would be representative of symptomatology of attention problems, oppos itional behavior, and conduct disorder. As previous factor analyses found th at several items did not load on any of the factors, it was hoped that this relevant clinical sample would be able to utilize all of the ECBI items, some of which may not have appeared as salient in the more ge neral samples in previous research. Secondarily, we expected that factors f ound via factor analysis would demonstrate good convergent and

PAGE 16

16 discriminant validity when assessed in rela tion to psychometrically established measures designed to measure similar constructs.

PAGE 17

17 CHAPTER 2 METHODS Data Collection Data for this study was collected from children (and their parents) who were involved with the interdisciplinary ADHD program at the University of Florida. This program is staffed by professionals from various disciplines (e.g. psyc hology, pediatrics, child psychiatry, medicine, nursing, speech pathology) from several departme nts at the University of Florida and Shands Teaching Hospital. Children are referred to the program for assessment and follow-up by parents, teachers and physicians from the loca l community and from throughout North Central Florida. Some children referred to the program have not received a complete assessment and have no official diagnosis at the time of referral. Others with a diagnos is have not undergone a detailed assessment such as that available at the University of Florida. When a child is referred to the program, the child and caregiver(s) particip ate in an all-day assessment. This includes parentand teacher-report (and sometimes child-re port depending on age) measures to assess for ADHD symptoms and possible comorbid features a clinical interview, intellectual and achievement testing, as well as other measures, as necessary. The data from the psychological assessment are presented at an ADHD Program meeting, along with data obtained by the pediatrician associated with the ADHD program. This information is used by the ADHD interdisciplinary team to determine whether additional assessment by other professionals is needed and to determine recommenda tions and the appropriate diagnosis. An application for exempt status was submitted to the University of Florida Institutional Review Board as data was collected without iden tifiers. Once exempt status was obtained, data were collected from the files and entered into a database for analyses. Demographic data collected included age, gender, ethnicity, edu cation level, and dia gnoses of the child.

PAGE 18

18 Information on the age, gender, education leve l, and family income of the caregiver who completed the forms was also recorded. Individual item data from the ECBI as well as scores from several other measures used for validity an alyses were also entered into the database. Measures The ECBI The ECBI is a 36-item parent-report measur e designed to assess conduct problems in both children and adolescents (Eyberg & Ross, 1978). The ECBI has two scales: a problem scale and a frequency scale. Caregivers ar e asked to indicate how often a particular behavior or problem occurs on a 7-point likert-type sc ale. The scale ranges from anc hors never to always. The problem scale asks caregivers to indicate in a yes or no format whether they consider the particular behavior to be a pr oblem regardless of the frequency. The ECBI has demonstrated good internal consistency, test-retest reliability, and sensitivity to treatment effects (Colvin et al., 1999; Eyberg & Robinson, 1983; Robinson et al., 1980). The CPRS and BASC As part of the childs assessment, caregiver s complete the Conners Parent Rating Scale (CPRS; Conners, 1997) and the Behavioral A ssessment System for Children (BASC; Reynolds & Kamphaus, 2004). The CPRS and BASC ar e broadband measures that have undergone psychometric validation in previous research. The CPRS is an 80-item questionnaire that assesses ADHD symptoms and associated features and provides measures used to assess the validity of factors found in the pr esent factor analysis. The BASC is a slightly longer broadband measure, completed by a parent or caregiver, wh ich assesses symptoms related to various types of child psychopathology and child difficulties that were also useful in assessing validity of derived factors in the present study. Caregivers may have completed either the first or second

PAGE 19

19 edition of the BASC as the measure underwent revision and was republishe d in 2004. However, both versions contain the subscales used in analyses for the current study. Participants Data were collected from 181 caregivers of children referred to the ADHD clinic at the University of Florida. The average age of th e child referred to the clinic was 8.75 years old ( SD =2.93) and 75% ( N =135) were male. Consistent w ith Alachua County census data, 75% ( N =136) of the sample identified as Caucasian, 18% ( N =32) identified as African-American, 2% ( N =3) identified as Hispanic, and the remaining 5% ( N =10) of the sample identified as another ethnicity, multiracial, or did not specify. The av erage level of education of the child was the third grade. Of the children on whom ECBIs were completed, 92% ( N =166) were given a diagnosis of ADHD. Of thos e diagnosed with ADHD, 79% ( N =131) were combined type, 17% ( N =28) were predominantly inattentive type, 2% ( N =4) were predominantly hyperactive/impulsive type, and 2% ( N =3) were categorized as ADHD NOS. A total of 22% ( N =40) of the children met diagno stic criteria for ODD and 3% ( N =5) met diagnostic criteria for CD. Statistical Analyses Factor Analyses In order to accomplish the aim of further exploring the factor structure of the ECBI, an exploratory factor analysis (EFA) was conducted on the individual item data of the ECBI frequency scale. EFA, as opposed to confirmatory factor analysis (CFA), was chosen for several reasons. First, while several studies have f ound multidimensional structures, these studies are few in number and were conducted on limited samples. While Burns and Patterson (1991, 2000) used large samples in their analyses, these participants represented community samples rather than samples drawn from a clinical population. The same is true of the study with Australian

PAGE 20

20 preschoolers (Werba, 2003). Even though Weis et al. (2005) performed a CF A, their sample was also not a specific clinical sample; perhaps a diffe rent model would have fit the data better had a more relevant sample been used. Here it can al so be noted that Weis et al. performed their CFA on the three-factor model from Burns and Patterson which excluded nine of the ECBI items. It was hoped that use of a relevant c linical sample would u tilize all 36 of the ECBI items. Hence, it was reasonable to conduct an EFA on the current c linical sample in order to explore the factor structure that would be most r obust in a sample likely to disp lay symptomatology of all three disorders. Secondly, while multivariate normality is generally helpful in providing a clear factor structure, the assumption is much more per tinent to CFA than EFA (Floyd & Widaman, 1995). As the current study sample is not a large sample multivariate normality is harder to achieve and the techniques employed in an EFA are not as sensitive to non-normally distributed data. Thus, the validity of the resulting f actor structure is somewhat incr eased by the use of EFA as opposed to CFA. A minimum sample size of 180 was chos en based on the genera lly accepted rule of needing five cases per item in order to conduct a valid analysis (Floyd & Widaman). Specifically, the study extracted factors via principal axis f actoring, a technique which is not thought to be as constrained by assumpti ons of multivariate normality (Floyd & Widaman, 1995). Two kinds of rotation were employed in the current study. First, all models were rotated via oblique rotation. As ADHD is highly como rbid with both ODD and CD it is logical to expect that factors representing symptoms of the three disorders woul d be correlated. All analyses were rerun using ort hogonal rotation to see if use of the independ ent rotation strategy resulted in a clearer factor stru cture. An initial EFA was run without specifying the number of factors. After investig ation of the scree plot and eigenvalues subsequent EFAs were run forcing specific-numbered solutions.

PAGE 21

21 Validity Analyses Pearson correlations were used to assess conve rgent and discriminant validity of factors found in factor analyses. Here, the ECBI fact ors were correlated with scores on various subscales of the CPRS and BASC. Specifica lly, the CPRS contains an ADHD Index and a DSM-IV Inattentive subscale wh ich were expected to correl ate highly with the proposed Attention Difficulties factor of the ECBI. The CPRS also contains an Oppositional Behavior subscale which was expected be highly correlated with the Oppositional Defiant Behavior factor of the ECBI. The BASC has an Attention Prob lems subscale which was expected to be highly correlated with the Attention Diffi culties factor of the ECBI. The BASC also has both a Conduct Problems and an Aggression subscale, which were hypothesized to be highly correlated with the Conduct Problems factor of the ECBI. The BA SC and CPRS both contain some form of an anxiety subscale and these were hyp othesized to not be correlated with any of the three proposed factors of the ECBI.

PAGE 22

22 CHAPTER 3 RESULTS Factor Analytic Findings Results of the initial factor analyses revealed nine factors with ei genvalues greater than one. Inspection of the scree plot (See Figure 31) demonstrated two point s of leveling-off, one point at three factors and one point at six factors. Thus, four more factor analyses were run forcing three-, four-, five-, and six-factor solu tions. This was completed using oblique rotation initially and then again with orthogonal rotation. Conceptually, results were best accounted for by the three-factor oblique ro tation with factors representi ng oppositional defiant behavior, attention problems, and conduct problems. Result s of this model are presented in Table 3-1. Items were considered to load on a factor if they had a loading of .4 or higher, a generally accepted level (Floyd & Widaman, 1995). This threefactor solution involved a total of 25 of the 36 ECBI items. Here it can be seen that f actor 1, labeled Oppositional Defi ant Behavior, consisted of 12 items which seemed to reflect the pattern of de fiant and negativistic behavior towards adults associated with ODD. The second factor, labeled Attention Difficulties, consisted of seven items that, at face value, appeared representative of the attention proble ms often experienced by children diagnosed with ADHD. Lastly, the third factor, labeled Conduct Problems, was comprised of six items which appeared to repr esent aggression and viol ation of the rights of others and societal norms, a symptom pattern that is associated w ith Conduct Disorder. Correlations among factors are presented in Ta ble 3-2. The EFA resulted in 11 items not loading on a factor at .4 or hi gher (See Table 3-3). Only two items cross-loaded on another factor at a factor loading of .2 or higher. These items were Whines and Interrupts.

