Sharing Our Predictive Validity Studies for 2020-2024
Law schools use the LSAT as an important component in their holistic admission process, because the LSAT provides important information about applicants’ skills and preparedness for the first year of law school.
Every year, the majority of law schools participate in correlation studies that LSAC conducts free-of-charge for each law school to evaluate the predictive validity of the LSAT in their admission process. The LSAT correlation studies evaluate the effectiveness of LSAT scores, undergraduate grade point average (UGPA), and the combination of LSAT score and UGPA in predicting a student's first year average in law school. We also aggregate the individual school results to provide a comprehensive summary of the LSAT’s predictive validity.
Today, we are releasing a research report covering the last five years of our predictive validity studies. This report provides summaries of the combined results for the law schools participating in the 2020, 2021, 2022, 2023, and 2024 LSAT Correlation Studies. Because the correlation studies look at first-year law school performance, these studies correspond to the incoming law school classes of 2019, 2020, 2021, 2022, and 2023; the correlation study for the incoming class of 2024 will be completed later this year. Recognizing that not all test takers applied to law school or matriculate in the same cycle in which they take the LSAT, these correlation studies represent a span of LSAT test takers from about 2015 to 2023.
Last five years confirm historic patterns of LSAT’s predictive validity
These studies document and support the predictive validity of LSAT scores for use in the law school admission process. The findings across all five correlation studies are consistent with all prior reports in showing that LSAT scores are the best single predictor of law school performance, and significantly better at predicting law school performance than UGPA alone. The findings also confirm that the combination of LSAT scores and UGPA continues to be the best predictor of first-year law school performance. The research has also shown that for the majority of schools, although not all, average LSAT score is a slightly better predictor of first-year performance than highest LSAT score.
Looking at the trends for predictive validity
Another interesting trend is how LSAT scores and UGPAs have changed in recent years. Both the mean LSAT scores and the mean UGPAs of matriculants have increased over the past 4 years. For LSAT scores, however, the standard deviation of matriculant LSAT scores has increased over time, meaning that the distribution of scores has expanded at the same time that the mean score has increased. By contrast, the standard deviation of matriculant UGPAs has decreased over the past 4 years, likely caused by a ceiling effect where a significant number of individuals are at the highest point. This ceiling effect limits one's ability to distinguish between individuals because there are so many individuals with very high UGPAs.
There is a lot of rich data contained in this report, so we encourage you to dig into any areas of particular interest.
Anna Topczewski is the director of assessment sciences at LSAC. Anna has been in the assessment field for more than 12 years and at LSAC for more than four years.