LSAT correlation to GRADES

A forum for applicants and admitted students to ask law students and graduates about law school and the practice of law.
005618502
Posts: 2577
Joined: Thu May 06, 2010 10:56 pm

LSAT correlation to GRADES

Postby 005618502 » Tue Jun 19, 2012 6:37 pm

So it has been talked about many times on here, LSAT is the single highest indicator of how you will do in law school. Even then, it is not that high of a correlation.

Here is the question, arent there to many "variables" to this?

I had a relatively high diagnostic, but my score only increased 5 points because RC was my hardest section.

Person 1: Diagnostic is a 160, RC kills. Even after taking a class and studying your ass off you get a 165 on test day. 4.0 gpa

Person 2: Diagnostic is a 155, games kill. After taking a class and studying your ass off you get a 171 on test day. 3.5 gpa

Who is better off if these two both end up at the same school?

I would imagine that person 1 has the advantage since the test is learnable, but the diagnostic shows true "intelligence" (as much as it tests intelligence that is). Or is it the opposite?

I am curious to know if people know many people who are in the bottom 25% LSAT at their school but end at top of the class

User avatar
smaug_
Posts: 2195
Joined: Mon Oct 17, 2011 5:06 pm

Re: LSAT correlation to GRADES

Postby smaug_ » Tue Jun 19, 2012 6:46 pm

AssumptionRequired wrote:
I am curious to know if people know many people who are in the bottom 25% LSAT at their school but end at top of the class


I would wager quite a few. Considering that the LSAT is meant to be correlate with the entire range of scores, most students at the same school are in a near dead heat. I don't see how anyone could feel comfortable picking one student to succeed over another.

User avatar
facile princeps
Posts: 421
Joined: Tue Jun 12, 2012 8:51 pm

Re: LSAT correlation to GRADES

Postby facile princeps » Tue Jun 19, 2012 7:18 pm

AssumptionRequired wrote:Person 1: Diagnostic is a 160, RC kills. Even after taking a class and studying your ass off you get a 165 on test day. 4.0 gpa

Person 2: Diagnostic is a 155, games kill. After taking a class and studying your ass off you get a 171 on test day. 3.5 gpa

Who is better off if these two both end up at the same school?

I would imagine that person 1 has the advantage since the test is learnable, but the diagnostic shows true "intelligence" (as much as it tests intelligence that is). Or is it the opposite?

In your hypothetical here i'd bet on person 2 to be more successful in law school. The games are more 'learnable' than RC, and from what i've read, the latter seems to be more important. Person 1 already tried hard with RC and still doesn't get it.

005618502
Posts: 2577
Joined: Thu May 06, 2010 10:56 pm

Re: LSAT correlation to GRADES

Postby 005618502 » Tue Jun 19, 2012 7:34 pm

facile princeps wrote:
AssumptionRequired wrote:Person 1: Diagnostic is a 160, RC kills. Even after taking a class and studying your ass off you get a 165 on test day. 4.0 gpa

Person 2: Diagnostic is a 155, games kill. After taking a class and studying your ass off you get a 171 on test day. 3.5 gpa

Who is better off if these two both end up at the same school?

I would imagine that person 1 has the advantage since the test is learnable, but the diagnostic shows true "intelligence" (as much as it tests intelligence that is). Or is it the opposite?

In your hypothetical here i'd bet on person 2 to be more successful in law school. The games are more 'learnable' than RC, and from what i've read, the latter seems to be more important. Person 1 already tried hard with RC and still doesn't get it.


That's interesting. I guess it's just hard to see how reading slightly faster would be that much more important. But I guess law school tests are all about time as well.

User avatar
dowu
Posts: 8334
Joined: Wed Mar 28, 2012 9:47 pm

Re: LSAT correlation to GRADES

Postby dowu » Tue Jun 19, 2012 7:38 pm

AssumptionRequired wrote:
facile princeps wrote:
AssumptionRequired wrote:Person 1: Diagnostic is a 160, RC kills. Even after taking a class and studying your ass off you get a 165 on test day. 4.0 gpa

Person 2: Diagnostic is a 155, games kill. After taking a class and studying your ass off you get a 171 on test day. 3.5 gpa

Who is better off if these two both end up at the same school?

I would imagine that person 1 has the advantage since the test is learnable, but the diagnostic shows true "intelligence" (as much as it tests intelligence that is). Or is it the opposite?

In your hypothetical here i'd bet on person 2 to be more successful in law school. The games are more 'learnable' than RC, and from what i've read, the latter seems to be more important. Person 1 already tried hard with RC and still doesn't get it.


