Processo de construção e evidências de validade do subteste de aritmética do teste de desempenho escolar : segunda edição (TDE-II)

• Ano de defesa: 2016
• Autor(a) principal: Viapiana, Vanisa Fante
• Orientador(a): Stein, Lilian Milnitsky
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul
• Programa de Pós-Graduação: Programa de Pós-Graduação em Psicologia
• Departamento: Escola de Humanidades
• País: Brasil
• Palavras-chave em Português: TESTES PSICOLÓGICOS; DESEMPENHO ESCOLAR; ARITMÉTICA; PSICOMETRIA; PSICOLOGIA
• Área do conhecimento CNPq: CIENCIAS HUMANAS::PSICOLOGIA
• Link de acesso: http://tede2.pucrs.br/tede2/handle/tede/6759

This dissertation consists of two studies that aimed the development and validation of the Arithmetic Subtest of the School Achievement Test – Second Edition (Teste de Desemenho Escolar -Segunda Edição: TDE -II). The first study addresses the items construction process carried out through consultations to the National Textbook Program Guide (Programa Nacional de Livro Didático- PNLD) of the brazilian government and the mathematics area experts. After the preparation of the items, a draft of the subtest that has been tested empirically was formulated. Three hundred and two students in grades 1st through 9th of elementary school of public and private schools in the metropolitan region of Porto Alegre/RS participated in the first stage of this study. The data were submitted to factor analysis, and showed the prevalence of two factors explaining 74% of the variance of the data. Both dimensions shown to be related to the complexity and difficulty level of the items. The subtest was therefore divided into two versions: one for students in grades 1 through 5 (Version A), and another, for those in grades 6 through 9 (Version B). Each of the versions was analyzed by Item Response Theory (IRT) models, based on item response theory (IRT) models, which suggested that the items provided a comprehensive measure of the latent trait. Later, in the empirical analysis, we selected the items that were included in the final version of the subtest of Arithmetic of TDE-II. The second study tested evidence of validity, type convergent of the final version of the instrument. The sample included one hundred and eleven students from public and private schools in the metropolitan region of Porto Alegre, RS. The results showed satisfactory correlational patterns of the Arithmetic Subtest with Subtests Reading and Writing of TDE -II, with the arithmetic subtest of the WISC -IV and with other neurocognitive assessment tasks. Performance in Version A of the Arithmetic Subtest was related to executive demands. And performance in Version B suggests a greater relationship with a specific semantic memory of mathematical procedures. We concluded that the Arithmetic Subtest of TDE-II features good psychometric properties for the assessment of arithmetic school performance over elementary school.