LSAT/SAT correlation + realistic potential score?

User avatar
JCFindley
Posts: 1283
Joined: Thu Mar 01, 2012 1:19 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby JCFindley » Mon Jun 25, 2012 11:10 pm

JJJ123 wrote:1. A single anecdote is insufficient to establish a statistical generalization, and so the evidence is insufficient.
2. There are an abundance of good/plausible reasons for believing a correlation exists.

Paragraph 3 was not part of the argument, but was a reflection on the futility and masturbatory nature of this discussion.

Lol.


A single exception IS in fact enough to show the OP that while a correlation may exist, it doesn't have to dictate how any one individual does on the LSAT.....

cc.celina wrote:OP wasn't interested in establishing a generalization, he was interested in becoming an exception to it. Sometimes common sense is helpful. OP's question was not "masturbatory," he was simply looking for reassurance that he is not doomed.


This....

JJ, It would be helpful if you would move your musings to the "Math for attorneys" section of this forum.... Someone might actually care there.

User avatar
Scotusnerd
Posts: 813
Joined: Sat Oct 01, 2011 7:36 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Scotusnerd » Mon Jun 25, 2012 11:29 pm

JJJ123 wrote:1. A single anecdote is insufficient to establish a statistical generalization, and so the evidence is insufficient.
2. There are an abundance of good/plausible reasons for believing a correlation exists.

Paragraph 3 was not part of the argument, but was a reflection on the futility and masturbatory nature of this discussion.

Lol.


It's not futile, and it's not masturbatory. In my previous post in this thread (before JJJ123), you might have noticed that I dealt with the issues of the intense insecurity surrounding this test. People stake their livelihoods on this test. There's gonna be fear and apprehension, and rightly so, I think. It's important to separate the useful fear (as a motivator) from the useless fear (which makes you waste time on useless endeavors and energy on being scared).

Not everyone understands the difference, and it's important to understand it, if one is going to take this test successfully. Performance anxiety is a big thing on this test. I did not get the feeling from the OP that that was the intent, and it is certainly not futile to point out a fallacy that others prey on.

It's just another trap to avoid, like shitty study materials and bad instructors.

JJJ123
Posts: 17
Joined: Thu Jun 03, 2010 12:08 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby JJJ123 » Tue Jun 26, 2012 12:04 am

cc.celina wrote:
JJJ123 wrote:1. A single anecdote is insufficient to establish a statistical generalization, and so the evidence is insufficient.

OP wasn't interested in establishing a generalization, he was interested in becoming an exception to it. Sometimes common sense is helpful. OP's question was not "masturbatory," he was simply looking for reassurance that he is not doomed.


But this proves my point, if the OP understood what we mean by a correlation he wouldn't need to look for reassurance. He would know that he is not necessarily doomed based on a poor performance on the SAT alone, because a correlative relationship does not yield absolutely predictive results (it is not causal)... Also, the masturbatory comment was not directed toward OP's concern, but rather the subsequent debate regarding the scope of the correlation between the LSAT/SAT.



JCFindley wrote:
JJJ123 wrote:1. A single anecdote is insufficient to establish a statistical generalization, and so the evidence is insufficient.
2. There are an abundance of good/plausible reasons for believing a correlation exists.

Paragraph 3 was not part of the argument, but was a reflection on the futility and masturbatory nature of this discussion.

Lol.


A single exception IS in fact enough to show the OP that while a correlation may exist, it doesn't have to dictate how any one individual does on the LSAT.....

[quote="cc.celina"]

A correlation is not a perfectly predictive relationship. That is a causal relationship. No one ever argued that high/low SAT scores dictate (cause) high/low LSAT scores, nor is this at all necessary for a correlation between the two to exist...



People, why would you let the existence of a correlation between the SAT and LSAT bother you? If you want to know if you'll do well on the LSAT, look at a goddamn LSAT prep book. Take a diagnostic. Try your hand at logic games. Don't look toward a score on a test you took in high school and resign yourself to a similar fate. And while you're at it please please please look up the Fallacies of Division and Composition. It will help you resolve what you may view as a contradiction between this paragraph and my previous posts (and its a fallacy that you will see on the LSAT too).

TERS
Posts: 161
Joined: Thu May 24, 2012 11:29 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby TERS » Tue Jun 26, 2012 12:17 am

Clearlynotstefan wrote:The bare minimum time between taking one and the other is about 4 years, and very often more. People change A LOT in four years, these four years in particular.


The bare minimum time between taking the SAT and LSAT is not “about 4 years,” as one could take the LSAT during their first year of university (read between the lines).

The only people who take the LSAT already have completed college, already show the drive to at least try getting into grad school.


Tell me you’re fucking joking. Seriously that’s so false it makes me wonder if you really did quit smoking pot and drinking.

