Abstract
Background: The purpose of this work was to collect construct and criterion-related evidence of validity for a reading comprehension test (TCL - Teste de Compreensão da Leitura) with three vertically scaled forms, designed to assess students from second, third and fourth grade. Method: Two studies were conducted. In the first (n = 1,229), a confirmatory factor analysis was performed to analyse the test dimensionality. In the second (n= 402), concurrent and predictive evidence of validity was analysed using correlations between TCL, other reading tests and academic achievement. Results: Confirmatory factor analysis results supported a one-factor structure. Correlation coefficients with other reading tests were low to moderate and statistically significant. The TCL forms were shown to be good predictors of students' reading comprehension as assessed by teachers and of the National Exams of Portuguese Language results. Conclusions: Present results provide empirical evidence for the validity of the TCL forms.
