Content and construct validities analysis of biology assessments instruments

• Autor(a) principal: Coppi, Marcelo Alves
• Data de Publicação: 2021
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Actio (Curitiba)
• Texto Completo: https://periodicos.utfpr.edu.br/actio/article/view/13715
• topic: Ensino de Ciências; Avaliação; Ensino Médio;; Validade de testes; Avaliação de testes; Instrumentos de avaliação; validade de conteúdo; Validade de construto; Test validity; Test evaluation; Assessment tools; Content Validity; Construct validity

The aim of this work was to analyze the validity of three classroom assessment instruments of the subject of biology a third grade of high school class in a private school in São Paulo, through an analysis of the content and construct validities. The content validity included the psychometric analysis of the parameters of difficulty, discrimination and random matching of the items, which were conceived and processed by the Item Response Theory (IRT). The construct validity was achieved through the verification of internal consistency, through the KR coefficient (α20), and the information function of the IRT. The results of the content validation showed that the items of the three classroom assessment instruments showed levels of discrimination between the moderate and high levels, considered as adequate, and levels of difficulty close to 50%, classified as appropriate for the latent skills of the students in the class analyzed. All items of the three assessment instruments were within the standard values of the random matching index. In the case of the construct validation, the three assessments were at the satisfactory level of construct validity and, therefore, in general, all assessments were considered valid.