intelligence vs LSAT Forum

[waker57]

What I am saying is that there is greater evidence for our claim, yet you disregard it, only to buy into one that has less evidence. AT NO POINT, did I say that money spent can not help you to get a greater score. Pointing out that a claim has been backed up with little evidence DOES NOT mean one rejects that claim. Remember that.

Also, I believe that the money spent is probably more of a random correlation than LSAT to 1L scores. Anyone could spend $1000 on a course, but a $700 course might prepare you better. There is no proof the more expensive one causes better grades. There's also the factors of how many books one buys, at what price. I bought all mine used on amazon, so they were alot cheaper, meaning I spent a relatively low amount of money, but my preparation yielded great results, but we digress.

What concerns me is you still, after all of these posts, cannot seem to defend your claim that

The LSAT has almost nothing to do with intelligence, and has even less to do with law school.

Any test that you can throw money at to do better on ceases to become a test of intellect.

How long are we going to have to wait to hear your explanation for the assertion that LSAT has "even less" than "almost nothing" to do with Law School performance?

[waker57]

LSAT and intelligence are probably positively correlated. I highly doubt they have a highly random or negatively correlated relationship.

Umm.. no kidding??

I also would predict they do have a correlation to how well someone winds up doing in law school, otherwise I would think over 200 ABA Approved Law Schools probably would not require the test.

This is some correlation, but the best is seen via considering UGPA and LSAT together.

Sorry, I just saw this post. Interesting that you'd take the sarcastic tone of "no kidding" when it was in response to the outrageous claim by rocketman that "The LSAT has almost nothing to do with intelligence".

[waker57]

I just wanted to highlight this. How does it feel to be the only person who doesn't think that money will get you a higher LSAT score?

I bet all these people buying preptests, buying PS bibles, paying for courses and tutors, are so misguided.

law of diminishing returns

That doesnt matter, he says money yielding a higher score is unproven. That's a statement that was unqualified.

Show me the evidence that money spent yields a higher score. Popular belief, (even though I agree with it) is not sufficient evidence.

[Rocketman11]

What I am saying is that there is greater evidence for our claim, yet you disregard it, only to buy into one that has less evidence. AT NO POINT, did I say that money spent can not help you to get a greater score. Pointing out that a claim has been backed up with little evidence DOES NOT mean one rejects that claim. Remember that.

Also, I believe that the money spent is probably more of a random correlation than LSAT to 1L scores. Anyone could spend $1000 on a course, but a $700 course might prepare you better. There is no proof the more expensive one causes better grades. There's also the factors of how many books one buys, at what price. I bought all mine used on amazon, so they were alot cheaper, meaning I spent a relatively low amount of money, but my preparation yielded great results, but we digress.

What concerns me is you still, after all of these posts, cannot seem to defend your claim that

The LSAT has almost nothing to do with intelligence, and has even less to do with law school.

Any test that you can throw money at to do better on ceases to become a test of intellect.

How long are we going to have to wait to hear your explanation for the assertion that LSAT has "even less" than "almost nothing" to do with Law School performance?

Lack of evidence to the contrary. The study from LSAC, for instance, uses UGPA and LSAT to show correlation with 1L grades across universities. Two inherent flaw is that 1L classes can vary, and we all know grading varies. Yale has the highest bracket of LSAT scorers, but with their Pass/Fail grading system do you really think they grade as competitively as an institution in the lower T14? From what I've heard and, admittedly this is heresay, once you get into Y/S it's not that tough to do well. Now I'm sure you'll say "well thats because they're all so smart! Not because classes are easier!" but no studies that I've read at least address that.

[waker57]

I would like to first concede one point, at least partially. You never originally said that as money spent by an individual increased, so did LSAT improvements, only that initially if money was spent, it would result in an increased score. Since almost everyone who preps for the LSAT is forced to spend greater than $0 in order to do so, you are technically right. I am not sure that is the point you were trying to make, but in the spirit of Scalia and textualism. I'll give you that one.

Due to the flaws you pointed out, there is then really no way to measure "success" in Law School objectively. Same can be said for life perhaps.

However, you made an assertion, which is that LSAT has less than almost nothing to do with law school, a statement for which you still have no basis.

