In the theory of measurement, there are two competing measurement frameworks, classical test theory and item response theory. The present study empirically examined, using large scale norm-referenced data, how the item and person statistics behaved under the two competing measurement frameworks. The study focused on two central themes: (1) How comparable are the item and person statistics derived from the item response and classical test framework? (2) How invariant are the item statistics from each measurement framework across examinee samples? The findings indicate that, in a variety of conditions, the two measurement frameworks produce similar item and person statistics. Furthermore, although proponents of item response theory have centered their arguments for its use on the property of invariance, classical test theory statistics, for this sample, are just as invariant.

In the theory of measurement, there are two competing measurement frameworks, classical test theory and item response theory. The present study empirically examined, using large scale norm-referenced data, how the item and person statistics behaved under the two competing measurement frameworks. The study focused on two central themes: (1) How comparable are the item and person statistics derived from the item response and classical test framework? (2) How invariant are the item statistics from each measurement framework across examinee samples? The findings indicate that, in a variety of conditions, the two measurement frameworks produce similar item and person statistics. Furthermore, although proponents of item response theory have centered their arguments for its use on the property of invariance, classical test theory statistics, for this sample, are just as invariant.

Submitted by Viviane Lima da Cunha (viviane@biblioteca.ufpb.br) on 2017-06-22T14:15:06Z No. of bitstreams: 1 arquivototal.pdf: 3580242 bytes, checksum: 63c3a2cbda0197a7117739490979581d (MD5) / Made available in DSpace on 2017-06-22T14:15:06Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 3580242 bytes, checksum: 63c3a2cbda0197a7117739490979581d (MD5) Previous issue date: 2016-02-19 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Usually, the people are well understood by somebody else, considering adjectives who describe their behaviour. This description or characterization is consider a singularity of these personality under your pairs. Some theoretical share the idea that traits are fundamental units of personality, presenting a high range of dispositions for answer the questions in certain ways. Therefore, the personality is a construct which strong influence under the behaviours, with influence in behaviours associate to healthy too. The Five Big Factors Associated with the Personality is considered a model well understood and with great replicability in many countries, who subsidized the formulation of instruments for personality rating. Taking this big concept for make a psicological evaluation, is necessary to use instruments that are not too long and whose items are easily understand by the subjects evaluated. Considering this needs and the importance of personality evaluation, the goal of this study was development and get evidences valid for a construct of a instrument for personality evaluation. Comparisons were made between two types of scale, one based on the Likert method and the other on the semantic differential method and another in differential semantic method. The research was count with a living togheter sample of 247 people in the Pilot Study and 571 attendees in the Final Study, of the both sexes and college students. The scales passed through semantic analysis with 10 subjects of lower extract of the population. It was also carried out the analysis of judges with three experts. In the Final Study were held 7 Factor Analyses (AF) for each scale, was also calculated Cronbach's alpha for each facet and dimensions in order to verify the reliability of the instruments. From the found psychometric results, it opted for the Semantic Differential scale to perform the analyzes of the items by the Item Response Theory. Therefore, the Semantic Differential Scale presented construct evidence of construct validity of as well as evidence of accuracy, be appropriate for personality evaluation. / Normalmente os indivíduos são caracterizados por outrem, levando em consideração adjetivos que descrevam seu comportamento. Essa descrição ou caracterização é considerada uma particularidade da personalidade daquele sujeito perante seus pares. Alguns teóricos compartilham a ideia de que os traços são unidades fundamentais da personalidade, representando disposições amplas para responder de determinadas maneiras. Dessa forma, a personalidade é um construto que tem forte influência sobre os comportamentos, influenciando também comportamentos em saúde. O modelo dos Cinco Grandes Fatores de Personalidade tem subsidiado a elaboração de instrumentos de avaliação da personalidade por ser considerado um modelo compreensivo de considerável replicabilidade em vários países. Para realizar uma avaliação psicológica levando em consideração este amplo conceito, faz-se necessária a utilização de instrumentos que não sejam demasiadamente longos, bem como instrumentos cujos itens sejam de fácil compreensão por parte dos sujeitos avaliados. Dadas as demandas e a relevância da avaliação da personalidade, o objetivo deste estudo foi desenvolver e obter evidências de validade de construto de um instrumento para a avaliação da personalidade. Foram realizadas comparações entre dois tipos de escalas de resposta, sendo uma baseada no método Likert e a outra no método do diferencial semântico. A pesquisa contou com uma amostra de conveniência de 247 sujeitos no Estudo Piloto e 571 participantes no Estudo Final, sendo eles sujeitos de ambos os sexos e universitários. Foi realizada a análise semântica das versões iniciais dos instrumentos com 10 sujeitos do extrato educacional mais baixo da população alvo. Também foi realizada a análise de juízes com três especialistas. O Estudo Piloto resultou na exclusão de itens a partir da análise do poder discriminativo dos itens. No Estudo Final foram realizadas 7 Análises Fatoriais (AF) para cada escala; também foi calculado o alfa de Cronbach para cada faceta e as dimensões com o objetivo de verificar a fidedignidade dos instrumentos. A partir dos resultados psicométricos encontrados, optou-se pela Escala de Diferencial Semântico para realizar as análises dos itens por meio da Teoria de Resposta ao Item. Dessa forma a Escala de Diferencial Semântico apresentou evidências de validade de construto, bem como evidências de precisão, sendo a mesma adequada para avaliação da personalidade.