User avatar
PARTY
Posts: 420
Joined: Sun Jan 08, 2012 5:54 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby PARTY » Tue Jun 26, 2012 12:21 am

i've always scored in the same range on all of the standardized tests.

User avatar
cc.celina
Posts: 602
Joined: Fri May 25, 2012 1:17 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby cc.celina » Tue Jun 26, 2012 12:25 am

JJJ123 wrote:No one ever argued that high/low SAT scores dictate (cause) high/low LSAT scores, nor is this at all necessary for a correlation between the two to exist...

You're right. No one did. But OP asked if it was true. I appreciate your attempt to edify us all on your superior statistical knowledge, but it was really unnecessary to jump in with a sour attitude and claim that people are ruining threads. Everything you say is true. This makes you knowledgeable. It does not, however, render you likable, or helpful to the OP. You can disagree with someone (and be right) without being rude.

User avatar
PARTY
Posts: 420
Joined: Sun Jan 08, 2012 5:54 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby PARTY » Tue Jun 26, 2012 12:26 am

cc.celina wrote:
JJJ123 wrote:No one ever argued that high/low SAT scores dictate (cause) high/low LSAT scores, nor is this at all necessary for a correlation between the two to exist...

You're right. No one did. But OP asked if it was true. I appreciate your attempt to edify us all on your superior statistical knowledge, but it was really unnecessary to jump in with a sour attitude and claim that people are ruining threads. Everything you say is true. This makes you knowledgeable. It does not, however, render you likable, or helpful to the OP. You can disagree with someone (and be right) without being rude.



welcome to the fucking internet. jeez.

longagofaraway30
Posts: 11
Joined: Sun Jun 24, 2012 6:47 am

Re: LSAT/SAT correlation + realistic potential score?

Postby longagofaraway30 » Tue Jun 26, 2012 1:34 am

Thanks for all the input you guys, I really appreciate it.

And I didn't mean to cause a spat. I understand how this topic may seem "masturbatory" (kind of along the lines of "console me, I had a blah SAT") but basically I just wanted to be realistic, that is: I certainly think I could have done better on the SAT had I studied, but I don't realistically think I would have achieved a perfect score even if I had studied to death.

One of the specific issues I was really curious about is how much of the test is based on inherent aptitude/potential vs how much of it is "learn-able" - and I know this is a topic of constant debate. It's probably a combination of both. Obviously even if it was mostly about inherent aptitude one can still maximize one's potential by studying...to an extent. I'm devoting this entire summer to the LSAT and will take all advice about study materials, thanks!

User avatar
Clearly
Posts: 4165
Joined: Sat Feb 11, 2012 4:09 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Clearly » Tue Jun 26, 2012 2:11 am

TERS wrote:
Clearlynotstefan wrote:The bare minimum time between taking one and the other is about 4 years, and very often more. People change A LOT in four years, these four years in particular.


The bare minimum time between taking the SAT and LSAT is not “about 4 years,” as one could take the LSAT during their first year of university (read between the lines).

The only people who take the LSAT already have completed college, already show the drive to at least try getting into grad school.


Tell me you’re fucking joking. Seriously that’s so false it makes me wonder if you really did quit smoking pot and drinking.


Do you always approach assumptions as though the person making them is a total asshole? Might wanna work on that. Let me be more specific. The vast majority of people take the LSAT in "about four years". Especially accounting the people that took them far further apart, taking time off before starting college. I took the lsat this year, and the SAT in 2004. You show me data that shows more students take the lsat within say 2 years of the SAT then do in "about four years" and then you can nitpick on a generalization I was making. If you have to "read between the lines", you're not on a good path to establishing correlation.

As to the other point, I should have been more specific, most (95%+ I'd imagine) are on path to graduation. I would imagine very few people end up with an LSAT score and no degree.

Finally, back to my first point. I stand by the points I was making, perhaps my language should have been more specific, but we're on a fucking internet forum man, I don't stay up at night writing personal attacks to people trying to reassure someone that they aren't doomed because they did poorly on the SAT. Maybe you should go live a little before you turn into the nitpicking asshole stereotype lawyer that people cant stand. The last sip of drink I had was 5 years and a day ago asshole.

User avatar
Clearly
Posts: 4165
Joined: Sat Feb 11, 2012 4:09 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Clearly » Tue Jun 26, 2012 2:14 am

longagofaraway30 wrote:Thanks for all the input you guys, I really appreciate it.

And I didn't mean to cause a spat. I understand how this topic may seem "masturbatory" (kind of along the lines of "console me, I had a blah SAT") but basically I just wanted to be realistic, that is: I certainly think I could have done better on the SAT had I studied, but I don't realistically think I would have achieved a perfect score even if I had studied to death.

