Predictive Validity of the LSAT: A National Summary of the 1997-1998 Correlation Studies (TR-99-01)
by Lisa C. Anthony, Jennifer R. Duffy, and Lynda M. Reese
Since the Law School Admission Test (LSAT) was first administered, the sponsors of the test have carried out predictive validity studies to evaluate the effectiveness of the test as well as other predictors in determining first-year law school performance. Over the years, increasingly large numbers of law schools have participated in these studies, commonly called correlation studies.
This report presents a summary of correlation study results for the 1997 and 1998 study years. This summary can serve as documentation supporting the generalizability of the claim of the validity of the LSAT for use in the law school admission process. More importantly, this study provides national longitudinal data for law schools to examine against their school-specific results to increase understanding of their own admission process. Since correlation studies are conducted for individual schools and school-specific results are reported only to the school whose data were analyzed, the results reported in this study may be used by schools as a benchmark in evaluating their own results.
The effectiveness of the LSAT score alone, undergraduate grade-point average (UGPA) alone, and the combination of these two variables for predicting first-year law school average (FYA) is evaluated. As previous studies have shown, results reported here indicate that the LSAT alone tends to be a better predictor of law school performance than the UGPA alone. The combination of LSAT and UGPA, however, continues to be superior to either predictor variable alone for predicting law school first-year average. These results support the validity of the LSAT for use in the law school admission process.