The main purpose of this study was to examine the prognostic validity of the School Readiness Test, which was constructed by Vlahović-Štetić et al. (1995). The subjects were students, of the fifth and sixth grade. Since the success of children is largely determined by academic self-efficacy, it was included as a covariate. Self-efficacy was measured with the Morgan-Jinks Student Efficacy Scale (MJSES), which was translated from English for the purpose of this study. A series of linear regression procedures (forward method) were conducted. By excluding self-efficacy from the model, it is possible on the basis of individual subtests of the School Readiness Test, to significantly predict grades in certain subjects in the fifth and sixth grade. The Numerical test proved to have the highest predictive value which predicts grades in Croatian language, mathematics, science and geography. The results are discussed in the framework of cognitive theory.

School Readiness Test, predictive validity