PAGE 23

23 Factor scores were computed as the sum of th e frequency ratings for the items loading at .4 or higher on the three factors. Summation was us ed as it is considered an adequate method of computing factor scores (Floyd & Widaman, 199 5; Tabachnick & Fidell, 2007) and it is consistent with the computation of factor scor es employed in prior st udies (Burns & Patterson, 2000; Weis et al., 2005). The 11 items that did not show significant loadi ngs on any factor were not included in the computation of factor scor es. In this sample, the Oppositional Defiant Behavior factor had a mean score of 53.96 ( SD =12.42), the Attention Difficulties factor had a mean score of 36. 98 ( SD =8.19), and the Conduct Problems factor had a mean score of 17.5 ( SD =7.95). Cronbachs alpha for the Oppositional De fiant Behavior, Attention Difficulties, and Conduct Problems factors were .92, .85, and .80 re spectively, suggesting adequate to very good internal consistency of the factors (Henson, 2001). Validity Analyses Factor scores were correlated with various subscales of the BASC and CPRS (See Table 3-4). Examination of the correlation matrix, as predicted, revealed that the ECBI factors correlated most highly with thei r conceptual counterpart subsca les of the BASC and CPRS. None of the ECBI factors corre lated significantly with the anxi ety subscales of the BASC and CPRS

PAGE 24

24 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Factor Number 12 10 8 6 4 2 0 Eigenvalue Figure 3-1. Scree plot of the ECBI individual frequency item data.

PAGE 25

25 Table 3-1. Factor loadings for pr incipal axis factoring via oblique rotation three-f actor analysis of the ECBI Factor loading Items 1 2 3 Factor 1 (Oppositional Defiant Behavior) 13. Has tempter tantrums .915 -.125 -.137 10. Acts defiant when told to do something .913 -.106 -.061 12. Gets angry when doesnt get own way .889 -.048 -.066 14. Sasses adults .776 -.067 .093 9. Refuses to obey until threatened with punishment .770 .084 -.027 11. Argues with parents about rules .735 -.032 .040 8. Does not obey house rules on own .724 .069 .013 17. Yells or screams .668 .021 .113 15. Whines .539 .219 -.098 5. Refuses to do chores when asked .529 .067 -.004 7. Refuses to go to bed on time .465 .189 -.091 4. Refuses to eat food presented .407 .065 .009 Factor 2 (Attention Difficulties) 30. Is easily distracted -.258 .867 .029 32. Fails to finish tasks or projects -.060 .850 -.113 31. Has short attention span .006 .794 -.130 34. Has difficulty concentrating on one thing .007 .700 -.046 29. Interrupts .263 .524 .055 1. Dawdles or lingers at mealtime -.031 .466 .192 2. Dawdles in getting dressed .055 .437 .004 Factor 3 (Conduct Problems) 23. Teases or provokes other children -.113 .016 .842 27. Physically fights with si sters and brothers -.078 -.032 .654 26. Physically fights with friends own age .097 -.192 .644 24. Verbally fights with friends own age .067 -.034 .637 25. Verbally fights with sist ers and brothers -.190 .060 .611 21. Steals .193 -.084 .447

PAGE 26

26 Table 3-2. ECBI factor correlations Factor 1 Factor 2 Factor 3 Factor 1 ---Factor 2 .52* --Factor 3 .46* .24* -* p <.01

PAGE 27

27 Table 3-3. Items not loading at .4 or higher in the three-factor obli que rotation factor analysis of the ECBI Factor loading Items 1 2 3 35. Is overactive or restless .399 .207 .103 18. Hits parents .288 .038 .219 36. Wets the bed .234 -.143 .116 34. Has difficulty entertaining self alone .039 .355 .081 6. Slow in getting ready for bed .253 .336 -.075 16. Cries easily .213 .327 -.086 28. Constantly seeks attention .186 .326 .122 19. Destroys toys or other objects .357 .058 .379 20. Is careless with toys a nd other objects .134 .202 .328 22. Lies .235 .237 .325 3. Has poor table manners .125 .217 .292

PAGE 28

28 Table 3-4. Pearson correlations between E CBI factors and CPRS and BASC subscales Subscale Oppositional Defiant Behavior factor Attention Difficulties factor Conduct Problems factor CPRS ADHD Index .14 .42** .12 CPRS Oppositional Behavior subscale .70** .37** .55** CPRS DSM-IV Inattention subscale .10 .46** .11 BASC Attention Problems subscale .20* .57** .17 BASC Conduct Problems subscale .57** .47** .63** BASC Aggression subscale .63** .37** .74** CPRS Anxious/Shy subscale .14 .10 .03 BASC Anxiety subscale .06 .09 -.01 *p<.05, **p<.01

PAGE 29

29 CHAPTER 4 SUMMARY AND CONCLUSIONS Consistent with hypotheses, resu lts of the factor an alysis strongly suggest that, when used with a clinical population including children with disruptive be havior disorders, the ECBI is a multidimensional measure. Results demonstrate that item variance is best accounted for by a three-factor solution, resulting in factors la beled Oppositional Defian t Behavior, Attention Difficulties, and Conduct Problems. Here it can be noted that cons idering the unrotated factors w ith eigenvalues greater than 1.0 (Kaiser, 1960) initially suggested a total of nine factors, alt hough that criterion is generally considered to significantly overestimate the number of factors (Zwi ck & Velicer, 1986). Subsequent inspection of the scree plot (C attell, 1966) suggested between three and six discernable factors; thus additional analyses were run forcing three-, four-, fiveand six-factor solutions. The three-factor model clearly appear ed to be the most conceptually meaningful, irrespective of whether oblique or orthogonal rotation stra tegies were employed. As a result, this solution was chosen for validity analyses. Factor analytic findings are mostly consistent with prev ious research. While it was predicted that use of a disruptive behavior disord er clinical sample would result in most ECBI items loading on at least one deri ved factor, a total of 11 items di d not. Of the 11 items that did not load on any factor in the current study, four of the items were included in Burns and Pattersons (2000) fourth j unk factor. Thus it app ears that even in this specific clinical sample, there are ECBI items that do not group together in a conceptually meaningful manner. In contrast to the previous resear ch, use of a four-factor model di d not produce more conceptually meaningful factors (Burns & Patterson).

PAGE 30

30 While previous work by Burns and Patterson (2 000) and Weis et al. (2005) found evidence for a multidimensional structure of the ECBI in community and general clinical samples, the current study aimed to determine the degree to which a multidimensional structure could be delineated in a more specific clin ical population using stricter fact or loading criteria than that employed in prior studies. Here it can be noted th at Burns and Patterson util ized a factor loading criterion of .3 or higher while th e current investigation considered loadings of .4 or higher to indicate a significant item loading on a factor. Results demonstrated that a basic three-factor model maintains even when using more stringent criteria. Thus, the present EFA supports a similar factor structure to that found by previous investigators even when using more rigorous criteria (Burns & Patterson; Weis et al.) continuing to demonstrate sound evidence of a multidimensional model of the ECBI. Results of validity analyses were also consiste nt with hypotheses. The three factors of the ECBI demonstrated adequate to very good intern al consistency. Strong evidence was also found for the convergent validity of factors. Here, as predicted, the Oppositional Defiant Behavior factor was found to correlate most highly with th e CPRS Oppositional Behavior subscale. This factor also correlated significantly with the BASC Conduct Problems and Aggression subscales, although these correlations were smaller. There was also a si gnificant correlation with the BASC Attention Problems subscale but, again, this correlation was smaller. The Attention Difficulties factor correlated most highly with the BASC Attention Problems subscale. This correlation was followed closel y by significant correlations w ith the CPRS ADHD Index and CPRS Inattention subscale. Somewhat surprising ly, the Attention Difficulties factor was also significantly correlated with the BASC Conduct Problems, Aggression, and Oppositional Behavior subscales. Lastly, the Conduct Problems factor was most highly correlated with the

PAGE 31

31 BASC Aggression subscale and also had a si gnificant correlation with the BASC Conduct Problems subscale, as predicted. This factor also correlated significantly with the CPRS Oppositional Behavior subscale but not with an y of the other BASC or CPRS subscales. Evidence for discriminant validity was also found as none of the th ree factors correlated significantly with either the BASC or CPRS anxiety subscales. As previous studies have only investigated specificity, sens itivity and other aspects of diagnostic and categorizing abilitie s of a three-factor model of th e ECBI, our study is the first to assess the convergent and discriminant validity of derived factors. Overall, results provide strong support for the multidimensional structure of the EBCI. While multiple factor analytic studies now provide support for some varia tion on a three-factor solution, the present investigation moves the field furt her ahead by not simply relying on the face validity of factor label designations. These validity analyses give empirical support that allows for more confident use of the Oppositional Defiant Behavior, Attent ion Difficulties, and Conduct Problems labels applied to the three factors of the ECBI in our study and in previous research. While our investigation adds to the literature and provides updated support for the threefactor structure of the ECBI, it is not without limitations. The present study used a sample size of 181, which meets minimum criteria of five cas es per item for a factor analysis (Floyd & Widaman, 1995). While meeting minimum requi rements, larger samples are generally considered better and some argue that a minimu m of 200 or 10 cases per item is required (Floyd & Widaman). Again sample size can also affect nor mality, which in turn can affect the clarity of the model, possibly explaining why the current st ructure did not perfectly match with previous findings. While EFA is more robust to such viola tions of normality than CFA, this aspect of the data cannot be ignored.

