Response process validity in the Cloze test

• Autor(a) principal: Cunha, Neide de Brito
• Data de Publicação: 2009
• Outros Autores: Santos, Acácia Aparecida Angeli dos
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Fractal : Revista de Psicologia
• Texto Completo: https://periodicos.uff.br/fractal/article/view/4768
• topic: psychological assessment; Cloze test; reading comprehension; psychometrics; response process

Based in the response process validity evidence, this study explored quality of children's mistakes on the Cloze test. The sample was composed by 266 children, boys and girls, aging from 8 to 13 years old, attending at the third and fourth grades of the elementary education of public and private schools of São Paulo, Brazil. Two protocols groups were formed: of the children with high and low means to construct a errors scale, for the least to most serious: blank, phonological, lexical, syntactic and semantic. The results showed that children with high means test presented lexical mistakes, while children with low means made semantic mistakes. Although the response process validity evidence was finding, by the homogeneously evaluation on the errors distribution types, it's necessary to extend the knowledge of psychometric characteristics of the Cloze test.