Comparative Difficulties Between the LSAT and GPA. Forum

[Fish127]

Looking back on my life, it occurred to me earlier that I spend most of my time bouncing back and forth between extreme arrogance and crippling self-doubt. One moment I will be perfectly confident about something, perhaps even overconfident, and then the next I will be unsure of myself, and blaming my previous arrogance for feelings that in all rights I should be confident with. I just wanted to give that bit of back-story to explain why I'm making what some may view as a strange thread, the information of which probably won't make much of a difference with my plans.

Anyway... I'm currently in my second Sophomore semester of UG, and my GPA is a 3.9. Needless to say, I've been pretty happy about my prospects for law school given this, however lately I've been starting to worry about taking the LSAT. I looked up some sample questions online and tried one of the logic games, and honestly, it scared the hell out of me. Now I haven't actually studied for it at all yet, or even looked into doing so. I'm planning to start that after this semester actually.

I'm basically making this thread to ask about other people's experience with the test. Has anybody ever done really well with UG grades, and then simply been unable to get comparable scores on the LSAT? Which do most people find more difficult to succeed with?

[esther0123]

Well this will spark an interesting debate I think...

And personally I'm a reverse splitter (3.98/169) in the sense that my LSAT doesn't really match my GPA in terms of its strength. I went to a pretty reputable program so I don't think the GPA was a result of grade inflation.

But I did do a B.Sc. in Life Sciences and I am ETL (I learned Korean and French before learning English). So I think the fact that I did not speak English until I was 13 perhaps had some influence on my relative difficulty with LSAT but who knows. My program also did not require much reading so I never got in the habit of it.

Getting good grades was honestly just about grit-and-grind for me. I just got through whatever work was there, or tried to, and when I did, I got good grades. Whereas with LSAT, I found it to be not such a linear progress in terms of my effort vs. performance.... the correlation was more logarithmic than linear...

So that's my two cents. I'm interested to hear what others have to say!

[Moonlight]

Content Deleted

[daryldixon]

This is a stupid question.

Success on the LSAT has nothing to do with success in UG.

[sjgonzalez3]

Using a cold diagnostic LG question as a measure of how you'll feel on the test is silly. It's by far the most "trainable section." I went from -15 on it to -0 within a month.

Regarding your more specific question about uGPA & LSAT correlation, I agree with a poster above me who said there is very little, if any. This is why splitters and reverse splitters exist. Bottom line though, study hard for the LSAT and you can get a score high enough that a beautiful GPA like that can get you into the T14 with very little trouble.

[guano]

Simple answer:
LSAT tests innate ability,or talent
GPA tests work ethic, or drive

You get a high GPA by working hard, while you get a high LSAT by having good logic/reasoning/etc.

Of course, this isn't actually true. Some people get a really high GPA while doing minimal work, others struggle to get Cs. Likewise, the LSAT is a learnable test and you can get a higher score by drilling down

But that's the concept. They both test different things, and people with low GPAs can get high LSATs and vice versa

[esther0123]

Yeah.. and to add to that, LSAT is a very specific test targeted towards your linguistic and linear logic ability.

There are more than one ways of being "intelligent" though. For instance, you could be excellent wrt your spatial intelligence but fairly poor in your linguistic intelligence. In this case, you would be at a disadvantage in LSAT but might do well in Calculus III.

To what extent the qualities being tested on LSAT is reflective of your future performance in law school is questionable, but I hear there is SOME correlation (0.3) which is considered "decent" in social sciences.

[guano]

Yeah.. and to add to that, LSAT is a very specific test targeted towards your linguistic and linear logic ability.

There are more than one ways of being "intelligent" though. For instance, you could be excellent wrt your spatial intelligence but fairly poor in your linguistic intelligence. In this case, you would be at a disadvantage in LSAT but might do well in Calculus III.