PAGE 32

32 Another limitation involves th e presence of cross-loading it ems. In our study, two items cross-loaded on another factor. These items were Whines, which loaded primarily on the Oppositional Defiant Behavior factor and sec ondarily on the Attention Problems factor, and Interrupts, which loaded primarily on the A ttention Problems factor and secondarily on the Oppositional Defiant Behavior factor. There ar e several reasons that may account for such cross-loadings. First, the Whines item may no t necessarily be associated with any of the constructs assessed by the three factors; it is not representati ve of any of the specific DSM-IV criteria for the behavioral di sorders (APA, 1994). As the E CBI was designed before the DSM included criteria for the disruptiv e behavior disorders, it does c ontain some items which, based on current DSM-IV criteria a nd diagnostic categories, may not clearly associate with one symptom category versus another. It was surprisi ng that the Interrupts item cross-loaded as this is a diagnostic criterion for ADHD, though it is grouped in the impulsivity category within the DSM division of symptoms. This item though, cross-loaded on the oppositional behavior factor. While interrup ting may not be a specific sympto m of ODD, it is an avenue through which to annoy authority figures and purposely annoying adults is a specific symptom of ODD. Again, sample size may have affected the clarity of results and use of a larger sample might eradicate the presence of cross-loadings. While the validity findings overall suggested that the three factors found in factor analysis demonstrate good convergent and discriminant va lidity when compared to psychometrically established measures of similar c onstructs, they were not perfectly in line with predictions. Both the Oppositional Defiant Behavior and Conduct Pr oblems factors, while correlating most highly with their conceptual counterparts, also co rrelated significantly with the CPRS and BASC subscales that related to CD a nd ODD, respectively. This limita tion of the validity results could

PAGE 33

33 be explained by the nature of the constructs the two factors represen t. As stated previously, prior discussions in the literature have focused on whether to consider th e two disorders on a continuum or view them as separate and distin ct constructs (Loeber et al., 1991). The DSM-IV itself notes that there is overlap between th e two disorders (e.g. disobedience) even while viewing them as two separate constructs (APA 1994). Hence, it is not surprising that two factors representing oppositional defiant behavior and conduct problems would correlate with measures of both categories, due to high levels of comorbidity between the two disorders. Still, the highest correlations with the more similar co nstructs of the CPRS and BASC highlight the distinctness between the two factors. A similar explanation exists for the fact that th e attention difficulties f actor correlated with all of the BASC and CPRS subscales except for the anxiety subscales. ADHD is highly comorbid with ODD and CD, and all three are categor ized as disruptive behavior disorders. As such, it would not be surprising to find correlations with constructs related to these disorders. These correlations were slightly higher than expected in the current study, though still generally smaller than the correlation with ADHD rela ted subscales. In a ddition, many behaviors associated with ADHD can be confused with symp toms of ODD. For instance, a child who is not paying attention because of atte ntion problems can be perceived as simply being disobedient. Lack of attention may cause children to need in structions or requests re peated to them several times. Items in the oppositional behavior factor of the ECBI represent these types of behaviors. If caregivers associate lack of a ttention with these behaviors (e.g. needing to repeat instructions) then a correlation between the Attention Diffi culties factor and ODD-related subscales is understandable.

PAGE 34

34 Regardless of the limitations of the study, data presented here continue to support the idea that the ECBI is best accounted for by a thre e-factor structure with validated factors. Nevertheless, more research is still needed. While Burns and Patterson (2000) suggested cutoff scores for the ECBI factors, the usefulness of these cutoff scores has only bee tested in the study by Weis et al. (2005). Here, the investigators focused on the ability of the ECBI factors to differentiate between specific behavior clusters in a restricted sample of young children. Given these limited findings, these cutoff scores should be validated with a wider age range as well as in both general clinical and more specific clinical samples si milar to the one used in our study. This would allow researchers to better ascertain th e utility of the factors to differentiate between children with different clinical problems (i.e. attention problems versus oppositional behavior, disruptive behavior problems versus depression or anxiety). Although re searchers have looked at specificity and sensitivity of the ECBI f actors (Weis et al.), the DBDRS used in these discriminative validity analyses is a clinicia n rating scale which does not include information often considered when deciding on a diagnosis (e.g. age of onset of symptoms, aspects of impairment, teacher data) and is also based partia lly on parent report. This, as the researchers note, introduces a degree of shared method vari ance between the ECBI and DBDRS ratings. The ECBI factors could be best evaluated by assessi ng their ability to differentiate between children who have been independently diagnosed with th ese various disorders. This would provide support for the usage of the ECBI as a pre-screen er to indicate areas wh ich may require further assessment. Future research should include continued validat ion of the three-factor structure. Using a larger specific clinical sample, with which mu ltivariate normality assumptions can be met, a CFA of the current results should be conducted. Again, these factors should be subjected to

PAGE 35

35 convergent and discriminant validity analyses, sensitivity and specificity tests in the manner described above, as well as furthe r assessment of internal consistency. Test-retest reliability of the factors should be ascertained along with th e individual factors se nsitivity to treatment effects. Once these analyses have been conducted, researchers and clinicians can begin to extract even more information from the ECBI than they can currently obtain.

PAGE 36

36 LIST OF REFERENCES American Psychiatric A ssociation (APA). (1994). Diagnostic and statistica l manual of mental disorders (4th ed.). Washington, DC: Author. Barkley, R.A. (1988). Attention-deficit/hyperactivity disorder: A handbook for diagnosis and treatment (2nd ed.). New York: Guilford Press. Barkley, R.A. (1997). Defiant Children New York: Guilford Press. Boggs, S.R., Eyberg, S., & Reynolds, L.A. (1990). Concurrent validity of the Eyberg Child Behavior Inventory. Journal of Clinical Child Psychology, 19, 75-78. Burns, G.L., & Patterson, D.R. (1991). Factor stru cture of the Eyberg Child Behavior Inventory: One-dimensional or multidimensional measure of disruptive behavior? Journal of Clinical Child Psychology, 20, 439-444. Burns, G., & Patterson, D.R. (2000). Factor struct ure of the Eyberg Child Behavior Inventory: A parent rating scale of oppositional defiant behavior towards adults, inattentive behavior, and conduct problem behavior. Journal of Clinical Child Psychology, 29, 569-577. Cattell, R.B. (1966). The scree test fo r the number of factors. Multivariate Behavioral Research, 1, 245-76. Colvin, A., Eyberg, S.M., & Adams, C.D. (1999). Restandardization of the Eyberg Child Behavior Inventory Gainesville, Florida: University of Florida, Child Study Laboratory. Conners, C.K. (1997). Conners Teacher and Parent Rating Scales Revised Toronto: MultiHealth Systems. Eyberg, S.M. (1992). Parent and teacher behavi or inventories for the assessment of conduct problem behaviors in children. In L. Va ndeCreek, S. Knapp, & T.L. Jackson (Eds), Innovations in clinical practice: A Source Book (Vol. 12, pp.377-382). Sarasota, FL: Professional Resource Exchange. Eyberg, S.M., & Robinson, E.A. (1983). Conduc t problem behavior: Standardization of a behavior rating scale with adolescents. Journal of Clinical Child Psychology, 11, 130 137. Eyberg, S.M., & Ross, A.W. (1978). Assessment of child behavior problems: The validation of a new inventory. Journal of Clinical Child Psychology, 7, 113-116. Floyd, F.J., & Widaman, K.F. (1995). Factor anal ysis in the development and refinement of clinical assessment instruments. Psychological Assessment, 7, 286-299. Funderburk, B.W., Eyberg, S.M., Rich, B.A ., & Behar, L. (2003). Further psychometric evaluation of the Eyberg and Behar rating scal es for parents and teachers of preschoolers. Early Education and Development, 14, 67-81.

PAGE 37

37 Henson, R. K. (2001). Understanding internal co nsistency reliability estimates: A conceptual primer on coefficient alpha. Measurement and Evaluation in Counseling and Development, 34, 177-189. Kaiser, H.F. (1960). The application of el ectronic computers to factor analysis. Educational and Psychological Measurement, 20, 141-151. Loeber, R., Lahey, B.B., & Thomas, C. (1991) Diagnostic conundrum of oppositional defiant disorder and conduct disorder. Journal of Abnormal Psychology, 100, 379-390. Reynolds, C.R., & Kamphaus, K.W. (2004). Behavioral assessment system for children (2nd ed.). Circle Pines, MN: Am erican Guidance Service. Robinson, E.A., Eyberg, S.M., & Ross, A.W. (19 80). The standardization of an inventory of child conduct problems. Journal of Clinical Child Psychology, 9, 22-29. Tabachnick, B.G., & Fidell, L.S. (2007). Using multivariate statistics (5th Ed.). Boston, MA: Pearson/Allyn & Bacon. Weis, R., Lovejoy, M.C., & Lundahl, B.W. (2005). Factor structure and discriminative validity of the Eyberg Child Behavior Inventory with young children. Journal of Psychopathology and Behavioral Assessment, 27, 269-278. Werba, B.E. (2003). Standardization and cross-cu ltural validity of the Ey berg Child Behavior Inventory with Australian preschoolers. (Doc toral dissertation, Univ ersity of Florida, 2003). Dissertation Abstra cts International, 63, 4391. Zwick, W.R., & Velicer, W.F. (1986 ). A comparison of five rules for determining the number of components to retain. Psychological Bulletin, 99, 432-442.