That's interesting. I guess it's just hard to see how reading slightly faster would be that much more important. But I guess law school tests are all about time as well.


Tag for the lulz.

005618502
Posts: 2577
Joined: Thu May 06, 2010 10:56 pm

Re: LSAT correlation to GRADES

Postby 005618502 » Tue Jun 19, 2012 7:49 pm

nmop_apisdn wrote:
AssumptionRequired wrote:
facile princeps wrote:
AssumptionRequired wrote:Person 1: Diagnostic is a 160, RC kills. Even after taking a class and studying your ass off you get a 165 on test day. 4.0 gpa

Person 2: Diagnostic is a 155, games kill. After taking a class and studying your ass off you get a 171 on test day. 3.5 gpa

Who is better off if these two both end up at the same school?

I would imagine that person 1 has the advantage since the test is learnable, but the diagnostic shows true "intelligence" (as much as it tests intelligence that is). Or is it the opposite?

In your hypothetical here i'd bet on person 2 to be more successful in law school. The games are more 'learnable' than RC, and from what i've read, the latter seems to be more important. Person 1 already tried hard with RC and still doesn't get it.


That's interesting. I guess it's just hard to see how reading slightly faster would be that much more important. But I guess law school tests are all about time as well.


Tag for the lulz.


Sad cause I seriously thought person #1 would be better off due to doing better initially. I just don't see how needing 5 more mins than someone else on a standardized test would actually make a legit difference. I've worked in a law office for a couple years. Even at trial your not in that kind of time crunch

User avatar
acrossthelake
Posts: 4432
Joined: Sat May 16, 2009 5:27 pm

Re: LSAT correlation to GRADES

Postby acrossthelake » Tue Jun 19, 2012 7:52 pm

Not a lot in the individual school since most students are within each other's score bands. Even the creators of the LSAT state that the test shouldn't be able to really say much within such a narrow band.

PKozi
Posts: 47
Joined: Mon Sep 12, 2011 4:29 pm

Re: LSAT correlation to GRADES

Postby PKozi » Tue Jun 19, 2012 7:57 pm

The fact that it is a weak correlation between LSAT and grades means is that if you had 100 pairings of Persons 1 and 2, in 55 of the pairings person 1 would do better and in 45 person 2 would do better. Or vice versa. Or with some other split that's close to 50/50. The point is that LSAT and GPA only explain a small amount of the variation in grades.

005618502
Posts: 2577
Joined: Thu May 06, 2010 10:56 pm

Re: LSAT correlation to GRADES

Postby 005618502 » Tue Jun 19, 2012 8:06 pm

PKozi wrote:The fact that it is a weak correlation between LSAT and grades means is that if you had 100 pairings of Persons 1 and 2, in 55 of the pairings person 1 would do better and in 45 person 2 would do better. Or vice versa. Or with some other split that's close to 50/50. The point is that LSAT and GPA only explain a small amount of the variation in grades.


This makes sense. Thanks for the post. That's what I meant by variables. How much do you get it? How hard do you study? How many other commitments do you have (work, wife, kids, etc.)

User avatar
laxbrah420
Posts: 2748
Joined: Fri Aug 12, 2011 1:53 am

Re: LSAT correlation to GRADES

Postby laxbrah420 » Tue Jun 19, 2012 8:15 pm

So I got a diagnostic 162 and 165 after studying a bit, killed on rc. I presume I'll do terrible in law school due to my inability to learn

005618502
Posts: 2577
Joined: Thu May 06, 2010 10:56 pm

Re: LSAT correlation to GRADES

Postby 005618502 » Tue Jun 19, 2012 8:23 pm

laxbrah420 wrote:So I got a diagnostic 162 and 165 after studying a bit, killed on rc. I presume I'll do terrible in law school due to my inability to learn


:lol:

User avatar
2014
Posts: 5831
Joined: Sat Jun 05, 2010 3:53 pm

Re: LSAT correlation to GRADES

Postby 2014 » Wed Jun 20, 2012 12:37 am

With the score band thing, the only way to really see a LSAT advantage is to look at students who are attending schools they are completely over qualified for. Splitters are useless because UGPA is the 2nd highest correlation to LS GPA and its rough trying to figure out whether the splitters shitty UGPA and related habits that led to it are playing a more prominent role than the LSAT or vice versa. So the only really relevant people would be a group controlled for UGPA, as in someone roughly median I guess, but way over the 75th of the LSAT. That just doesn't seem to happen a whole lot.

