How do LSAT scores correlate with class standing? Forum

[Mousebandit]

I'm wondering if there is a resource that correlates LSAT scores, with class standings at end of 1L? The purpose of the LSAT is to predict"success" in law school, right?

Since first year grades and class standings determine OCI and summer internships, which have huge influence on post-grad employment, this would be the point in school to look at for correlation, right?

Mousebandit

[lymenheimer]

There is a slight correlation but that does not mean that it is, or can be, relied upon to any degree. Also, you don't know the LSAT+GPA combos of your fellow classmates. Also, they are usually so close that it doesn't make that much of a difference (Exactly how much better would a 166 be than a 165?).

[nick417]

I didn't know many of my classmates LSAT scores, however, the ones who did share, I saw very little correlation to 1L grades.

My issue with the LSAT is that it doesn't really test any type of knowledge that you need for law school. It loosely grades your writing capability (which I would argue is the best predictor of your 1L grades) and instead focuses on your memory (reading comp) and quick thinking skills (logic games). Neither are important for law school exams.

Law school exams are about knowing the material (whatever the Professor has taught you), spotting the legal issues, and drafting a well thought out and logical answer.

[nick417]

I wanted to add, I understand the purpose of the LSAT (there needs to be some neutral test given to potential law students to rank them) but I don't see much of a correlation. If you are scoring in the 170s, you are probably pretty intelligent already and should do well in law school if you put in the effort. However, if you are in the high 150s or 160s, I don't know how valuable the LSAT score is to determine your class standing

The one thing I have noticed about law school is the people at the top of the class tend to be the better writers. Since law school exams are writing intensive (and usually not multiple choice) good writers tend to flourish, especially if they can issue spot effectively.

[jdmonkey]

Very little. The LSAT is similar to an IQ test. Whereas # of hours productively devoted to study seems to predict grades. The people that know the law best will spot more issues and give more in-depth analysis. Also the LSAT doesn't take memory into account or speed of typing, which are both relevant.

[Mousebandit]

Thanks! The skills you're listing for success on the exams are probably what I'm ultimately looking for.

FWIW, I'm looking at going to a regional school, shooting for a full ride, but still will need to have excellent class standing to get the employment I'm wanting. I think, LOL. I'm just starting to research it all from a numbers point of view. I'm looking for public service employment in Alaska after graduation, and basically must have full-ride or very close to full-ride for COA, to make it reasonable. My GPA is going to be very low - 2.9-3.1 (I am non-trad student, have a 2.9 from 20 years ago and not enough credits still needed to bring it any higher than a 3.1), but my LSAT back in the day was very high, and I'm intending to devote a lot of effort to ensuring that it's at least that high (173) this time around. I'll be applying in the 2017-2018 cycle. So far, I'm looking at U of Seattle and Gonzaga as potentials. That's based on PNW location and ties to Alaska law. I haven't researched enough yet to come up with any other schools, and haven't done the employment stats research on those yet either. So, I am mostly still in the dark, LOL.

Anyways, THANK YOU! I'll make sure and include books on exam-taking skills in my reading this year!

[LikelyThrowaway]

2013 LSAC paper on the subject.

LSAT is a "substantially better" predictor for first-year GPA than undergrad GPA is. Using both is most predictive.

Median validity for LSAT: 0.37
Median validity for uGPA: 0.27

Median validity for using both: 0.47

As a reminder, a validity of zero would suggest zero correlation between two variables, and a validity of one would mean perfect correlation. So basically, LSAT gives you a hint at what's going to happen, and a bit better one than uGPA does, but it's a pretty weak hint if you're trying to figure out how you'll do as a 1L.

[oshberg28]

2013 LSAC paper on the subject.

LSAT is a "substantially better" predictor for first-year GPA than undergrad GPA is. Using both is most predictive.

Median validity for LSAT: 0.37
Median validity for uGPA: 0.27

Median validity for using both: 0.47

As a reminder, a validity of zero would suggest zero correlation between two variables, and a validity of one would mean perfect correlation. So basically, LSAT gives you a hint at what's going to happen, and a bit better one than uGPA does, but it's a pretty weak hint if you're trying to figure out how you'll do as a 1L.

Yeah, it's going to vary. My 1L GPA was almost dead on with my UGPA, and my 1L GPA it was significantly better than my LSAT score would have suggested.

[rav17]

This is just an anecdote but my cousin matriculated to Harvard Law several years ago with a 172 which is just slightly below median.
He ended up graduating magna cum laude, near the top of his class. In his case his LSAT score didn't correlate with his performance.