[jrs12]

This is a silly argument, since there's nothing close to a consensus on the definition of intelligence. It's worth noting that pretty much every contemporary educational psychologist subscribes to some notion of "multiple intelligences." The LSAT tests specific intelligences, which may or may not go hand-in-hand with other intelligences. I'm hopeless at foreign languages, but LSAT-type reasoning was a breeze for me. I also doubt that you can really game the test. I bought 3 books, worked through them, and improved, but I never felt that there were any tricks that helped me get better. I was simply sharper, in the same way that I am sharper for a math exam when I spend the weekend working practice problems.

The LSAT correlates well with law school grades, but that doesn’t mean that it indicates what kind of lawyer one will become. It certainly doesn’t measure emotional intelligence, which is probably the most important ability for anyone arguing before a jury.

(By the way, causation is irrelevant. Of course the LSAT doesn’t cause you to get good grades. Those Michigan students who may get accepted without taking the LSAT—they’ll still do well. Both LSAT and grades are “caused,” in some sense, by intelligence. You would have to ask yourself about these hypothetical Pepsi drinkers, “why did they all do well?” It doesn’t matter that Pepsi preferences are irrelevant to law school. Insurance companies do this all the time.)

[totalidiot]

The LSAT has almost nothing to do with intelligence, and has even less to do with law school.

Any test that you can throw money at to do better on ceases to become a test of intellect.

oh boy, somebody's bitter-- we at tls are sorry that you didn't receive the score you wanted (maybe you should have thrown more money at it? like the 200 dollars most spend on prepbooks, but i guess that doesn't include the bribes most folks send the lsac in order to get the score they want, right?).

the lsat has a strong correlation to 1l grades.

First, I haven't received my score back, which I'm fairly optimistic about. I hope you were able to do better on assumption questions on the test, because you clearly can't spot you own.

I'm glad the LSAT correlates positively with grades. Show me that it is causal and you'll have an argument. If everyone who got A's on an exam drank a can of Pepsi the night before, I'm sure you'd be out there crafting an ad that says "Pepsi gets you A's!"

Oh boy again-- who says there's a causal link between LSAT and 1L grades? The correlation is what's important, because they are both considered to be products of the same cause, i.e. intelligence level. This is something the entire admissions system is telling you-- the LSAT is privileged because it has that correlation and is taken to imply a causal relationship between intelligence and LSAT.

[vvv4]

--

[Rocketman11]

Oh boy again-- who says there's a causal link between LSAT and 1L grades? The correlation is what's important, because they are both considered to be products of the same cause, i.e. intelligence level. This is something the entire admissions system is telling you-- the LSAT is privileged because it has that correlation and is taken to imply a causal relationship between intelligence and LSAT.

Uh, false.

First rule of Research Methods: Correlation does not imply causation.

[jrs12]

"Is taken to imply," is different than "implies."

[totalidiot]

"Is taken to imply," is different than "implies."

TiTCR

Rocketman-- First Rule of Reading Comprehension: Actually read what you're criticizing.

[izcanzbelawyrnow?]

The LSAT predicts IL performance about 70% of the time, and is also one of the best standardized tests for predicting performance.

Now let's watch as everyone fights over the remaining 30%!!!!!!!!!

[bigben]

The LSAT predicts IL performance about 70% of the time, and is also one of the best standardized tests for predicting performance.

Now let's watch as everyone fights over the remaining 30%!!!!!!!!!

Very wrong.

http://www.top-law-schools.com/forums/v ... =5&t=37902

[The_Wall]

The LSAT is barely an aptitude test. It's definitely not an intelligence test.

As for Secretary Rice, bear in mind that when she was applying to law schools (mid eighties?) the US News Rankings did not exist. The first year they did (1987, perhaps ~3-6 years later), Michigan was ranked third. Some would argue it was in fact the best, or perhaps second best, law school in the country in the early eighties.

I doubt very seriously the Secretary passed on law school because she couldn't get in.

[vasiok]

This has been discussed before, iirc. There is a very loose correlation between LSAT score and IQ, however there are so many different factors that play in that neither is even a vaguely reliable tool in predicting performance of the other.

TITCR.

