All Validities are Not Created Equal: Determinants of Variation in SAT Validity Across Schools

Previous research has demonstrated that cognitive test validities are generalizable and predictive of academic performance across situations. However, even after accounting for statistical artifacts (e.g., sampling error, range restriction, criterion reliability), substantial variability often remains around estimates of cognitive test-performance relationships suggesting the presence of additional moderators. In the present study, we examine the sources of institutional variation in Scholastic Aptitude Test (SAT) validity across a sample of 110 institutions. Institutional characteristics moderated the size of SAT validities, such that more selective schools and schools that emphasize traditional assessment techniques (i.e., school records, standardized tests) showed higher SAT validities while schools that were larger and where students demonstrated more financial need, schools that emphasized the usage of alternative assessment techniques (i.e., essays, letters of recommendations, extracurricular activities), and schools that enrolled higher percentages of historically disadvantaged minority students generally exhibited lower SAT validities. Future directions in the understanding of situational influences on SAT-grade point average validities are discussed.