PAGE 38

38 BIOGRAPHICAL SKETCH Melissa Kay Stern was born on March 27, 1981 in Boston, Massachusetts. The youngest of two, she grew up in the greater Boston area graduating from Needha m High School in 1999. She earned her B.A. in psychology from Brandeis University. Upon graduating with high honors in 2003, Melissa took a position as a research coordinator at the Massachusetts General Hospital (MGH) and Freedom Trail Clinic in Boston, Mass achusetts. She spent 15 months coordinating research projects at the hospita ls schizophrenia and severe me ntal illness outpatient clinic. In September of 2004, Melissa left MGH fo r a position at the National Center for Posttraumatic Stress Disorder (PTSD) at Bo ston University and the Boston Veterans Administration, where she was employed as a resear ch technician for projects investigating the psychophysiological and cognitive effects of nicotin e on veterans with PTSD. Melissa relocated to Gainesville, Florida in August of 2005 to begi n her graduate career in clinical psychology. Upon completion of the M.S., Melissa will contin ue on in her program and apply for doctoral candidacy in the Department of Clinical and Health Psychology at the University of Florida.


Permanent Link: http://ufdc.ufl.edu/UFE0020122/00001

Material Information

Title: Assessing Dimensions of Disruptive Child Behavior with the Eyberg Child Behavior Inventory
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: UFE0020122:00001

Permanent Link: http://ufdc.ufl.edu/UFE0020122/00001

Material Information

Title: Assessing Dimensions of Disruptive Child Behavior with the Eyberg Child Behavior Inventory
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: UFE0020122:00001


This item has the following downloads:


Full Text





ASSESSING DIMENSIONS OF DISRUPTIVE CHILD BEHAVIOR WITH THE EYBERG
CHILD BEHAVIOR INVENTORY





















By

MELISSA K. STERN


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

UNIVERSITY OF FLORIDA

2007


































2007 Melissa K. Stern









ACKNOWLEDGMENTS

I thank my supervisory committee chair, Dr. James H. Johnson, for his invaluable guidance

and mentoring. I would also like to thank the members of my committee for their suggestions

and comments. I thank the students and research assistants of the ADHD study lab and members

of the child area in the Clinical and Health Psychology Department for their help and support.










TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS .............. ................3

L IS T O F T A B L E S ......... ............ ..................................... ............................................. 5

F IG U R E .............. ..... ............ ................................... ............................................ 6

ABSTRAC T ........................................................................... 7

CHAPTER

1 INTRODUCTION ............... .............................................................9

ADHD and the Disruptive Behavior Disorders............................................. .......................9
The Eyberg Child Behavior Inventory (ECBI).....................................................................10

2 M E T H O D S .......................................................................................................17

D ata C o lle ctio n ................................................................................................................. 1 7
M e a su re s ................... ...................1...................8..........
T h e E C B I ................... ...................1...................8..........
The C PR S and B A SC ........................................................................................... .. ...... ..... 18
P artic ip an ts .........................................................................19
S statistic al A n aly se s ........................................................................................................... 19
F a cto r A n a ly se s ...................................................................................................... 1 9
V validity A n aly ses ................................................................2 1

3 R E SU L T S .............. ... ................................................................22

F actor A nalytic F findings ................................................................22
V validity A nalyses................................................... 23

4 SUMMARY AND CONCLUSIONS ................................. ................................ 29

L IST O F R EFE R E N C E S ............................................................................... 36

B IO G R A PH IC A L SK E T C H ................................................................................................... 38












4









LIST OF TABLES


Table Page

3-1 Factor loadings for principal axis factoring via oblique rotation three-factor analysis
o f th e E C B I ...........................................................................................2 5

3-2 E C B I factor correlations .......................................................................... ....................26

3-3 Items not loading at .4 or higher in the three-factor oblique rotation factor analysis of
th e E C B I............... .............. .............. ................................................2 7

3-4 Pearson correlations between ECBI factors and CPRS and BASC subscales...................28









FIGURE


Figure Page

3-1 Scree plot of the ECBI individual frequency item data. .................................................24









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

ASSESSING DIMENSIONS OF DISRUPTIVE CHILD BEHAVIOR WITH THE EYBERG
CHILD BEHAVIOR INVENTORY

By

Melissa K. Ster

May 2007

Chair: James H. Johnson
Major: Psychology

The Eyberg Child Behavior Inventory (ECBI) is a well-validated parent rating scale that

assesses behavior problems in both children and adolescents. While originally conceptualized as

a one-dimensional measure of disruptive behavior, there are discrepancies in the literature

regarding its factor structure, with some studies finding a single factor and others finding support

for a multidimensional structure. While previous research samples have been diverse, additional

research is needed on other clinical samples. The primary goal of our study was to further assess

the factor structure of this measure using data obtained from a clinical sample displaying a wide

range of disruptive behaviors such as children referred for evaluation of Attention Deficit

Hyperactivity Disorder (ADHD) and associated conditions.

Data from parents of 181 children referred to an ADHD Program were used for the

analyses. Upon initial assessment in the ADHD clinic, parents completed a number of

assessment measures that included the ECBI as well as measures used in validity analyses. A

factor analysis using principal axis factoring and promax rotation was performed on the

individual item data of the ECBI. Analysis of eigenvalues greater than one, item distribution

breakdown, and inspection of the scree plot revealed that the ECBI was best accounted for by a

three-factor structure with factors representing oppositional defiant behavior, attention problems,









and conduct problems. Convergent and discriminant validity analyses using measures of

relevant constructs from psychometrically sound measures demonstrated good validity for all

three factors. Current research continues to demonstrate that the ECBI is a valid tool with a

multidimensional structure, which could increase the measure's utility in both clinical and

research contexts.









CHAPTER 1
INTRODUCTION

ADHD and the Disruptive Behavior Disorders

Attention Deficit Hyperactivity Disorder (ADHD) is one of the most commonly diagnosed

childhood behavioral disorders occurring in approximately 3 to 5% of the general child

population (American Psychiatric Association (APA), 1994). Barkley (1998) suggests that

children with this disorder also comprise approximately 40% of referrals to outpatient clinics.

The guidelines defined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth

Edition (DSM-IV) are generally used to make diagnostic decisions regarding the presence or

absence of ADHD. Current criteria include clinically significant levels of symptoms in the

domains of attention, hyperactivity and impulsivity; symptom onset before 7 years of age;

symptoms that are present across situations; that cause impairment in social, academic, or

occupational functioning; and symptoms that have occurred apart from other mental disorders.

Subtypes of the disorder are based on these dimensions and include the predominantly

inattentive type, the predominantly hyperactive/impulsive type, and the combined type.

ADHD also has high rates of comorbidity with Oppositional Defiant Disorder (ODD) and

Conduct Disorder (CD). ODD is defined as a pattern of hostile, defiant, and negativistic

behavior towards adults (APA, 1994). Symptoms include arguing with adults, actively defying

rules or requests, as well as often being angry and vindictive. To meet diagnostic criteria for

ODD, a child must have exhibited the pattern of defiant behavior for at least 6 months, must

show evidence of impairment, and the symptoms must not be due to another mental disorder.

Conduct Disorder is characterized by a pattern of behavior that violates the basic rights of others

or age-appropriate societal rules (APA). Symptoms are grouped into the categories of aggression

towards people or animals, destruction of property, deceitfulness or theft, and serious violations









of rules such as running away. Subtypes for conduct Disorder are based on age of onset and

include childhood-onset type and adolescent-onset type. All three of the described disorders are

viewed as disruptive behavior or externalizing disorders and are highly comorbid with one

another (APA). Indeed, the high degree of comorbidity between ODD and CD has prompted

some researchers to suggest that they are not, in fact, separate disorders, although a variety of

studies clearly support distinctions between these two conditions (Loeber, Lahey, & Thomas,

1991).

While several measures exist that assess for symptoms of each of these disorders (e.g.

structured interviews, Child Behavior Checklist, Diagnostic Interview Schedule for Children), no

widely used brief questionnaire or screener assesses all of these domains. A short measure that

accomplishes such a task has the potential for great clinical and research utility.

The Eyberg Child Behavior Inventory (ECBI)

The Eyberg Child Behavior Inventory (ECBI) is a parent rating scale which assesses

disruptive behavior problems in both children and adolescents (Eyberg & Robinson, 1983). The

ECBI was designed to differentiate normal children and adolescents from those with conduct

problems and as a follow-up instrument for assessing treatment effects for a broad age range

(Eyberg & Ross, 1978). While demonstrating good psychometric characteristics in previous

research (Boggs, Eyberg, & Reynolds, 1990; Funderburk, Eyberg, Rich & Behar, 2003;

Robinson, Eyberg, & Ross, 1980), there are discrepancies regarding the factor structure of the

ECBI. Eyberg designed the ECBI as a one-dimensional assessment scale of general conduct

problems and studies have found support for such a structure (Colvin, Eyberg, & Adams, 1999;

Eyberg, 1992). Other researchers, however, have found evidence of a multidimensional structure

of the ECBI. In an examination of the cross-cultural validity of the ECBI in a sample of

Australian preschoolers, Werba (2003) found evidence for a two-factor structure of the ECBI.