[vicsinad]

Considering that scientists are always bickering about what intelligence actually means, and considering that it's one of those things that (in my opinion) is currently immeasurable, I don't see how there can be any meaninful correlation between overall intelligence and any standarized test. I guess the LSAT does measure, to an extent, your ability to reason through small written passages within a time constraint. But would there be a difference for those who are "more intelligent" at hearing the passages being read to them instead of reading them? Or if there were written responses to the questions instead of circling the correct answer?

I don't know. But I do think you can find a better correlation between LSAT/GPA combination and success in law school than you can between LSAT/GPA and overall intelligence. But I doubt it's as strong as the limited number of statistics they use to make this correlation.

[izcanzbelawyrnow?]

The LSAT predicts IL performance about 70% of the time, and is also one of the best standardized tests for predicting performance.

Now let's watch as everyone fights over the remaining 30%!!!!!!!!!

Very wrong.

http://www.top-law-schools.com/forums/v ... =5&t=37902

True, the source that I learned this from (my pre-law advisor) could be very wrong.

I would like to see where this information is referenced from before I admitted defeat - while I have noticed that my pre-law advisor has been out of touch with her advice at times, she's rarely so blatantly wrong. Even still, it wouldn't knock me over with surprise if she was off by such a gross amount.

[ianpantz]

I just wanted to highlight this. How does it feel to be the only person who doesn't think that money will get you a higher LSAT score?

I bet all these people buying preptests, buying PS bibles, paying for courses and tutors, are so misguided.

law of diminishing returns

Econ terms don't exactly lend themselves very well to arguments of measuring intelligence.

[izcanzbelawyrnow?]

The LSAT predicts IL performance about 70% of the time, and is also one of the best standardized tests for predicting performance.

Now let's watch as everyone fights over the remaining 30%!!!!!!!!!

Very wrong.

http://www.top-law-schools.com/forums/v ... =5&t=37902

True, the source that I learned this from (my pre-law advisor) could be very wrong.

I would like to see where this information is referenced from before I admitted defeat - while I have noticed that my pre-law advisor has been out of touch with her advice at times, she's rarely so blatantly wrong. Even still, it wouldn't knock me over with surprise if she was off by such a gross amount.

Interesting...

http://www3.interscience.wiley.com/jour ... 2/abstract

http://eric.ed.gov/ERICDocs/data/ericdo ... /70/a5.pdf

--LinkRemoved--

Looks more like 40%; predictive ability increases to about 50-55% in combination with UGPA.

Even further, it looks like predictability of the LSAT and the LSAT w/ UGPA varies amongst schools. So your LSAT score may predict your performance better at School A than at School B.

"For example, looking at the 2004 correlation study data, the average correlation between LSAT and FYA (edit: performance in first-year) is .32 for schools that have a negative correlation between the LSAT and GPA (edit: aka - those with more splitter numbers), but it increases to .47 for schools that show a positive correlation greater than .2 between those variables."

Soooo...yep.

[bigben]

You found it. I confess I don't know what a .4 correlation means exactly but the people in the other thread seemed to come to different conclusions than you.

[awesomepossum]

You guys will be happy to know that nobody gives a shit about the LSAT once you're in law school.

[Dead Ringer]

We could settle this debate by simply coming up with a judgment that the majority can agree on, which I will try to offer. I tutor the LSATS and can tell you that, on the level, the people who would strike you as intelligent after even a single conversation do much better than those who strike one as boring and dull witted. I have met probably a hundred students now and really can't think of a counter example. On the other hand, I play a lot of chess, and there I don't really see as strong a correlation between brightness and good play, which could indicate that chess tests a much more compartmentalized kind of intelligence than the LSATS, or at least a sort that doesn't conform to the lay definition of intelligence. Plainly, I think being "sharp" is necessary for success on the LSATs.

Also, if you think a subjective impression is useless for determining intelligence, it's based on a belief that there is very little difference between being smart and being able to act smart.

[edcrane]

The LSAT predicts IL performance about 70% of the time, and is also one of the best standardized tests for predicting performance.

Now let's watch as everyone fights over the remaining 30%!!!!!!!!!

Very wrong.

http://www.top-law-schools.com/forums/v ... =5&t=37902

True, the source that I learned this from (my pre-law advisor) could be very wrong.

I would like to see where this information is referenced from before I admitted defeat - while I have noticed that my pre-law advisor has been out of touch with her advice at times, she's rarely so blatantly wrong. Even still, it wouldn't knock me over with surprise if she was off by such a gross amount.