User avatar
arkansawyer
Posts: 222
Joined: Wed Nov 03, 2010 5:11 pm

Re: LSAT correlation to GRADES

Postby arkansawyer » Wed Jun 20, 2012 12:44 am

My 2 cents as an LSAT instructor. RC tests a lot of things, but reading speed is the predictive variable. All other things equal, a faster reader probably has an advantage in law school.

User avatar
dowu
Posts: 8334
Joined: Wed Mar 28, 2012 9:47 pm

Re: LSAT correlation to GRADES

Postby dowu » Wed Jun 20, 2012 12:54 am

arkansawyer wrote:My 2 cents as an LSAT instructor. RC tests a lot of things, but reading speed is the predictive variable. All other things equal, a faster reader probably has an advantage in law school.


How so?

Say for instance it take one 3 minutes to read a passage and another person takes 4 minutes to read that same passage. Are you saying that such a marginal difference in reading speed would fare much better for one in LS?

ETA: didn't mean to ask a loaded question lol but you know the gist of what I'm asking.
Last edited by dowu on Wed Jun 20, 2012 1:23 am, edited 1 time in total.

User avatar
arkansawyer
Posts: 222
Joined: Wed Nov 03, 2010 5:11 pm

Re: LSAT correlation to GRADES

Postby arkansawyer » Wed Jun 20, 2012 1:08 am

Haha I know what you're getting at. I'm just a 0L so I can't pretend to have any expert knowledge about law school; however, it seems intuitive to me that the person who reads faster will have more time contemplate the material (cases, supplements, hornbooks). Obviously, this won't be much of an advantage with the 3 vs 4 minute reading time that you examples above, but my experience is that the reading speed gulf between the top 25% and bottom 25% in my classes was extreme. Still, might not make a huge difference in a class with tight score bands.

User avatar
ben4847
Posts: 789
Joined: Mon Nov 07, 2011 11:38 pm

Re: LSAT correlation to GRADES

Postby ben4847 » Wed Jun 20, 2012 1:10 am

Well, I have an LSAT score, and I have grades. So I'd say the correlation is 1:1

clone22
Posts: 354
Joined: Sat Oct 23, 2010 3:38 pm

Re: LSAT correlation to GRADES

Postby clone22 » Wed Jun 20, 2012 1:14 am

Although there's some correlation, it is far from perfect. People with identical LSATs can end up on opposite ends of the curve. And reading fast isn't necessarily that useful in law school. You'll have more than enough time to read your material. Comprehension is what's important.

Also typing speed. If you're not amazing at typing, start doing typing drills now. In some exams there will be far more issues to spot/material to write about than you can physically type out in the time allotted. Having 80+ wpm typing speeds will save your ass in those finals.

User avatar
arkansawyer
Posts: 222
Joined: Wed Nov 03, 2010 5:11 pm

Re: LSAT correlation to GRADES

Postby arkansawyer » Wed Jun 20, 2012 1:17 am

Wormfather wrote:
arkansawyer wrote:My 2 cents as an LSAT instructor. RC tests a lot of things, but reading speed is the predictive variable. All other things equal, a faster reader probably has an advantage in law school.


I can see how this works if you have a fast reader who knows what they are looking for, which to some extent is what happens in the LSAT. However to be able to read something, comprehend it, and be able to recall it later when it becomes relevant is something the LSAT doesn't capture.

Kind of like the difference between the kids who always ask "Is this going to be on the test?" and the ones who look to have an understanding of the whole document or what have you.


I think that RC does test recall and comprehension. It's definitely mental quick-draw, but I imagine students who can do it during the LSAT will be able to do it during law school.

bruss
Posts: 470
Joined: Tue May 24, 2011 3:58 am

Re: LSAT correlation to GRADES

Postby bruss » Wed Jun 20, 2012 1:33 am

Stupid

User avatar
PDaddy
Posts: 2073
Joined: Sat Jan 16, 2010 4:40 am

Re: LSAT correlation to GRADES

Postby PDaddy » Wed Jun 20, 2012 2:17 am

AssumptionRequired wrote:...but the diagnostic [LSAT] shows true "intelligence"...


Because the LSAT is a learnable test (as you concede), the diagnostic cannot test intelligence. It can only measure how fast you read and how innately adept you are at employing the specific skills needed tpo score high on the test.

I personally believe that, until the LSAC publishes the actual data supporting its claim that LSAT performance correlates with 1L grades, everyone should view the claim with suspicion. LSAC has published the "results" of the data (graphs and the like), but never the actual data supporting this claim. That's very suspicious, especially when the council that produces the LSAT and all of the official prep materials has every reason to misrepresent the test's accuracy as a predictor of law performance.

