Thesis (Ph. D.)--Georgia State University, 2007. / Title from file title page. Carolyn F. Furlow , committee chair; Phillip Gagne, T. Chris Oshima, Valerie A. Miller, committee members. Electronic text (113 p. : ill.) : digital, PDF file. Description based on contents viewed June 24, 2008. Includes bibliographical references (p. 93-100).

Thesis (Ph. D.)--Michigan State University. Measurement and Quantitative Methods, 2008. / Title from PDF t.p. (Proquest, viewed on Aug. 20, 2009) Includes bibliographical references (p. 96-101). Also issued in print.

Thesis (Ph. D.)--University of North Texas, Dec., 2008. / Title from title page display. Includes bibliographical references.

Thesis (Ph.D.)--University of Nebraska-Lincoln, 2008. / Title from title screen (site viewed Nov. 25, 2008). PDF text:vii, 125 p. : ill. ; 29 Mb. UMI publication number: AAT 3317288. Includes bibliographical references. Also available in microfilm and microfiche formats.

Thesis advisor: Walter Vispoel, Won-Chan Lee. Includes bibliographic references (p. 204-209).

Thesis (Ph. D.)--Ohio State University, 2007. / Title from first page of PDF file. Includes bibliographical references (p. 116-120).

abstract: Item response theory (IRT) and related latent variable models represent modern psychometric theory, the successor to classical test theory in psychological assessment. While IRT has become prevalent in the assessment of ability and achievement, it has not been widely embraced by clinical psychologists. This appears due, in part, to psychometrists' use of unidimensional models despite evidence that psychiatric disorders are inherently multidimensional. The construct validity of unidimensional and multidimensional latent variable models was compared to evaluate the utility of modern psychometric theory in clinical assessment. Archival data consisting of 688 outpatients' presenting concerns, psychiatric diagnoses, and item level responses to the Brief Symptom Inventory (BSI) were extracted from files at a university mental health clinic. Confirmatory factor analyses revealed that models with oblique factors and/or item cross-loadings better represented the internal structure of the BSI in comparison to a strictly unidimensional model. The models were generally equivalent in their ability to account for variance in criterion-related validity variables; however, bifactor models demonstrated superior validity in differentiating between mood and anxiety disorder diagnoses. Multidimensional IRT analyses showed that the orthogonal bifactor model partitioned distinct, clinically relevant sources of item variance. Similar results were also achieved through multivariate prediction with an oblique simple structure model. Receiver operating characteristic curves confirmed improved sensitivity and specificity through multidimensional models of psychopathology. Clinical researchers are encouraged to consider these and other comprehensive models of psychological distress. / Dissertation/Thesis / Ph.D. Psychology 2011