The author found similar results when repeating the exploratory factor analysis on a sample of

parents of preschoolers in the United States. Results from both analyses, while not identical,

indicated the factors represented the constructs of ODD and ADHD.

Taking a closer look at individual items from the ECBI, they seem representative of three

specific categories and similar to specific DSM-IV (APA, 1994) criteria for ODD, ADHD, and

CD. For example, the ECBI item "Has short attention span" is similar to the DSM-IV criterion

"Often has difficulty sustaining attention in tasks or play activities." Similarly, the ECBI item

"Acts defiant when told to do something" is similar to the DSM-IV ODD criterion "Often

actively defies or refuses to comply with adult's requests or rules." Another ECBI item, "Teases

or provokes other children," appears at face-value to be similar to the DSM-IV CD criterion

"Often bullies, threatens, or intimidates others." When continuing to compare the ECBI item by

item to DSM-IV criteria for the three disorders, many other examples are revealed. As ODD,

CD, and ADHD comprise the primary childhood disruptive behavior disorders, it is not

surprising that a measure such as the ECBI may have a factor structure representing aspects of

these three areas.

Some researchers have, in fact, found evidence for a three-factor structure of the ECBI.

Burns and Patterson (1991) collected data from parents of 1,526 children in outpatient clinics in

Montana, Washington, Idaho, and Oregon. In their exploratory factor analysis they utilized

principal components analysis with varimax rotation on the frequency scores of the ECBI.

Results indicated a three-factor structure with factors representing oppositional-defiant behavior,

attentional difficulties, and violations of others' rights via aggression.

Burns and Patterson (1991) attempted to replicate their three-factor findings of the ECBI

using a random sample of children from an urban school district in Washington. Again using









principal components analysis and varimax rotation, Bums and Patterson performed an

exploratory factor analysis, though limiting the resulting model to three factors as they had an

a priori hypothesis based on results from their first study. Findings supported a three-factor

solution representing the categories they had identified in previous work. Bums and Patterson

(2000) later combined their urban and original sample and divided them into two subsequent

random samples in order to further evaluate the factor structure of the ECBI. Using exploratory

factor analysis with maximum likelihood extraction and oblique rotation, they examined two- to

seven-factor solutions as results indicated seven eigenvalues greater than one. Here, Bums and

Patterson found a four-factor solution to be most relevant. The first three factors again

represented symptomatology of ODD, attention problems, and CD, but the fourth factor did not

represent a discernible dimension. The researchers explained this fourth factor as a possible

response bias effect which resulted in a meaningless factor as can happen when items with low

but significant loadings on the first factor separate and form their own separate factor. Burns and

Patterson chose the four-factor solution because it resulted in the three primary factors becoming

more conceptually clear.

Burns and Patterson (2000) next took their second random sample and performed a

confirmatory factor analysis on the ECBI using the first three factors of the four-factor model

(the fourth factor was not used as it was judged conceptually meaningless). Bums and Patterson

also compared the three-factor solution to a one-factor solution and a two-factor solution (which

combined the ODD and CD factors). Results from the initial three-factor confirmatory analysis

indicated adequate to good fit, and a significantly better fit compared to the one- and two-factor

models. Burns and Patterson also identified 90th percentile cutoff scores from their community









sample to identify children with significant levels of oppositional behavior, attention difficulties,

and conduct problems.

Still, other researchers continue to find support for a one-factor structure of the ECBI

(Colvin et al., 1999). Using data collected for the restandardization of the ECBI, researchers

found, via four different methods of factor analysis, that the ECBI was best accounted for using a

one-factor model. These researchers utilized principal components analysis with both varimax

and oblimin rotations, and principal axis factoring with varimax and oblimin rotations. A

conceptually meaningful one-factor model was found with all four methods of analysis.

In 2005, Weis, Lovejoy, & Lundahl performed a confirmatory factor analysis using

individual ECBI item data to compare the previously described Bums and Patterson (2000)

three-factor model to a one- and two-factor model using a large general clinical sample of young

children. In this analysis, the researchers used data from different studies of stress, affect, and

parenting which included ECBIs completed by a predominantly Caucasian sample of parents of

children ages two to six. The study demonstrated that Burns and Patterson's three-factor model

was an adequate fit to the combined data set. A one-factor model demonstrated a poor fit to the

data. The two-factor model (collapsing the ODD and CD categories) did demonstrate adequate

fit. It can be noted, however, that the three-factor model was a significantly better fit to the data

than either the one- or two-factor model.

In addition to their confirmatory factor analysis, Weis et al., (2005) completed a second

study assessing the discriminative validity of the three-factor model of the ECBI. In this

investigation, the researchers recruited caregivers of 115 children referred to an outpatient

psychology clinic. In order to assess the discriminative validity of the Burns and Patterson cutoff

scores described earlier, Weis et al. had caregivers complete an ECBI. The primary aim of this









study was to assess the ability of cutoff scores to categorize children with and without significant

disruptive behavior problems. The second aim of the study was to discriminate between children

who do or do not display symptoms of ADHD, ODD and CD as determined by the Disruptive

Behavior Disorder Rating Scale (DBDRS; Barkley, 1997). Results demonstrated that the three

factors had adequate negative predictive power and poor positive predictive power. Sensitivity

for the three factors was poor (.63 to .77) but results demonstrated good specificity (.91 to .94).

Results of a MANOVA using the factor scores of the ECBI as the multivariate dependent

variable and the symptom groups from the DBDRS as the independent variables suggested that

the Attention Difficulties factor did not appear to discriminate children with attention problems

from children with ODD or CD; the same was true of the oppositional factor. The Conduct

Problems factor did appear to adequately differentiate children with aggressive behaviors from

those with attention problems or ODD. While these researchers did extend the work of Bums

and Patterson, the limited age range and nonspecific general clinical sample may limit results.

The second sample used in the study was also not large and individual diagnostic groups were

not well represented. Thus, while the study adds to the literature, further research is needed.

In these previous studies, samples have included participants recruited through pediatric

clinics and schools, general outpatient psychology clinics, childcare centers, family service

agencies, and newspaper advertisements. While researchers have previously attempted to assess

the fit of a three-factor model on a general clinical sample (Weis et al., 2005), additional research

is needed on other clinical samples. In addition, it is important to assess the validity of factors

resulting from such research and, perhaps later, the value of cutoff scores for well validated

factors. Research along these lines might provide the basis for using the ECBI as a brief,

preliminary screener for specific problem areas, rather than just as an overall index of disruptive









behavior. Additionally, it can be noted that the ECBI was also designed to assess change in

conduct problem behavior over the course of treatment (Eyberg & Ross, 1978). Here, the utility

of the ECBI could potentially be increased threefold in terms of providing change data in

multiple areas that may or may not be directly targeted by treatment (e.g., ODD, CD, and ADHD

related symptoms). This, in turn, might decrease time spent in the assessment process by

reducing the need for multiple pre-screeners and follow-up assessments. Clinicians, researchers,

and patients will save time if such a measure is available. While the ECBI has already been

validated as a brief screening instrument for both children and adolescents with conduct

problems, it may be able to provide even more information.

The current study aimed to add to the literature by further assessing the factor structure of

the ECBI and conducting validity analyses on the factors of the model. While previous

investigations have studied diverse samples, more research is needed on clinically relevant

samples. One such sample would be one that could be expected to display symptomatology for

all three of the disorders purported to be assessed by the ECBI. As ADHD is highly comorbid

with both ODD and CD, a sample of children referred to an ADHD clinic would likely display

the symptomatology of all three disorders. Specifically, it was expected that multiple factors of

the ECBI would be identified using such a sample and that these factors would be representative

of symptomatology of attention problems, oppositional behavior, and conduct disorder. As

previous factor analyses found that several items did not load on any of the factors, it was hoped

that this relevant clinical sample would be able to utilize all of the ECBI items, some of which

may not have appeared as salient in the more general samples in previous research. Secondarily,

we expected that factors found via factor analysis would demonstrate good convergent and









discriminant validity when assessed in relation to psychometrically established measures

designed to measure similar constructs.









CHAPTER 2
METHODS

Data Collection

Data for this study was collected from children (and their parents) who were involved with

the interdisciplinary ADHD program at the University of Florida. This program is staffed by

professionals from various disciplines (e.g. psychology, pediatrics, child psychiatry, medicine,

nursing, speech pathology) from several departments at the University of Florida and Shands

Teaching Hospital. Children are referred to the program for assessment and follow-up by

parents, teachers and physicians from the local community and from throughout North Central

Florida. Some children referred to the program have not received a complete assessment and

have no official diagnosis at the time of referral. Others with a diagnosis have not undergone a

detailed assessment such as that available at the University of Florida. When a child is referred

to the program, the child and caregiver(s) participate in an all-day assessment. This includes

parent- and teacher-report (and sometimes child-report depending on age) measures to assess for

ADHD symptoms and possible comorbid features, a clinical interview, intellectual and

achievement testing, as well as other measures, as necessary. The data from the psychological

assessment are presented at an ADHD Program meeting, along with data obtained by the

pediatrician associated with the ADHD program. This information is used by the ADHD

interdisciplinary team to determine whether additional assessment by other professionals is

needed and to determine recommendations and the appropriate diagnosis.