To the extent that time plays such an important role in performance on the exam, one's performance is likely determined more by reading speed than any other factor. The LSAC admits that greater than 50% of all test takers could get 80-85% of the questions correct if given the proper amount of time. As it stands now, only 15% of test takers do so. Time constraints thus determine the score distributions. This suggests that the LSAT's value as a measure of actual reasoning ability is overestimated, and that the test is more about reading speed and "nerve management" than anything else.

Reading quickly is not always a great asset in law school, and it can cause problems in real-world practice. Comprehension matters most. I think the LSAT should be computer-adaptive, as are the GMAT and GRE. That would reward speed, accuracy and degree of difficulty.

User avatar
smaug_
Posts: 2195
Joined: Mon Oct 17, 2011 5:06 pm

Re: LSAT correlation to GRADES

Postby smaug_ » Wed Jun 20, 2012 2:49 am

PDaddy wrote:
Reading quickly is not always a great asset in law school, and it can cause problems in real-world practice. Comprehension matters most. I think the LSAT should be computer-adaptive, as are the GMAT and GRE. That would reward speed, accuracy and degree of difficulty.


I was under the impression that this was to avoid compression of scores at the top end of the range. (And even then it doesn't really work well for the GRE)

User avatar
Ohiobumpkin
Posts: 465
Joined: Fri Oct 21, 2011 9:50 am

Re: LSAT correlation to GRADES

Postby Ohiobumpkin » Thu Jun 21, 2012 10:51 am

I would predict that person 2 would do better in law school because RC is probably a more important skill than AR. Having trouble with RC is pretty bad news bears in law school. That being said, I believe if I remember correctly, the LSAT has a predictive value of .41 for your first year law school GPA, while your UGGPA has a predictive value of .09 (or something comparatively low). So the prospect that somebody scored on the LSAT at the 25% at their LS class and ended up at the top of their class is certainly possible.

I think part of the problem with the predictive value of both the LSAT and UGGPA is the fact that so many variables exist that influence one's first year LS GPA. What section were you in (consequently, the professors you took)? Did you have relationship or social issues? State of mind coming into law school (sane or bat shit crazy :p )? How quickly did you "get" law school work? You get the picture.

Until I get my grades back and compare them with my LS comrades, I will assume all of them are 4.0 UGGPA and 180 LSAT score holders. It will keep me motivated to rack my brain day in and day out for three and half months twice a year for three years.

User avatar
jkpolk
Posts: 895
Joined: Thu Nov 10, 2011 10:44 am

Re: LSAT correlation to GRADES

Postby jkpolk » Thu Jun 21, 2012 11:19 am

the person who does better at law school will do better at law school.

User avatar
DC_Patent_Law
Posts: 139
Joined: Fri Feb 25, 2011 7:03 pm

Re: LSAT correlation to GRADES

Postby DC_Patent_Law » Fri Jun 22, 2012 12:08 am

LOL, I'll go with A.

Wormfather wrote:
Ohiobumpkin wrote:I would predict that person 2 would do better in law school because RC is probably a more important skill than AR. Having trouble with RC is pretty bad news bears in law school. That being said, I believe if I remember correctly, the LSAT has a predictive value of .41 for your first year law school GPA, while your UGGPA has a predictive value of .09 (or something comparatively low). So the prospect that somebody scored on the LSAT at the 25% at their LS class and ended up at the top of their class is certainly possible.

I think part of the problem with the predictive value of both the LSAT and UGGPA is the fact that so many variables exist that influence one's first year LS GPA. What section were you in (consequently, the professors you took)? Did you have relationship or social issues? State of mind coming into law school (sane or bat shit crazy :p )? How quickly did you "get" law school work? You get the picture.

Until I get my grades back and compare them with my LS comrades, I will assume all of them are 4.0 UGGPA and 180 LSAT score holders. It will keep me motivated to rack my brain day in and day out for three and half months twice a year for three years.


Ohiobumpkin's feelings towards LSAT reading comprehension can best be described as which of the following?

A. Resigned Acceptance
B. Plausible Resignation
C. Academic Ambivalence
D. Evenhanded Acceptance
E. Unbridled Enthusiasm

You have 30 seconds to decide.

User avatar
warandpeace
Posts: 301
Joined: Sat Mar 12, 2011 1:43 pm

Re: LSAT correlation to GRADES

Postby warandpeace » Fri Jun 22, 2012 12:24 am

has to be d, brah




Return to “Ask a Law Student / Graduate”

Who is online

Users browsing this forum: No registered users and 7 guests