Search the Lounge


« Losing an Academic Career to ME/CFS | Main | Are Law Schools Exploiting Part-Time Students? »

November 29, 2017


Feed You can follow this conversation by subscribing to the comment feed for this post.

Ralph Clifford

A one- or two-point change on the LSAT percentile of a school is not statistically significant. As the error band reported by the LSAC is 2.6 points, your statistical argument for both Albany and NYLS doing more poorly or St. Johns doing better is not sustainable.

David Frakt

I beg to differ Ralph. A one or two point difference for an individual may not be statistically significant, because some students will be slightly better and some slightly stronger than their score. But when the whole class (or the bottom half of the class, where most of the failures come from) is weaker or stronger (and not just on LSATs, but also UGPA), then lower or higher performance can be predicted. Admission credentials may not account for all of the difference in bar performance, but it is certainly a significant factor.


I have not been following your posts regularly but you seem to assume there is a 100% correlation between LSAT score and bar passage. As I am sure you are aware, LSAT scores are designed to predict first-year grades not bar passage, and if you were right, it might just be possible to skip the whole law school thing. My sense is law school grades are quite a good predictor of bar passage, but the correlation between LSAT and law school grades (without factoring in UGPA) is not particularly strong. I share the concern that law schools are broadening their admission standards primarily out of financial rather than academic interests, but I just wanted to suggest you avoid overstating the connection between LSATS and bar passage.

David Frakt

I would encourage you to go back and read all my posts, as I have addressed this topic several times in previous posts. Your "sense" is incorrect. The correlation between LSAT and bar passage is very strong for law school classes as a whole, and even stronger with UGPA factored in, which is why I look to see if schools are balancing lower LSAT scores with higher UGPAs. (They seldom do.) Although first year grades are a better predictor of bar passage rates than LSAT scores, LSAT scores are by far the best predictor of bar passage that are available to law schools at the point of reviewing applications for admission.

Although the report is somewhat dated (1998), the LSAC National Longitudinal Bar Pass Study has some very useful information about the correlation between bar passage and LSAT scores. You can find the study here:

See also this more recent article from the NCBE which discusses the correlation of LSAT scores and MBE scores., and includes a table with average MBE score by LSAT score.

I would also recommend Rosin, Gary S., Unpacking the Bar: Of Cut Scores, Competence and Crucibles. 1st Annual Conference on Empirical Legal Studies Paper; Journal of the Legal Profession, Vol. 32, 2008. Available at SSRN:

For an analysis of eventual bar pass rates by LSAT score, see here:

The comments to this entry are closed.


  • StatCounter
Blog powered by Typepad