An application for exempt status was submitted to the University of Florida Institutional

Review Board as data was collected without identifiers. Once exempt status was obtained, data

were collected from the files and entered into a database for analyses. Demographic data

collected included age, gender, ethnicity, education level, and diagnoses of the child.









Information on the age, gender, education level, and family income of the caregiver who

completed the forms was also recorded. Individual item data from the ECBI as well as scores

from several other measures used for validity analyses were also entered into the database.

Measures

The ECBI

The ECBI is a 36-item parent-report measure designed to assess conduct problems in both

children and adolescents (Eyberg & Ross, 1978). The ECBI has two scales: a problem scale and

a frequency scale. Caregivers are asked to indicate how often a particular behavior or problem

occurs on a 7-point likert-type scale. The scale ranges from anchors "never" to "always." The

problem scale asks caregivers to indicate in a yes or no format whether they consider the

particular behavior to be a problem regardless of the frequency. The ECBI has demonstrated

good internal consistency, test-retest reliability, and sensitivity to treatment effects (Colvin et al.,

1999; Eyberg & Robinson, 1983; Robinson et al., 1980).

The CPRS and BASC

As part of the child's assessment, caregivers complete the Conners Parent Rating Scale

(CPRS; Conners, 1997) and the Behavioral Assessment System for Children (BASC; Reynolds

& Kamphaus, 2004). The CPRS and BASC are broadband measures that have undergone

psychometric validation in previous research. The CPRS is an 80-item questionnaire that

assesses ADHD symptoms and associated features and provides measures used to assess the

validity of factors found in the present factor analysis. The BASC is a slightly longer broadband

measure, completed by a parent or caregiver, which assesses symptoms related to various types

of child psychopathology and child difficulties that were also useful in assessing validity of

derived factors in the present study. Caregivers may have completed either the first or second









edition of the BASC as the measure underwent revision and was republished in 2004. However,

both versions contain the subscales used in analyses for the current study.

Participants

Data were collected from 181 caregivers of children referred to the ADHD clinic at the

University of Florida. The average age of the child referred to the clinic was 8.75 years old

(SD=2.93) and 75% (N=135) were male. Consistent with Alachua County census data, 75%

(N=136) of the sample identified as Caucasian, 18% (N=32) identified as African-American, 2%

(N=3) identified as Hispanic, and the remaining 5% (N=10) of the sample identified as another

ethnicity, multiracial, or did not specify. The average level of education of the child was the

third grade. Of the children on whom ECBIs were completed, 92% (N=166) were given a

diagnosis of ADHD. Of those diagnosed with ADHD, 79% (N=131) were combined type, 17%

(N=28) were predominantly inattentive type, 2% (N=4) were predominantly

hyperactive/impulsive type, and 2% (N=3) were categorized as ADHD NOS. A total of 22%

(N=40) of the children met diagnostic criteria for ODD and 3% (N=5) met diagnostic criteria for

CD.

Statistical Analyses

Factor Analyses

In order to accomplish the aim of further exploring the factor structure of the ECBI, an

exploratory factor analysis (EFA) was conducted on the individual item data of the ECBI

frequency scale. EFA, as opposed to confirmatory factor analysis (CFA), was chosen for several

reasons. First, while several studies have found multidimensional structures, these studies are

few in number and were conducted on limited samples. While Bums and Patterson (1991, 2000)

used large samples in their analyses, these participants represented community samples rather

than samples drawn from a clinical population. The same is true of the study with Australian









preschoolers (Werba, 2003). Even though Weis et al. (2005) performed a CFA, their sample was

also not a specific clinical sample; perhaps a different model would have fit the data better had a

more relevant sample been used. Here it can also be noted that Weis et al. performed their CFA

on the three-factor model from Burns and Patterson which excluded nine of the ECBI items. It

was hoped that use of a relevant clinical sample would utilize all 36 of the ECBI items. Hence, it

was reasonable to conduct an EFA on the current clinical sample in order to explore the factor

structure that would be most robust in a sample likely to display symptomatology of all three

disorders. Secondly, while multivariate normality is generally helpful in providing a clear factor

structure, the assumption is much more pertinent to CFA than EFA (Floyd & Widaman, 1995).

As the current study sample is not a large sample, multivariate normality is harder to achieve and

the techniques employed in an EFA are not as sensitive to non-normally distributed data. Thus,

the validity of the resulting factor structure is somewhat increased by the use of EFA as opposed

to CFA. A minimum sample size of 180 was chosen based on the generally accepted rule of

needing five cases per item in order to conduct a valid analysis (Floyd & Widaman).

Specifically, the study extracted factors via principal axis factoring, a technique which is

not thought to be as constrained by assumptions of multivariate normality (Floyd & Widaman,

1995). Two kinds of rotation were employed in the current study. First, all models were rotated

via oblique rotation. As ADHD is highly comorbid with both ODD and CD it is logical to

expect that factors representing symptoms of the three disorders would be correlated. All

analyses were rerun using orthogonal rotation to see if use of the independent rotation strategy

resulted in a clearer factor structure. An initial EFA was run without specifying the number of

factors. After investigation of the scree plot and eigenvalues, subsequent EFAs were run forcing

specific-numbered solutions.









Validity Analyses

Pearson correlations were used to assess convergent and discriminant validity of factors

found in factor analyses. Here, the ECBI factors were correlated with scores on various

subscales of the CPRS and BASC. Specifically, the CPRS contains an ADHD Index and a

DSM-IV Inattentive subscale which were expected to correlate highly with the proposed

Attention Difficulties factor of the ECBI. The CPRS also contains an Oppositional Behavior

subscale which was expected be highly correlated with the Oppositional Defiant Behavior factor

of the ECBI. The BASC has an Attention Problems subscale which was expected to be highly

correlated with the Attention Difficulties factor of the ECBI. The BASC also has both a Conduct

Problems and an Aggression subscale, which were hypothesized to be highly correlated with the

Conduct Problems factor of the ECBI. The BASC and CPRS both contain some form of an

anxiety subscale and these were hypothesized to not be correlated with any of the three proposed

factors of the ECBI.









CHAPTER 3
RESULTS

Factor Analytic Findings

Results of the initial factor analyses revealed nine factors with eigenvalues greater than

one. Inspection of the scree plot (See Figure 3-1) demonstrated two points of leveling-off, one

point at three factors and one point at six factors. Thus, four more factor analyses were run

forcing three-, four-, five-, and six-factor solutions. This was completed using oblique rotation

initially and then again with orthogonal rotation. Conceptually, results were best accounted for

by the three-factor oblique rotation with factors representing oppositional defiant behavior,

attention problems, and conduct problems. Results of this model are presented in Table 3-1.

Items were considered to load on a factor if they had a loading of .4 or higher, a generally

accepted level (Floyd & Widaman, 1995). This three-factor solution involved a total of 25 of the

36 ECBI items.

Here it can be seen that factor 1, labeled Oppositional Defiant Behavior, consisted of 12

items which seemed to reflect the pattern of defiant and negativistic behavior towards adults

associated with ODD. The second factor, labeled Attention Difficulties, consisted of seven items

that, at face value, appeared representative of the attention problems often experienced by

children diagnosed with ADHD. Lastly, the third factor, labeled Conduct Problems, was

comprised of six items which appeared to represent aggression and violation of the rights of

others and societal norms, a symptom pattern that is associated with Conduct Disorder.

Correlations among factors are presented in Table 3-2. The EFA resulted in 11 items not

loading on a factor at .4 or higher (See Table 3-3). Only two items cross-loaded on another

factor at a factor loading of .2 or higher. These items were "Whines" and "Interrupts."









Factor scores were computed as the sum of the frequency ratings for the items loading at .4

or higher on the three factors. Summation was used as it is considered an adequate method of

computing factor scores (Floyd & Widaman, 1995; Tabachnick & Fidell, 2007) and it is

consistent with the computation of factor scores employed in prior studies (Burs & Patterson,

2000; Weis et al., 2005). The 11 items that did not show significant loadings on any factor were

not included in the computation of factor scores. In this sample, the Oppositional Defiant

Behavior factor had a mean score of 53.96 (SD=12.42), the Attention Difficulties factor had a

mean score of 36. 98 (SD=8.19), and the Conduct Problems factor had a mean score of 17.5

(SD=7.95). Cronbach's alpha for the Oppositional Defiant Behavior, Attention Difficulties, and

Conduct Problems factors were .92, .85, and .80 respectively, suggesting adequate to very good

internal consistency of the factors (Henson, 2001).

Validity Analyses

Factor scores were correlated with various subscales of the BASC and CPRS (See Table

3-4). Examination of the correlation matrix, as predicted, revealed that the ECBI factors

correlated most highly with their conceptual counterpart subscales of the BASC and CPRS.