To what extent the qualities being tested on LSAT is reflective of your future performance in law school is questionable, but I hear there is SOME correlation (0.3) which is considered "decent" in social sciences.

I'm not going to look it up, but I recall the correlation was about 0.51, which means that roughly half of the people at the 75% LSAT will be top 25% of the class, and half the people at the 25% LSAT will be at the bottom 25% of the class

[cron1834]

NO WAY is .51 the "true" coefficient. That's a high-end estimate. From LSAC:

Correlations between LSAT scores and first-year law
school grades ranged from .12 to .56 (median is .36). The
correlations between UGPA and first-year law grades ranged
from .09 to .45 (median is .28)

ETA - I often wonder what percentage of these studies are performed by folks with a vested interest in having the correlations look meaningful, as opposed to legit third-party/neutral sources. I guess you'd have to find a meta analysis and source it to figure this out. My hunch is that the true number is slightly below their reported medians here.

[drawstring]

Simple answer:
LSAT tests innate ability,or talent
GPA tests work ethic, or drive

You get a high GPA by working hard, while you get a high LSAT by having good logic/reasoning/etc.

Of course, this isn't actually true. Some people get a really high GPA while doing minimal work, others struggle to get Cs. Likewise, the LSAT is a learnable test and you can get a higher score by drilling down

But that's the concept. They both test different things, and people with low GPAs can get high LSATs and vice versa

My own experience reflected this. Generally speaking, I found that changes in how hard I worked had a much greater impact on my GPA than my LSAT score. When I was working hard and putting the time in I almost always earned As or A+s, and when I slacked off my marks would drop. While hard work played an important role in getting the LSAT score I did, my scores did not change as strongly in proportion to the effort I put in.

And I know it's almost axiomatic on here that most people can get a 170+ just through hard work and smart studying, but I don't agree.

[esther0123]

Yeah.. and to add to that, LSAT is a very specific test targeted towards your linguistic and linear logic ability.

There are more than one ways of being "intelligent" though. For instance, you could be excellent wrt your spatial intelligence but fairly poor in your linguistic intelligence. In this case, you would be at a disadvantage in LSAT but might do well in Calculus III.

To what extent the qualities being tested on LSAT is reflective of your future performance in law school is questionable, but I hear there is SOME correlation (0.3) which is considered "decent" in social sciences.

I'm not going to look it up, but I recall the correlation was about 0.51, which means that roughly half of the people at the 75% LSAT will be top 25% of the class, and half the people at the 25% LSAT will be at the bottom 25% of the class

I think 0.5 was LSAT combined with GPA in predicting law school performance. I think LSAT was marginally better at predicting law school performance than GPA, but neither was all that great of a predictor. And also, just like what the previous comment says, we don't now how or who's conducted the study so the extent of how biased it is remains unknown. I too would think that some people are heavily invested in inflating LSAT's predictive ability....

[scoobysnax]

The LSAT is like any subject in UG. Some people have a natural aptitude for it, while others don't. But whether or not you have a knack for it, it requires work, and hard work can make up some difference in aptitude. I think it's possible for most people to get a decent score 160+ as long as they put in the time.

Also, don't use your first LG to judge your ability. There's a pretty steep learning curve (for me anyway)

[Pneumonia]

This is a stupid question.

Success on the LSAT has nothing to do with success in UG.

[Fish127]

This is a stupid question.

Success on the LSAT has nothing to do with success in UG.

I never claimed that it did, nor was that what I was asking. I was simply asking about people's experience with both.

[furrrman]

Has anybody ever done really well with UG grades, and then simply been unable to get comparable scores on the LSAT?

Yeah, I know people like this. That said, its way too early to get discouraged. It took me a couple months of hard studying before I started getting good scores on my PTs.

[KD35]

There are way too many variables that go into GPA and LSAT (especially GPA) for there to be any sort of statistically significant correlation.

[cron1834]

There are way too many variables that go into GPA and LSAT (especially GPA) for there to be any sort of statistically significant correlation.

