The Predictive Validity of the Star: A Need for Local Validation

Abstract

The Screening Test of Academic Readiness (STAR) was administered to 59 incoming kindergarten boys at a private school with superior academic standards. To assess the predictive validity of the STAR, the Metropolitan Readiness Test (MRT) and teachers' academic ranking of the students were obtained as criterion measures toward the end of the kindergarten year. Except for a low but significant correlation (.24) between the STAR and the MRT Numbers subtest, there was no substantial relationship between the STAR and MRT scores. The Spearman rank-difference correlation between the STAR and teachers' ranking of the kindergarten (.23) was statistically significant but low. The findings suggested that the STAR may not be an effective screening test for kindergarten pupils in a population of children with superior IQs. The results call for local validation of tests to be used for admissions purposes.