None of the ECBI factors correlated significantly with the anxiety subscales of the BASC and

CPRS
















12-



10-



8-



C 6-
0)


4-



2-



0-

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Factor Number

Figure 3-1. Scree plot of the ECBI individual frequency item data.










Table 3-1. Factor loadings for principal axis factoring via oblique rotation three-factor analysis
of the ECBI
Factor loading


1 2 3


Items
Factor 1 (Oppositional Defiant Behavior)
13. Has tempter tantrums
10. Acts defiant when told to do something
12. Gets angry when doesn't get own way
14. Sasses adults
9. Refuses to obey until threatened with punishment
11. Argues with parents about rules
8. Does not obey house rules on own
17. Yells or screams
15. Whines
5. Refuses to do chores when asked
7. Refuses to go to bed on time
4. Refuses to eat food presented

Factor 2 (Attention Difficulties)
30. Is easily distracted
32. Fails to finish tasks or projects
31. Has short attention span
34. Has difficulty concentrating on one thing
29. Interrupts
1. Dawdles or lingers at mealtime
2. Dawdles in getting dressed

Factor 3 (Conduct Problems)
23. Teases or provokes other children
27. Physically fights with sisters and brothers
26. Physically fights with friends own age
24. Verbally fights with friends own age
25. Verbally fights with sisters and brothers
21. Steals


.915
.913
.889
.776
.770
.735
.724
.668
.539
.529
.465
.407


-.258
-.060
.006
.007
.263
-.031
.055


-.113
-.078
.097
.067
-.190
.193


-.125
-.106
-.048
-.067
.084
-.032
.069
.021
.219
.067
.189
.065


.867
.850
.794
.700
.524
.466
.437


.016
-.032
-.192
-.034
.060
-.084


-.137
-.061
-.066
.093
-.027
.040
.013
.113
-.098
-.004
-.091
.009


.029
-.113
-.130
-.046
.055
.192
.004


.842
.654
.644
.637
.611
.447











Table 3-2. ECBI factor correlations
Factor 1 Factor 2 Factor 3
Factor 1 -- -
Factor 2 .52* --
Factor 3 .46* .24* --
*p<.01











Table 3-3. Items not loading at .4 or higher in the three-factor oblique rotation factor analysis of
the ECBI
Factor loading
Items 1 2 3
35. Is overactive or restless .399 .207 .103
18. Hits parents .288 .038 .219
36. Wets the bed .234 -.143 .116
34. Has difficulty entertaining self alone .039 .355 .081
6. Slow in getting ready for bed .253 .336 -.075
16. Cries easily .213 .327 -.086
28. Constantly seeks attention .186 .326 .122
19. Destroys toys or other objects .357 .058 .379
20. Is careless with toys and other objects .134 .202 .328
22. Lies .235 .237 .325
3. Has poor table manners .125 .217 .292











Table 3-4. Pearson


correlations between ECBI factors and CPRS and BASC subscales
Oppositional Defiant Attention Difficulties Conduct Problems
Behavior factor factor factor


.37**


Subscale
CPRS ADHD Index
CPRS Oppositional
Behavior subscale
CPRS DSM-IV
Inattention subscale
BASC Attention
Problems subscale
BASC Conduct
Problems subscale
BASC Aggression
subscale
CPRS Anxious/Shy
subscale
BASC Anxiety
subscale
*p<.05, **p<.01


.55**


.57**

.47**

.37**


.74**


.20*

.57**









CHAPTER 4
SUMMARY AND CONCLUSIONS

Consistent with hypotheses, results of the factor analysis strongly suggest that, when used

with a clinical population including children with disruptive behavior disorders, the ECBI is a

multidimensional measure. Results demonstrate that item variance is best accounted for by a

three-factor solution, resulting in factors labeled Oppositional Defiant Behavior, Attention

Difficulties, and Conduct Problems.

Here it can be noted that considering the unrotated factors with eigenvalues greater than

1.0 (Kaiser, 1960) initially suggested a total of nine factors, although that criterion is generally

considered to significantly overestimate the number of factors (Zwick & Velicer, 1986).

Subsequent inspection of the scree plot (Cattell, 1966) suggested between three and six

discernable factors; thus, additional analyses were run forcing three-, four-, five- and six-factor

solutions. The three-factor model clearly appeared to be the most conceptually meaningful,

irrespective of whether oblique or orthogonal rotation strategies were employed. As a result, this

solution was chosen for validity analyses.

Factor analytic findings are mostly consistent with previous research. While it was

predicted that use of a disruptive behavior disorder clinical sample would result in most ECBI

items loading on at least one derived factor, a total of 11 items did not. Of the 11 items that did

not load on any factor in the current study, four of the items were included in Burns and

Patterson's (2000) fourth "junk" factor. Thus it appears that even in this specific clinical sample,

there are ECBI items that do not group together in a conceptually meaningful manner. In

contrast to the previous research, use of a four-factor model did not produce more conceptually

"meaningful" factors (Burs & Patterson).









While previous work by Bums and Patterson (2000) and Weis et al. (2005) found evidence

for a multidimensional structure of the ECBI in community and general clinical samples, the

current study aimed to determine the degree to which a multidimensional structure could be

delineated in a more specific clinical population using stricter factor loading criteria than that

employed in prior studies. Here it can be noted that Burns and Patterson utilized a factor loading

criterion of .3 or higher while the current investigation considered loadings of .4 or higher to

indicate a significant item loading on a factor. Results demonstrated that a basic three-factor

model maintains even when using more stringent criteria. Thus, the present EFA supports a

similar factor structure to that found by previous investigators even when using more rigorous

criteria (Burns & Patterson; Weis et al.), continuing to demonstrate sound evidence of a

multidimensional model of the ECBI.

Results of validity analyses were also consistent with hypotheses. The three factors of the

ECBI demonstrated adequate to very good internal consistency. Strong evidence was also found

for the convergent validity of factors. Here, as predicted, the Oppositional Defiant Behavior

factor was found to correlate most highly with the CPRS Oppositional Behavior subscale. This

factor also correlated significantly with the BASC Conduct Problems and Aggression subscales,

although these correlations were smaller. There was also a significant correlation with the

BASC Attention Problems subscale but, again, this correlation was smaller. The Attention

Difficulties factor correlated most highly with the BASC Attention Problems subscale. This

correlation was followed closely by significant correlations with the CPRS ADHD Index and

CPRS Inattention subscale. Somewhat surprisingly, the Attention Difficulties factor was also

significantly correlated with the BASC Conduct Problems, Aggression, and Oppositional

Behavior subscales. Lastly, the Conduct Problems factor was most highly correlated with the









BASC Aggression subscale and also had a significant correlation with the BASC Conduct

Problems subscale, as predicted. This factor also correlated significantly with the CPRS

Oppositional Behavior subscale but not with any of the other BASC or CPRS subscales.

Evidence for discriminant validity was also found as none of the three factors correlated

significantly with either the BASC or CPRS anxiety subscales.

As previous studies have only investigated specificity, sensitivity and other aspects of

diagnostic and categorizing abilities of a three-factor model of the ECBI, our study is the first to

assess the convergent and discriminant validity of derived factors. Overall, results provide

strong support for the multidimensional structure of the EBCI. While multiple factor analytic

studies now provide support for some variation on a three-factor solution, the present

investigation moves the field further ahead by not simply relying on the face validity of factor

label designations. These validity analyses give empirical support that allows for more confident

use of the Oppositional Defiant Behavior, Attention Difficulties, and Conduct Problems labels

applied to the three factors of the ECBI in our study and in previous research.

While our investigation adds to the literature and provides updated support for the three-

factor structure of the ECBI, it is not without limitations. The present study used a sample size

of 181, which meets minimum criteria of five cases per item for a factor analysis (Floyd &

Widaman, 1995). While meeting minimum requirements, larger samples are generally

considered better and some argue that a minimum of 200 or 10 cases per item is required (Floyd

& Widaman). Again sample size can also affect normality, which in turn can affect the clarity of

the model, possibly explaining why the current structure did not perfectly match with previous

findings. While EFA is more robust to such violations of normality than CFA, this aspect of the

data cannot be ignored.









Another limitation involves the presence of cross-loading items. In our study, two items

cross-loaded on another factor. These items were "Whines," which loaded primarily on the

Oppositional Defiant Behavior factor and secondarily on the Attention Problems factor, and

"Interrupts," which loaded primarily on the Attention Problems factor and secondarily on the

Oppositional Defiant Behavior factor. There are several reasons that may account for such

cross-loadings. First, the "Whines" item may not necessarily be associated with any of the

constructs assessed by the three factors; it is not representative of any of the specific DSM-IV

criteria for the behavioral disorders (APA, 1994). As the ECBI was designed before the DSM

included criteria for the disruptive behavior disorders, it does contain some items which, based

on current DSM-IV criteria and diagnostic categories, may not clearly associate with one

symptom category versus another. It was surprising that the "Interrupts" item cross-loaded as

this is a diagnostic criterion for ADHD, though it is grouped in the impulsivity category within

the DSM division of symptoms. This item though, cross-loaded on the oppositional behavior

factor. While interrupting may not be a specific symptom of ODD, it is an avenue through

which to annoy authority figures and purposely annoying adults is a specific symptom of ODD.