[Mousebandit]

I hadn't really given much thought to correlating uGPA with 1L GPA because they are graded on such different criteria. Most of uGPA classes, at least mine, weren't graded on curve - it was right answer / wrong answer, did you grasp the material. LS exams seem so much dicier, because you can theoretically, grasp all the material, express it well in your writing, and have 95% of the rest of the class do the same, and you end up in bottom 5%, LOL. I understand that's the nature of the game. Just was hoping for something to help me feel a little better about it, I guess, LOL.

STUDY, STUDY, STUDY. That's gonna be my mantra for the next few years. STUDY for the LSAT, STUDY so I don't divebomb my uGPA any worse than it already is, and STUDY even more in 1L!

[thsmthcrmnl]

they are usually so close that it doesn't make that much of a difference (Exactly how much better would a 166 be than a 165?).

This is right. It's an issue of range compression.

[ChildWithADove]

FWIW, my LSAT/GPA score placed me somewhere around the 30-40th percentile in my school. But I finished 1L in the 90th percentile of my class. That said, my school's admissions dpt. mentioned that they track LSAT/GPA scores with 1L performance. They said that it more often than not shares a strong correlation with 1L performance.

[jbagelboy]

I was the median LSAT at my school and placed several standard deviations away from that in my law school class. I knew students with below 25th LSATs that were at the very top of the class, and students with above 75th LSATs that were below median. I think statistically there is a slight correlation but its not very strong, unless your LSAT is very, very low for your class (like 5+ points below median), in which case you might be at a natural disadvantage.

[cyj43]

I think that even the correlations are misleading as it is a lot more complex than merely a score.

I.e. Someone with a diagnostic of 146 scores in the high 160's after x amount of time studying. This person, in my opinion, has not increased his or her chance of succeeding in law school through rigorous studying of material that is quite different from the material given in law school.

What about another person whose diagnostic is 156 and scores 164 as an example. My intuition leads me to believe that the person with the higher diagnostic but a lower final score may actually be just as likely if not more to succeed in law school, especially if the diagnostic score is primarily assessed from LG and RC.

Personally, i have a high LSAT score after a year of studying from a very low diagnostic. And i don't think my percentage or correlation for success in law school has gone up because i cracked a test after repeated exposure and endless amount of hours of studying. My RC is not better imo. I just know how to read for structure and sometimes i can anticipate certain excerpts in the passage to appear as questions--skills that are surely not something i will use in law school. I think?



Any thoughts on this?

[ChildWithADove]

I think that even the correlations are misleading as it is a lot more complex than merely a score.

I.e. Someone with a diagnostic of 146 scores in the high 160's after x amount of time studying. This person, in my opinion, has not increased his or her chance of succeeding in law school through rigorous studying of material that is quite different from the material given in law school.

What about another person whose diagnostic is 156 and scores 164 as an example. My intuition leads me to believe that the person with the higher diagnostic but a lower final score may actually be just as likely if not more to succeed in law school, especially if the diagnostic score is primarily assessed from LG and RC.

Personally, i have a high LSAT score after a year of studying from a very low diagnostic. And i don't think my percentage or correlation for success in law school has gone up because i cracked a test after repeated exposure and endless amount of hours of studying. My RC is not better imo. I just know how to read for structure and sometimes i can anticipate certain excerpts in the passage to appear as questions--skills that are surely not something i will use in law school. I think?


Any thoughts on this?

Law school is much more about long term studying than natural intelligence. I think that the person that can go from 146 to 170 has the best chance to do well in law school compared to a person that went from 168 to 170. The former person knows how to study and appreciates the process of learning - and they are smart as hell too. The latter person is just smart.

[cyj43]

Slightly disagree here w the above statement in the context of 1L.

The person who goes from 168-->170 is more likely satisfied w the first score and thus does not need to study long term as someone who had a 20+ point increase. And in the context of 1L, you really don't get a "long term" chance to study and figure law school out. SA positions hinges on 1L grades and unlike LSAT, there is no retake or taking when ready.

I would actually think the 168 diagnostic is smarter than the 140's diagnostic but this is heavily arbitrary and dependent from person to person.

Also as prev mentioned, as someone who has had a 20+ jump over a long term study, i realize that LSAT requires mastery of techniques more than comprehension and critical thinking. I have used technique for certain passages i could not comprehend in both LR and RC, and i'm sure many have done the same as these techniques are quite common and discussed through LSAT prep materials.

Consequentially, this has saved me from perhaps 6-8 questions, equivalent to maybe a 4-5point increase. Thus, i'm not sure if someone like me can be seen with a favorable light in law school over the person w a high diagnostic and lower final Lsat score, who understood the entirety of the passage/questions but did not develop a technique/mastery of LSAT skills

[PeanutsNJam]

The correlation is poor enough that it would be foolish to make a gamble on your 1L performance based on your LSAT score. Would you bet $200,000 with a 20% chance of it paying off?