THE LAW SCHOOL ADMISSION TEST AS A PREDICTOR OF LAW SCHOOL GRADES*

Abstract

SUMMARY AND CONCLUSIONSThe effectiveness of the Law School Admission Test (LSAT) as a predictor of first‐year law school grades grades was evaluated in a series of studies initiated in the summer of 1949. This report is based on the findings of studies is eighteen law schools, including twelve schools which provided data on some measure of the previous academic success of their students. The eighteen law schools on which this report is based include all schools now represented on the Policy Committee of the Law School Admission Test which had test scores on 25 or more students.The principal finding of this study is that Law School Admission Test scores are distinctly useful alone or preferably as supplementary to evidence of previous academic success in predicting first‐year grades of law students. This conclusion, is based on the fact than the predictive effectiveness of the combined predictors is represented by a correlation coefficient of .52, as compared with a value of .40 for the test alone and of .38 for previous academic success alone. These figures summarize results based on 1725 day students in 12 law schools. In ten of those twelve schools the test score equalled or excelled the college record as a predictor of first‐year lay grades.The Law School Admission Teal had a validity coefficient of .44 based on the pooled results from 18 groups of day students and three groups of evening students. A coefficient of this size indicates definite usefulness in prediction of graces for the best used alone. It would reasonably be expected that the coefficients obtained would have been higher if the students had net been selected, at least in part, on the basis of test scores.In a number of law schools, attention was given to the possibility that the number of years of pre‐law college work might be considered an additional determiner in predicting success. It was found, however, that no appreciable gains in predicting success could be obtained by using this information. Similarly, it was found that no appreciable gain in prediction could be obtained by taking into account whether or not the student had received his undergraduate education at the same university in which he was attending law school.