THERE IS a statistically significant correlation. I cited the coefficients above. What on earth are you talking about? That's not even up for debate. The question is what do we make of it, and where does causality lie?

[guano]

There are way too many variables that go into GPA and LSAT (especially GPA) for there to be any sort of statistically significant correlation.

THERE IS a statistically significant correlation. I cited the coefficients above. What on earth are you talking about? That's not even up for debate. The question is what do we make of it, and where does causality lie?

I think KS meant correlation between LSAT and UGPA, not LSAT and law school GPA

[cron1834]

There are way too many variables that go into GPA and LSAT (especially GPA) for there to be any sort of statistically significant correlation.

THERE IS a statistically significant correlation. I cited the coefficients above. What on earth are you talking about? That's not even up for debate. The question is what do we make of it, and where does causality lie?

I think KS meant correlation between LSAT and UGPA, not LSAT and law school GPA

Ah. I guess the "too many variables" line threw me. Wtf does that even mean? We can easily correlate uGPA with jdGPA. uGPA and LSAT? No idea what that looks like. I'd be surprised if there were NO coefficient, though.

[TigerDude]

Recognize that the "good score" of 170+ puts you in the top 2.5% or so of all test takers. That's pretty rare air - a sampling of people who want to go to law school is probably a tougher crowd than the SAT.

People believe the LSAT is a learnable skill. That is probably true up to your max, whatever that is.

[TheSpanishMain]

That's pretty rare air - a sampling of people who want to go to law school is probably a tougher crowd than the SAT.

Probably, but I'd guess not by a lot. If my LSAT administration is representative, a LOT of complete idiots take the LSAT. As far as the idea of a "max", I tend to agree. I don't think everyone is capable of breaking 170, but I do think that every reasonably intelligent person who puts in the work should be able to land in the 160s, which is about where the whole law school decision starts to change from "horrible idea" to "potentially good investment."

[zhenders]

In comparing the LSAT with my former teaching experience and comparing it to undergrad, I would have to both agree and disagree with the comments above. My opinion is as follows:

GPA: strongly work ethic correlated; no doubt about it.

LSAT: equally work ethic correlated, BUT strong (very strong) English comprehension ability is necessary for a top (170+) score. It is not in any way sufficient, but without it, no amount of study -- unless that study is focused specifically on one's command of the English language -- is going to get you into the 99th.

While this may be less true for LG, if one hopes to score consistently well (-0/-2) on LR or RC, it is imperative that you can read accurately, quickly and comprehensively, with zero vocabulary issues, no trouble parsing complex sentences, and the ability to immediately spot e subtle nuances between words (a great barrier for many in LR).

[KD35]

There are way too many variables that go into GPA and LSAT (especially GPA) for there to be any sort of statistically significant correlation.

THERE IS a statistically significant correlation. I cited the coefficients above. What on earth are you talking about? That's not even up for debate. The question is what do we make of it, and where does causality lie?

I think KS meant correlation between LSAT and UGPA, not LSAT and law school GPA

Correct, meant LSAT and uGPA...important detail I didn't add

[cron1834]

There are way too many variables that go into GPA and LSAT (especially GPA) for there to be any sort of statistically significant correlation.

THERE IS a statistically significant correlation. I cited the coefficients above. What on earth are you talking about? That's not even up for debate. The question is what do we make of it, and where does causality lie?

I think KS meant correlation between LSAT and UGPA, not LSAT and law school GPA

Correct, meant LSAT and uGPA...important detail I didn't add

I would be curious to see what the #s are.

[drawstring]

What I've wondered is if one's scores on certain sections are better correlated with law school GPA than other sections. For example, are LG scores more strongly/weakly related with GPA than LR or RC scores are?

I always felt like LG was somehow disconnected from the type of work done in law school, while reading comp and logical reasoning were far more relevant.