Tuesday, December 29, 2015
The recent issue of preLaw magazine offers a positive justification for admissions departments relying on that test. It is familiar, and you can read it here. More interesting is the comment that Professor Ralph Brill posted at the end of the article. I quote:
The analysis is incomplete. 1. The predictive value of the LSAT (med. .36) is not "success" in the first year of law school, unless the author only means "you won't flunk out." It has no predictive value on your GPA at the end of the first year; it has no predictive value on whether you will ultimately graduate; it has no predictive value on whether you will finish in the top 10 or 15% of your class. 2. Most schools combine the UGPA and the LSAT (LSAC does it for you) to increase the predictive value for getting past the first year, but it isn't the total of the separate predictive values of each -- there is a separate combined predictive value, usually something like .46. So, I have seen students with 170 LSATs flunk out or finish very low in their classes, and students with 143 LSATs top the graduating class. It is not a great predictor, but, with the UGPA, it is just the best one we have. Interviews, age or applicants, job experience, college major, colleges from which applicants graduated, all can add value to decisions, but obviously involve much more work for admissions officers.