Demonstrating the validity of three general scores of PET in predicting higher education achievement in Israel

Background: The Psychometric Entrance Test (PET), used for admission to higher education in Israel together with the Matriculation (Bagrut), had in the past one general (total) score in which the weights for its domains: Verbal, Quantitative and English, were 2:2:1, respectively. In 2011, two additional total scores were introduced, with different weights for the Verbal and the Quantitative domains. This study compares the predictive validity of the three general scores of PET, and demonstrates validity in terms of utility. Method: Sample: 100,863 freshmen students of all Israeli universities over the classes of 2005-2009. Regression weights and correlations of the predictors with FYGPA were computed. Simulations based on these results supplied the utility estimates. Results: On average, PET is slightly more predictive than the Bagrut; using them both yields a better tool than either of them alone. Assigning differential weights to the components in the respective schools further improves the validity. Conclusion: The introduction of the new general scores of PET is validated by gathering and analyzing evidence based on relations of test scores to other variables. The utility of using the test can be demonstrated in ways different from correlations.