Dissertação (mestrado)—Universidade de Brasília, Instituto de Psicologia, Departamento de Psicologia Social e do Trabalho, Programa de Pós-Graduação em Psicologia Social, do Trabalho e das Organizações, 2002. / Submitted by Tania Milca Carvalho Malheiros (tania@bce.unb.br) on 2009-08-12T15:10:21Z No. of bitstreams: 1 2002_Frederico_Neves_Condé_Dissertação.pdf: 757326 bytes, checksum: d2e4997cc97c4e45cf7a0d658e7f00f5 (MD5) / Approved for entry into archive by Guimaraes Jacqueline(jacqueline.guimaraes@bce.unb.br) on 2009-08-12T16:10:44Z (GMT) No. of bitstreams: 1 2002_Frederico_Neves_Condé_Dissertação.pdf: 757326 bytes, checksum: d2e4997cc97c4e45cf7a0d658e7f00f5 (MD5) / Made available in DSpace on 2009-08-12T16:10:44Z (GMT). No. of bitstreams: 1 2002_Frederico_Neves_Condé_Dissertação.pdf: 757326 bytes, checksum: d2e4997cc97c4e45cf7a0d658e7f00f5 (MD5) Previous issue date: 2002-10-18 / A Teoria de Resposta ao Item (TRI) assume a existência da propriedade de invariância dos parâmetros, que permite estimar a habilidade dos sujeitos (theta) independentemente da forma do teste utilizado. Esta propriedade se baseia em pelo menos duas condições relacionadas aos itens do teste: estar na mesma escala e atender ao pressuposto de unidimensionalidade. O objetivo do presente estudo é o de investigar se a estimativa de theta independe da dificuldade dos itens utilizados para estimá-la bem como verificar em que medida a unidimensionalidade da prova influencia nesta propriedade. Foram utilizados os dados secundários de 26 formas de prova de Matemática de 8a Série do Ensino Fundamental (E.F.) do Sistema Nacional de Avaliação da Educação Básica (Saeb), aplicada em 1997, em uma amostra de 18.806 estudantes brasileiros de escolas públicas e particulares de cada uma das 27 Unidades da Federação brasileiras. Essas formas de prova foram respondidas por 26 grupos diferentes de estudantes, equivalentes em termos de habilidades. Foram correlacionados os resultados médios de índices de dificuldade das provas e habilidade dos estudantes calculados pela Teoria Clássica dos Testes (TCT) e pela TRI. Os resultados apontam para a existência de uma dependência do theta em relação à dificuldade dos cadernos (r = 0,68, com o valor p; r = - 0,69 com o parâmetro b), menor que a do índice de habilidade calculado pela TCT, o escore total, em relação à dificuldade (r = - 0,95, com o parâmetro b). A dependência entre o theta e a dificuldade diminui quando são excluídos da prova os itens com cargas fatoriais inferiores a 0,20 no fator único, que praticamente não contribuem para a unidimensionalidade. Observou-se, neste caso, um coeficiente de correlação com o valor p de 0,60 e, com o parâmetro b, de – 0,57. Conclui-se que o theta estimado depende da dificuldade dos itens que são utilizados para estimá-lo, não confirmando a propriedade de invariância dos parâmetros. O theta apresenta uma dependência menor com a dificuldade, que a observada pelo escore total. Por sua vez esta estimativa da TRI apresenta uma diminuição da dependência com relação à dificuldade quando a prova se aproxima da unidimensionalidade. Os resultados indicam ser vantajosa a utilização da TRI para estimar a habilidade dos estudantes, quando são utilizadas formas de teste com dificuldades diferentes, pois o theta é menos dependente da dificuldade que o escore total. Faz-se necessário um maior rigor no controle da condição de unidimensionalidade da prova para a obtenção de estimativas de theta mais invariantes. _______________________________________________________________________________ ABSTRACT / The Item Response Theory (IRT) assumes the existence of a property of invariance of the parameters, which implies that the parameter that characterizes the ability of an examinee (theta) does not depend on the set of items of the test forms. This property is based on at least two conditions related to the test items: being in a common scale and being in accordance with the assumption of unidimensionality. The objective of the present study is to investigate whether the estimation of theta does not depend on the difficulty of the items used to estimate it, as well as how much the unidimensionality of the test items influences this property. In this study, secundary data were used of 26 test forms of Mathematics, eight grade, from the National System of Evaluation of the Basic Education (Saeb), applied in 1997, in a sample of 18.806 Brazilian students of public and particular schools of the 27 Brazilian states. The test forms were answered by 26 different groups of students, equivalent in terms of abilities. The results of indexes of the tests' difficulty and the students' ability were calculated and correlated using both the Classic Theory of Tests (CTT) and the IRT. The results point to the existence of a dependence of the theta in relation to the difficulty of the booklets (r = 0,68, for the p value; r = - 0,69 for the parameter b), smaller than the one of the ability index calculated by CTT, the total score, in relation to the difficulty (r = - 0,95, with the parameter b). The dependence between theta and the difficulty decreases when items with factor loadings less than 0,20 in the only factor of the test, that practically don't contribute to the unidimensionality, are excluded. A correlation with the p value of 0,60 and, with the parameter b, of - 0,57, was observed, in this case. It was concluded that the estimated of theta is related to the difficulty of the items, not confirming the property of parameter invariance. This relation is less strong than the one between the total score and the difficulty of the items. The estimated of theta presents a dependence that decreases with the difficulty when the test approaches the unidimensionality. The use of IRT is considered advantageous to esteem the students' ability, when test forms present different difficulties, theta is less dependent on the difficulty than the total score. It is considered necessary a greater rigidity on the control of the conditions for the unidimensionality of the tests in order to obtain more invariant estimates of theta.