One of the specific issues I was really curious about is how much of the test is based on inherent aptitude/potential vs how much of it is "learn-able" - and I know this is a topic of constant debate. It's probably a combination of both. Obviously even if it was mostly about inherent aptitude one can still maximize one's potential by studying...to an extent. I'm devoting this entire summer to the LSAT and will take all advice about study materials, thanks!


The test is not an aptitude test. The closest thing is the ability to read and understand English critically, which I should hope if graduating college in this country, you can. The ability to detect patterns, and the dedication to study like a mad man is what the LSAT really tests. My diagnostic was 54th%ile, and my highest PT was a perfect 180. The vast majority of 170+ scorers did not diagnose at 170. You'll do fine. This test is a learned ability, believe that, and be ready to learn your ass off.

User avatar
Clearly
Posts: 4165
Joined: Sat Feb 11, 2012 4:09 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Clearly » Tue Jun 26, 2012 2:31 am

JJJ You are out of your mind. You realize there are degrees of correlation right? Any two events could have some correlation, we are discussing the variability of that causation. I provided examples to demonstrate that any correlation that exists is likely weak due to the infinite number of variables not being controlled, and those that by natural progression (aging) are practically guaranteed to effect the correlation. I didn't argue for NO correlation, I just said one is not sufficient to even attempt to predict the other. You also realize that your counter argument relies on disproving the weakness of the correlation by taking advantage of outliers that most statisticians would avoid even including in the equation to prove the strength of a correlation. Of course the top 1% of SAT takers would be likely to have a higher LSAT score (though not guaranteed), to use that as evidence to apply a correlation to the other 99% of test takers is exactly the type of logical bullshit you seem so convinced I've committed.

You seem to be countering my argument as if I implied 0 correlation. I merely contested both correlation and to the extreme, causation by providing several examples that weaken the correlation.



Thank you to those who came to my defense while I was away from the computer doing things besides addressing overzealous logicians who need to impose their extreme attitude on an internet forum post I made at 2 in the fucking morning. Understand two things that will determine your success as a lawyer: your ability to reason and process logical arguments and thoughts, and having a personality that doesn't make 8 strangers cringe every time you type a paragraph. Your halfway there.

TERS
Posts: 161
Joined: Thu May 24, 2012 11:29 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby TERS » Tue Jun 26, 2012 7:20 pm

Clearlynotstefan wrote: Let me be more specific.


I seriously just laughed out loud. I think you meant to say "let me correct my mistakes." How am I being nitpicky by pointing out that your statements are patently false? Nobody likes someone who can't admit when they're wrong. You should work on that, skipper.

JJJ123
Posts: 17
Joined: Thu Jun 03, 2010 12:08 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby JJJ123 » Tue Jun 26, 2012 7:51 pm

Clearlynotstefan wrote:JJJ You are out of your mind.


Uninspired ad hominem attack, I would have preferred something a little more like "JJJ has not had an original idea since 1990, and that was under prenatal influences." If you're going to go the ad hominem route, the least you could do is be a little clever.

Clearlynotstefan wrote: You realize that there are degrees of correlation right?


I loled. You clearly did not read/understand any of my previous posts. Each one assumed that the SAT/LSAT correlation was less than 1 , but I argued that it was still a significant correlation. And I find that my position has the greater rational basis, despite your vehement, emotional opposition.


Clearlynotstefan wrote: Of course the top 1% of SAT takers would be likely to have a higher LSAT score


...lol

Clearlynotstefan wrote: Understand two things that will determine your success as a lawyer: your ability to reason and process logical arguments and thoughts, and having a personality that doesn't make 8 strangers cringe every time you type a paragraph. Your halfway there.


Thanks for the advice on what makes a great lawyer, as a prospective law school applicant, you seem uniquely qualified to dispense such valued wisdom. Please share more. Indeed, I suggest that you should put your immense knowledge into practice: go take the bar exam and begin your career as a great lawyer, law school seems like an unnecessary stepping stone for someone like you!

Now let me give you a bit of advice. Arrogance has ruined far greater men than you or me. Be wary of that.


P.S. You shouldn't take internet forums as seriously as you seem to do.

User avatar
Clearly
Posts: 4165
Joined: Sat Feb 11, 2012 4:09 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Clearly » Tue Jun 26, 2012 8:24 pm

JJJ123 wrote:
Clearlynotstefan wrote:JJJ You are out of your mind.


Uninspired ad hominem attack, I would have preferred something a little more like "JJJ has not had an original idea since 1990, and that was under prenatal influences." If you're going to go the ad hominem route, the least you could do is be a little clever.

Clearlynotstefan wrote: You realize that there are degrees of correlation right?


I loled. You clearly did not read/understand any of my previous posts. Each one assumed that the SAT/LSAT correlation was less than 1 , but I argued that it was still a significant correlation. And I find that my position has the greater rational basis, despite your vehement, emotional opposition.