Again, sample size may have affected the clarity of results and use of a larger sample might

eradicate the presence of cross-loadings.

While the validity findings overall suggested that the three factors found in factor analysis

demonstrate good convergent and discriminant validity when compared to psychometrically

established measures of similar constructs, they were not perfectly in line with predictions. Both

the Oppositional Defiant Behavior and Conduct Problems factors, while correlating most highly

with their conceptual counterparts, also correlated significantly with the CPRS and BASC

subscales that related to CD and ODD, respectively. This limitation of the validity results could









be explained by the nature of the constructs the two factors represent. As stated previously, prior

discussions in the literature have focused on whether to consider the two disorders on a

continuum or view them as separate and distinct constructs (Loeber et al., 1991). The DSM-IV

itself notes that there is overlap between the two disorders (e.g. disobedience) even while

viewing them as two separate constructs (APA, 1994). Hence, it is not surprising that two

factors representing oppositional defiant behavior and conduct problems would correlate with

measures of both categories, due to high levels of comorbidity between the two disorders. Still,

the highest correlations with the more similar constructs of the CPRS and BASC highlight the

distinctness between the two factors.

A similar explanation exists for the fact that the attention difficulties factor correlated with

all of the BASC and CPRS subscales except for the anxiety subscales. ADHD is highly

comorbid with ODD and CD, and all three are categorized as disruptive behavior disorders. As

such, it would not be surprising to find correlations with constructs related to these disorders.

These correlations were slightly higher than expected in the current study, though still generally

smaller than the correlation with ADHD related subscales. In addition, many behaviors

associated with ADHD can be confused with symptoms of ODD. For instance, a child who is

not paying attention because of attention problems can be perceived as simply being disobedient.

Lack of attention may cause children to need instructions or requests repeated to them several

times. Items in the oppositional behavior factor of the ECBI represent these types of behaviors.

If caregivers associate lack of attention with these behaviors (e.g. needing to repeat instructions)

then a correlation between the Attention Difficulties factor and ODD-related subscales is

understandable.









Regardless of the limitations of the study, data presented here continue to support the idea

that the ECBI is best accounted for by a three-factor structure with validated factors.

Nevertheless, more research is still needed. While Burns and Patterson (2000) suggested cutoff

scores for the ECBI factors, the usefulness of these cutoff scores has only bee tested in the study

by Weis et al. (2005). Here, the investigators focused on the ability of the ECBI factors to

differentiate between specific behavior clusters in a restricted sample of young children. Given

these limited findings, these cutoff scores should be validated with a wider age range as well as

in both general clinical and more specific clinical samples similar to the one used in our study.

This would allow researchers to better ascertain the utility of the factors to differentiate between

children with different clinical problems (i.e. attention problems versus oppositional behavior,

disruptive behavior problems versus depression or anxiety). Although researchers have looked

at specificity and sensitivity of the ECBI factors (Weis et al.), the DBDRS used in these

discriminative validity analyses is a clinician rating scale which does not include information

often considered when deciding on a diagnosis (e.g. age of onset of symptoms, aspects of

impairment, teacher data) and is also based partially on parent report. This, as the researchers

note, introduces a degree of shared method variance between the ECBI and DBDRS ratings. The

ECBI factors could be best evaluated by assessing their ability to differentiate between children

who have been independently diagnosed with these various disorders. This would provide

support for the usage of the ECBI as a pre-screener to indicate areas which may require further

assessment.

Future research should include continued validation of the three-factor structure. Using a

larger specific clinical sample, with which multivariate normality assumptions can be met, a

CFA of the current results should be conducted. Again, these factors should be subjected to









convergent and discriminant validity analyses, sensitivity and specificity tests in the manner

described above, as well as further assessment of internal consistency. Test-retest reliability of

the factors should be ascertained along with the individual factors' sensitivity to treatment

effects. Once these analyses have been conducted, researchers and clinicians can begin to extract

even more information from the ECBI than they can currently obtain.









LIST OF REFERENCES


American Psychiatric Association (APA). (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.

Barkley, R.A. (1988). Aiieunl,-,l-JejyL 11iy'i i/ ii\'n disorder: A handbookfor diagnosis and
treatment (2nd ed.). New York: Guilford Press.

Barkley, R.A. (1997). Defiant Children. New York: Guilford Press.

Boggs, S.R., Eyberg, S., & Reynolds, L.A. (1990). Concurrent validity of the Eyberg Child
Behavior Inventory. Journal of Clinical Child Psychology, 19, 75-78.

Burns, G.L., & Patterson, D.R. (1991). Factor structure of the Eyberg Child Behavior Inventory:
One-dimensional or multidimensional measure of disruptive behavior? Journal of Clinical
Child Psychology, 20, 439-444.

Burns, G., & Patterson, D.R. (2000). Factor structure of the Eyberg Child Behavior Inventory: A
parent rating scale of oppositional defiant behavior towards adults, inattentive behavior,
and conduct problem behavior. Journal of Clinical Child Psychology, 29, 569-577.

Cattell, R.B. (1966). The scree test for the number of factors. Multivariate Behavioral Research,
1, 245-76.

Colvin, A., Eyberg, S.M., & Adams, C.D. (1999). Restandardization of the Eyberg Child
Behavior Inventory. Gainesville, Florida: University of Florida, Child Study Laboratory.

Conners, C.K. (1997). Conners' Teacher and Parent Rating Scales -Revised. Toronto:
MultiHealth Systems.

Eyberg, S.M. (1992). Parent and teacher behavior inventories for the assessment of conduct
problem behaviors in children. In L. VandeCreek, S. Knapp, & T.L. Jackson (Eds),
Innovations in clinical practice: A Source Book (Vol. 12, pp.377-382). Sarasota, FL:
Professional Resource Exchange.

Eyberg, S.M., & Robinson, E.A. (1983). Conduct problem behavior: Standardization of a
behavior rating scale with adolescents. Journal of Clinical Child Psychology, 11, 130 137.

Eyberg, S.M., & Ross, A.W. (1978). Assessment of child behavior problems: The validation of a
new inventory. Journal of Clinical Child Psychology, 7, 113-116.

Floyd, F.J., & Widaman, K.F. (1995). Factor analysis in the development and refinement of
clinical assessment instruments. Psychological Assessment, 7, 286-299.

Funderburk, B.W., Eyberg, S.M., Rich, B.A., & Behar, L. (2003). Further psychometric
evaluation of the Eyberg and Behar rating scales for parents and teachers of preschoolers.
Early Education and Development, 14, 67-81.









Henson, R. K. (2001). Understanding internal consistency reliability estimates: A conceptual
primer on coefficient alpha. Measurement and Evaluation in Counseling and Development,
34, 177-189.

Kaiser, H.F. (1960). The application of electronic computers to factor analysis. Educational and
Psychological Measurement, 20, 141-151.

Loeber, R., Lahey, B.B., & Thomas, C. (1991). Diagnostic conundrum of oppositional defiant
disorder and conduct disorder. Journal ofAbnormal Psychology, 100, 379-390.

Reynolds, C.R., & Kamphaus, K.W. (2004). Behavioral assessment system for children (2nd
ed.). Circle Pines, MN: American Guidance Service.

Robinson, E.A., Eyberg, S.M., & Ross, A.W. (1980). The standardization of an inventory of
child conduct problems. Journal of Clinical Child Psychology, 9, 22-29.

Tabachnick, B.G., & Fidell, L.S. (2007). Using multivariate statistics (5th Ed.). Boston, MA:
Pearson/Allyn & Bacon.

Weis, R., Lovejoy, M.C., & Lundahl, B.W. (2005). Factor structure and discriminative validity
of the Eyberg Child Behavior Inventory with young children. Journal ofPsychopathology
and Behavioral Assessment, 27, 269-278.

Werba, B.E. (2003). Standardization and cross-cultural validity of the Eyberg Child Behavior
Inventory with Australian preschoolers. (Doctoral dissertation, University of Florida,
2003). Dissertation Abstracts International, 63, 4391.

Zwick, W.R., & Velicer, W.F. (1986). A comparison of five rules for determining the number of
components to retain. Psychological Bulletin, 99, 432-442.









BIOGRAPHICAL SKETCH

Melissa Kay Stem was born on March 27, 1981 in Boston, Massachusetts. The youngest

of two, she grew up in the greater Boston area, graduating from Needham High School in 1999.

She earned her B.A. in psychology from Brandeis University. Upon graduating with high honors

in 2003, Melissa took a position as a research coordinator at the Massachusetts General Hospital

(MGH) and Freedom Trail Clinic in Boston, Massachusetts. She spent 15 months coordinating

research projects at the hospital's schizophrenia and severe mental illness outpatient clinic.

In September of 2004, Melissa left MGH for a position at the National Center for

Posttraumatic Stress Disorder (PTSD) at Boston University and the Boston Veteran's

Administration, where she was employed as a research technician for projects investigating the

psychophysiological and cognitive effects of nicotine on veterans with PTSD. Melissa relocated

to Gainesville, Florida in August of 2005 to begin her graduate career in clinical psychology.

Upon completion of the M.S., Melissa will continue on in her program and apply for doctoral

candidacy in the Department of Clinical and Health Psychology at the University of Florida.