Predictive Validity of the LSAT: A National Summary of the 1990-1992 Correlation Studies (RR-93-05)
by Linda F. Wightman

Executive Summary

For approximately 45 years, the sponsors of the Law School Admission Test (LSAT) have offered to conduct studies of the effectiveness of the test as well as other predictors of law school performance used in the admission process. These studies are offered free of charge to participating law schools, and schools have been encouraged to avail themselves of the service.

Over the years, increasingly large numbers of law schools have participated in the LSAT correlation studies. During the three-year period from 1990 through 1992, 183 different schools participated in the studies and 527 correlation studies were conducted. Among the 183 schools, 167 schools participated for each of the three years-1990, 1991, and 1992. The summary data presented in this report are only from those 167 schools. By examining data based on the same set of schools over multiple years, trends can be examined and meaningfully interpreted.

One purpose of this study is to summarize data across schools to provide documentation of the generalizability of the claim of validity of the LSAT for use in the admission process. A more important purpose is to provide national longitudinal data for law schools to examine against their school-specific data to help them increase their understanding of their own admission process. Correlation studies are conducted for individual schools and school-specific results are reported exclusively to the schools whose data were analyzed. Thus, schools know how well the test and other predictors are performing within their own admission process, but they have no benchmark against which to evaluate their results.

The correlation studies provide valuable information to LSAT score users. One task frequently assigned to those responsible for law school admission is that of identifying from large groups of law school applicants, those who are most likely to succeed in law school. A limited amount of information usually is available from which to make that decision. Almost universally across all ABA-accredited law schools, and English language common-law law schools in Canada, both LSAT score and undergraduate grade point average are among the available data. Both are quantifiable measures that are potentially useful in making admissions decisions and many schools use this information extensively. If this (or any other) quantifiable information is relied on in the selection process, the burden is 00 the score user to obtain evidence that there is a relationship between the quantified variables and the outcome of interest to the admission committee—usually success in law school. The correlation studies can provide that evidence for participating schools. An additional value of the correlation studies is that they provide score users with quantifiable information about how their admission process is working, and about the make-up of their entering class.

Predictive Validity of the LSAT: A National Summary of the 1990-1992 Correlation Studies (RR-93-05)

Research Report Index