LSAT Repeater Data

The LSAC has a lot of data on performance of repeat takers of the LSAT, including this very interesting table, which lists the  number of people at each LSAT score who repeat the test, the average of their new score, the standard deviatoin, the number who increase, those with no change, and those who increase their scores -- along with a distribtuion of scores for the new administration.  As the heading to the table says, "The table on this page presents distribution for 'Repeaters'; that is, candidates who took the LSAT in 2012–2013 after having taken it previously. . . .  Most people take the test only once; last year 66.8 percent of the total number of test takers took the LSAT just one time; 26.1 percent took the test twice; and approximately 7.1 percent took the LSAT more than twice."  I think this is important data for people contemplating taking the exam another time and it also gives a sense of just how much one can expect to increase a score at each point on the scale.

And if that's of interest to you, you will also likely be interested in The Performance of Repeat Test Takers on the Law School Admission Test: 2006–2007 Through 2012–2013 Testing Years by Laura A. Lauth, Andrea Thornton Sweeney, Christian Fox, Lynda M. Reese.   Cribbing now from their paper:

The average percentages of first-, second-, and third-time test takers over these 7 testing years were about 71%, 25%, and 4%, respectively.  Within testing years, the percentages of first-time and repeat test takers have followed a cyclic pattern. On average, the percentage of first-time test takers was about 83% in June, 76% in September/October, and 61% in both December and February.  Test takers who repeated the LSAT gained an average of 2.8 points the second time they took the test and 2.2 points the third time they took the test (compared to the second time).  The percentage of test administrations in which second-or third-time test takers have a higher mean than first-time test takers is larger in this report than in the previous report (Marcus et al., 2011). For a closer inspection of this pattern, see Tables A-5a and A-6.

Comments

[Ron]

Glad to see other that kids recognize how formulaic law school admissions are. I remember feeling like an idiot for thinking that there was anything holistic about the admissions process after someone introduced me to lawschoolnumbers.com. As it turned out, I was nothing more than an LSAT score attached to an LSAC GPA. Didn't matter that I majored in physics or had actual experience working as a patent agent. To be honest, I am perfectly fine with the way law school admissions works. Just wish prelaw advisors recognized this.

[Ralph Clifford]

The sad thing revealed by the table are the three people who got a 179 on the test who repeated it, hoping for the perfect score.

[Former Editor]

Ralph,

A perfect score might be worth thousands to tens of thousands of dollars in scholarship money to that taker. Why not take it again (particularly if the first time was far enough out not to delay an application)?

[Ralph Clifford]

If any school is distinguishing between a 179 and a 180 -- both within the same error band -- then I wonder as much about that school as I do about the perfect-seeking student.

[Former Editor]

I'll bet you a nickle that at some schools it makes the difference between a full and partial scholarship. Whether that makes any sense is a different matter.

[anon]

Because it’s above the 75th percentile at every school, it really doesn’t make sense why the difference should matter.

Unless the school applies a formulaic index score—perhaps those retakers were gunning for Cooley.

[Nathan A]

What anon @ 06:11 PM said. Law school admissions offices do a good job of adhering to their mission of maximizing the median LSAT/GPA of their entering classes.

Take UVA, for example. Anyone with a 168 is always rejected. Anyone with a 169 or above is almost never rejected. Apparently, if you had a 168, you would have been much more qualified for the study of law at UVA if you had just guessed right on a single question. Almost no applicants above the previous year's median GPA OR the previous year's median LSAT are ever rejected. If USNWR didn't factor in a school's acceptance rate into its rankings, UVA would probably automatically admit anyone with a GPA OR an LSAT score higher than one of the previous class' medians.

The ONLY applicants worth throwing serious scholarship money are those students who have BOTH an LSAT score and a GPA higher than a school's medians. Enrolling one of these kids allows you enroll another kid whose LSAT/GPA are below both medians (legacies, affirmative action, etc ...) without hurting your class profile.

Long story shot - given how well admissions offices play the game, I see almost no reason for them to care if you score a 180. If you have a 179 already, the best thing you can do for yourself is to take easy classes and boost your GPA.

[Kyle McEntee]

A few other reasons for a 179 to retake for a 180: the person works for a test prep company and has to retake every so often, or the person wants to advertise a perfect score so as to increase perception/price.

[BoredJD]

Scooped by Kyle.

I'm surprised and disheartened that the percentage of retakers is so low. Because of USNWR, law schools only have to report the highest score, and admissions are so numbers driven that even the average score increase of 2-3 points can be worth tens of thousands of dollars or admission to a better tier of school. There is almost no practical disadvantage to not retaking the first few times, and I can't imagine 60-70% of takers get an acceptable score where an increase will lead to no practical benefit, like a student who wants to practice small law in a local market who gets a score sufficient for a full-ride at their local school.

Any pre-law counselors lurking who can comment on whether they routinely advise applicants of the strategic value of retaking? I did not receive anything close to that advice from my pre-law counselor, but I could also see many low-information students just deciding that they don't want the stress of a second testing. It could also be that a disproportionate amount of one-time takers don't apply to law school, though we don't have numbers on that.

[Barry]

Former Edictor: "I'll bet you a nickle that at some schools it makes the difference between a full and partial scholarship. Whether that makes any sense is a different matter."

Or admission to HYS, perhaps.

[confused by your post]

BoredJD,

The real problem with retaking is that it is difficult to invest the time/money in adequately preparing AGAIN for the LSAT. That said, it a good idea for most to make that investment if they are able and really want to go to law school

[BoredJD]

The money thing is certainly an issue, especially if you care about trying to make the process as class-blind as possible, but this is not unique to law school- any academic program is going to have applicants who can take $4000 prep classes. For the self-studier there's a fair amount of free advice on the internet (TLS is the site I used), and the best prep books are relatively cheap. Most of the money is spent on buying prep tests, 20 are recommended. LSAC should be required to release that number of tests for free, but of course that would contravene their purpose of squeezing a little extra out of people in their early 20's looking for a better life- like price gouging lemons on the Mayflower.

We're not talking a huge investment of time here, 15-20 prep tests, 4 hours per test over 3-6 months (although again it's relative and the upper-middle class kid who doesn't have to work two jobs on campus is going to have an easier time). I think more people would invest the time if they better understood (1) the benefits of retaking, (2) how learnable its is- especially the logic games section, and (3) more about the nature of the legal job market and how your school plays into that. Prelaw advisors can really stress all this to applicants.

[confused by your post]

Wanted to say thanks to Professor Brophy for this brief but important post. My takeaway from it is that the LSAT is not especially learnable. Second time takers generally improve their scores only between 2-3 points. While such an improvement is certainly valuable to most takers, it is a modest one in relative terms. The average retaker does not dramatically improve upon the original score.

[BoredJD]

Confused - there's enough anecdotal evidence out there to suggest the test is learnable with big improvements around 10-15 prep tests. Especially logic games. The question is how much prep or study time an applicant can put into it and if they are using the right strategies.

It could be the case that most people do minimal or no studying between retakes. It's not an uncommon view among young people that standardized tests measure immutable aspects of intelligence. Or maybe students are peaking before they take the first test.