Interesting...

http://www3.interscience.wiley.com/jour ... 2/abstract

http://eric.ed.gov/ERICDocs/data/ericdo ... /70/a5.pdf

--LinkRemoved--

Looks more like 40%; predictive ability increases to about 50-55% in combination with UGPA.

Even further, it looks like predictability of the LSAT and the LSAT w/ UGPA varies amongst schools. So your LSAT score may predict your performance better at School A than at School B.

"For example, looking at the 2004 correlation study data, the average correlation between LSAT and FYA (edit: performance in first-year) is .32 for schools that have a negative correlation between the LSAT and GPA (edit: aka - those with more splitter numbers), but it increases to .47 for schools that show a positive correlation greater than .2 between those variables."

Soooo...yep.

The correlation coefficient is the square of the predictive value. .4 correlation does not equate to 40% "predictive value;" it equates to 16% predictive value. Even in schools where the coefficient is large (most likely where dispersion is significant), you're probably only looking at something that explains 36% of the variance in 1L GPAs.

More on point: the LSAT undoubtedly correlates to IQ--just like reciting a string of digits backwards correlates to IQ. Quickly ordering letters or distilling paragraphs of text into logical relationships takes mental horsepower. The problem with the OP is that it insinuates that high levels of intelligence are required for all sorts of high level success--I disagree. While there is a minimum amount of intelligence required to compose coherent sentences and write meaningful papers, I don't think having a 99th percentile IQ is a prerequisite for becoming a skilled defense secretary, etc. Raw computational power is likely to be overwhelmed by extraordinary diligence (a quality Rice is said to possess). Also consider Sarah Palin. She's obviously not that bright (SAT score ~900) and did horribly when she was unprepared; yet when you look at the old Alaska debate footage, she was easily able to keep pace and answer questions on the fly. (Obviously it's a different story when dealing with really brain-intensive pursuits like coming up with innovative math proofs and the like.)

[tome]

The LSAT is an apptitude test, not an IQ test. It measures certain apptitudes. As it happens, some of those apptitudes are also tested by IQ tests. However, it does not test everything that an IQ test tests.

So there is bound to be some correlation between LSAT scores and IQ scores.

I realize that this is a rather crude analysis, and that IQ =/= intelligence. But I think the relationship is equally crude, though definitely there.

[nicola.kirwan]

Interesting points. I have a question.

1) "LSAT is (or isn't) a strong predictor of 1L success." Okay...with respect to what? Given that law school grades are curved, and that every entering class has a different level of aptitude, are these studies saying that the higher the LSAT in comparison with the rest of the class, the higher the class rank?

Another question. Does anyone know if entering statistics are good predictors of law school success? For instance, if a student enters in the bottom 25% LSAT, do they basically stay there? And what about schools which range in the 160s, meaning that people are bright, but mostly not absolutely brilliant. Can you judge outright a substantial difference in ability between a 162 and a 168?

2) The poster who said that being "sharp" is required for success on the LSAT. I agree. As far as general measures of intelligence are concerned, one theory measures the "g" factor, which is basically the sheer speed with which an individual makes mental connections. Someone who is very sharp in this respect is undoubtedly intelligent, but I don't think that academia at least demands this type of brilliance. What standardized test cannot by nature test is capacity for understanding outside of certain time restraints. This seems to mean that it might work to compare test-takers, but above a certain score isn't predictive of whether a student is actually able to do the work. Like, does it really take a 175-180 to be a noted attorney at a BigLaw firm, or a highly competent government figure? Doesn't seem like it. At what point are extra LSAT points no longer really relevant in terms of what's actually needed to perform well? Is there such a point?

(I have a good friend who scored literally off the charts with her IQ, and in high school ranked third in the class next to two people who were smart, but nowhere near as brilliant. But the subject matter didn't require off-the-chart IQ and they seemed to have just as much of a capacity to wholly master the material as she did. Just wondering what this threshold is in law school)

3) While the LSAT itself might be a good predictor of intelligence (not concluding either way), those studies seem to assume that everyone comes to that test equally prepared. Perhaps that missing 30% (which actually seems like a fairly high number) for whom the LSAT is not correlative with their success is due to those who did not spend months preparing, who had other obligations, who may have reviewed the materials but more or less showed up to the test cold.