I just took an actual LSAT from February 1993 and scored a 167 (raw score 82). How accurate of a predictor is this test? I have a book of ten LSAT’s from the early 1990’s (92-95). How much has the test changed? Will the tests from 1995 be more accurate to today’s test? I also noticed in the book raw scores for different test dates equal different 120-180 LSAT scores. Why is this? Change in conversion does not seem to be directly related to test date.
Predictor of what? Success in law school? In my experience (and in speaking to a friend is an undergraduate admissions officer), standardized tests can be generally predictive of success in school if you work roughly as hard as an average student. If you don’t, like I didn’t as an undergraduate, you’ll do a lot worse than your peers with similar scores.
–Cliffy
I looked up the entry for LSAT in the Mental Measurement Yearbook published by the Buros Institute. (The MMY is an almanac of sorts with information on hundreds of standardized tests. It is not available in full text online for free [I have access through my grad school] but you can find a copy at a university library or order information about a specific test from their web site at http://www.unl.edu/buros/index.html ).
According to the review by McMorris and Bringsjord:
So the scores you earn on tests published since 1991 should be roughly equal.
What you are asking about is the reliability of the LSAT (if a test is reliable, you will get roughly the same score every time you take it). Here’s what MMY has to say about that:
Elsewhere, the MMY review seems to indicate that scaled scores will vary about 2-3 points each time you take the test. So I think you could assume that if you are taking an old LSAT under approximate testing conditions (timing yourself, not using reference books to help you, etc), your score will be pretty reliable.
Disclaimer: I am almost done with my M.A. in Education, but I haven’t done a lot of coursework in testing and evaluation. If I’ve made any errors here, hopefully someone will come along and help us out 
hill
More than one of my acquaintances had done very badly on their first LSAT, then gone for tutoring classes, and then did very well. They also did well in school. Then they tanked when it came to actually working with lawyers.
None of these people had any real interest in practicing law, however, they were just gung how about “being a lawyer.”
Successful lawyers I know like to say it’s a hard work thing, but good lawyers don’t like to give anything away so take that answer with a grain of salt.
I should clarify, how accurate a predictor is an old LSAT as a predictor of a score on the current LSAT.
I took the test using the time limits. etc.
Accuracy of the LSAT? I dunno…how can you ascribe accuracy to something that’s supposed to be applied to everyone? There’s always going to be exceptions, and there’s always going to be people who are less knowledgeable but have the ability to do well / less well on standardized tests.
Test taking is a skill unto itself. Can it be used as a predictor of future success? Generally I’d say yes… Should it be taken as gospel? I’d say no way…
D.
Accuracy of the LSAT? I dunno…how can you ascribe accuracy to something that’s supposed to be applied to everyone? There’s always going to be exceptions, and there’s always going to be people who are less knowledgeable but have the ability to do well / less well on standardized tests.
Test taking is a skill unto itself. Can it be used as a predictor of future success? Generally I’d say yes… Should it be taken as gospel? I’d say no way…we all know "c’ level students or underachievers that are brilliant in practice, just not in theory…
Does scoring low on the LSAT predict that you’ll be a horrible lawyer? I wouldn’t jump to that conclusion…
Everyone’s different.
D.