The purpose of this study was to develop a core HIV/AIDS knowledge assessment (CHAKA) for students enrolled in counseling-related degree programs. Although there are studies that examined counseling HIV/AIDS knowledge, the instruments that were used were limited in ways that may compromise the accuracy of the inferences that were made. This study was carried out in three phases. Phase 1 involved developing an initial pool of items; Phase 2 involved an expert review for content validation as well as a pilot-test; Phase 3 involved field testing the CHAKA. The field-test involved 343 undergraduate and graduate students at Southern Illinois University. Item response theory (IRT) was used to analyze the data. Before the data were analyzed, they were examined to see if the CHAKA was a unidimensional test. Results of the factor analysis performed was that the CHAKA may not be unidimensional; however the internal consistency was decent (α= .734). A two-parameter logistic (2PL) model was fit to the data. Results from the item parameter estimates displayed relatively low discrimination and difficulty parameters in addition to some problematic items (i.e., negative discrimination estimates, unusually large difficulty values). Additional analyses revealed that locally dependent items may have accounted for the possible multidimensionality, low discrimination indices, and inflated difficulty values. The low discrimination values likely affected the information values of the items and the test. All item information values were less than 1. Last, both uniform and non-uniform differential item functioning (DIF) was present between undergraduate and graduate students. IRT appears to be a promising approach to instrument development in counseling-related programs. Although the CHAKA properties were not ideal, further revisions and a larger sample size may contribute to the overall improvement of this instrument.

BOSI,Marcos Antonio.Um Estudo Sobre O Grau de Maturidade e a Evolução da Gestão Pela Qualidade Total no Setor deTransformação Cearense por Meio da Teoria da Resposta ao Item.2010.137f.Dissertação(Mestre em Gestão Logística)-Coordenação do Curso de Pós-Graduação em Logística e Pesquisa Operacional,Universidade Federal do Ceará,Fortaleza,2010. / Submitted by Winne Gomes da Silva (winnegomez87@gmail.com) on 2012-06-15T12:21:22Z No. of bitstreams: 1 2010_dis_mabosi.pdf: 4362190 bytes, checksum: 04328d1ba720329c8d37c10600612c80 (MD5) / Approved for entry into archive by Nirlange Queiroz(nirlange@gmail.com) on 2012-06-20T11:55:14Z (GMT) No. of bitstreams: 1 2010_dis_mabosi.pdf: 4362190 bytes, checksum: 04328d1ba720329c8d37c10600612c80 (MD5) / Made available in DSpace on 2012-06-20T11:55:14Z (GMT). No. of bitstreams: 1 2010_dis_mabosi.pdf: 4362190 bytes, checksum: 04328d1ba720329c8d37c10600612c80 (MD5) Previous issue date: 2010 / The objective of this text is to apply the Item Response Theory (IRT) to evaluate the maturity degree of the manufacturing industry of medium and large scale based in the Ceará State, Northeastern Brazil, on the use of Total Quality Management (TQM) and its evolution over the past decade, based on comparative study carried out by Alexander (1999). The methodology for data collection was based on the completion of a quantitative survey covering this universe, through a structured questionnaire based on Likert scale of five categories. Based on a maturity scale constructed from the parameters of the IRT, it was possible to observe and compare the two moments and conclude that the evolution of the maturity degrees on that period focused on the industrial park was a kind of shy, especially if we consider the intensification of market competition in recent decades, especially in the last ten years with the expansion of globalization and quality as a competitive top to stay on the market. / O objetivo deste trabalho é aplicar a Teoria da Resposta ao Item (TRI) na avaliação do grau de maturidade da indústria de transformação e de construção civil de portes médio e grande sediadas no Estado do Ceará, quanto à utilização da Gestão pela Qualidade Total (GQT) e a sua evolução nos últimos dez anos, tendo como base comparativa estudo realizado por Alexandre (1999). A metodologia de pesquisa para a coleta de dados teve por base a realização de uma pesquisa quantitativa abrangendo esse universo, através de um questionário estruturado com a utilização da escala de Likert de 5 categorias. Através de uma escala de maturidade construída neste trabalho a partir dos parâmetros da TRI, foi possível constatar e comparar os dois momentos e concluir que a evolução do grau de maturidade desse parque industrial no período enfocado se mostrou tímida, principalmente se for considerado o acirramento da concorrência de mercado nas últimas décadas, notadamente nos últimos dez anos com a expansão da globalização, e a qualidade como um diferencial competitivo superior para se manterem no mercado.