Clearlynotstefan wrote: Of course the top 1% of SAT takers would be likely to have a higher LSAT score


...lol

Clearlynotstefan wrote: Understand two things that will determine your success as a lawyer: your ability to reason and process logical arguments and thoughts, and having a personality that doesn't make 8 strangers cringe every time you type a paragraph. Your halfway there.


Thanks for the advice on what makes a great lawyer, as a prospective law school applicant, you seem uniquely qualified to dispense such valued wisdom. Please share more. Indeed, I suggest that you should put your immense knowledge into practice: go take the bar exam and begin your career as a great lawyer, law school seems like an unnecessary stepping stone for someone like you! ad hominem

Now let me give you a bit of advice. Arrogance has ruined far greater men than you or me. Be wary of that.


P.S. You shouldn't take internet forums as seriously as you seem to do.


Take at look at the last two pages, then tell me who takes it too seriously. :roll:

It has nothing to do with law at all actually, two key aspects to success at anything, being good at what you do, and not pissing people off along the way. I give you credit, at least you try to engage in actual discussion, you could be like TERS here.

The point is, the kid was worried he wouldn't do well on the LSAT, because he didn't do well on the SAT. To my knowledge none of you are crazy enough to agree with him. I merely provided examples in which one could expect to see significant variance between the two. As far as I'm concerned, everything after that turned into pissing contest. Truce?


@Ters. As I said, my language was too strong. However, I still stand by my points covering the vast majority of test takers. When you can prove a majority of test takers take the LSAT and the SAT 2 years apart, or that a significant number of LSAT takers never went on to complete an undergrad degree, I'll change my stance. In the mean time, the points I made still apply to the majority of applicants, and to dismiss them outright because I used language that was too strong is foolish.

Excellent117
Posts: 194
Joined: Fri Dec 23, 2011 4:44 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Excellent117 » Tue Jun 26, 2012 8:35 pm

That first formula in the OP fit me pretty well actually...

1520/21 + 101 = 173.38 (got a 174 on the LSAT)

That being said, this SAT-to-LSAT "formula" sounds like BS of the highest order. Might as well throw ACT scores in there for an uber-accurate prediction... :roll:

TERS
Posts: 161
Joined: Thu May 24, 2012 11:29 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby TERS » Tue Jun 26, 2012 8:45 pm

(this is directed at Stefan)

There's nothing to discuss about the points I mentioned. I was pointing out that a number of statements you made were false, not just "strongly-worded." I'm not taking a stance, I'm pointing out facts that contradict what you stated. Please understand this. I do not need to "prove" anything.

You said "the bare minimum" gap between taking the SAT and LSAT is "about four years." This is false. A less-strongly-worded version of this is not "the majority of individuals who take the SAT and LSAT take the latter about four years after taking the former." That is a completely different proposition.

Second, you stated that "the only people who take the LSAT already have completed college," which is false because many people have taken the LSAT without having completed college. A less-strongly-worded version of this proposition is not "the majority of individuals who take the LSAT have or will complete college."

I'm done with this.

JJJ123
Posts: 17
Joined: Thu Jun 03, 2010 12:08 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby JJJ123 » Tue Jun 26, 2012 8:56 pm

I fear that you're repeatedly missing the mark. Futile as it may be, I'll quickly respond.

Clearlynotstefan wrote: Take at look at the last two pages, then tell me who takes it too seriously.


Just because I clearly articulated and defended an argument, and rebutted objections to that argument, does not mean I took the discussion very seriously. In fact, I was not and currently am not personally or emotionally invested in this thread. I enjoy debating and I found that certain posters (including you) were employing faulty reasoning. I pointed it out, and then a bunch of you got very upset. I found it pretty amusing; now I see why people deliberately "flame" on this site (although that was not my intention).

Clearlynotstefan wrote: It has nothing to do with law at all actually, two key aspects to success at anything, being good at what you do, and not pissing people off along the way. I give you credit, at least you try to engage in actual discussion, you could be like TERS here.


Your absurd egoism has broken free from the reigns of good reason, and is now running amok all over this thread... You now purport to know the key ingredients to success in life- of any endeavor whatsoever. You must be infinitely wise. So, where did you acquire this knowledge? By what divine means was it bestowed upon you? If you have such understanding, why are you considering a career in law over all the other careers in which you could reach the pinnacle of achievement? Why not finance, academia, politics, etc.?

Refer to my last post if you want some actually useful advice.

User avatar
Clearly
Posts: 4165
Joined: Sat Feb 11, 2012 4:09 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Clearly » Wed Jun 27, 2012 12:11 am

Jesus Christ I'm done. Can't do it any more. It's common sense that being good at what you do will help your career in doing that thing. It's likewise common knowledge that people don't like assholes... You don't have to be a prophet to realize being unskilled and an asshole is likely to negatively impact ones career. For the record, I'm not addressing you specifically. I'm merely generalizing on assholes.

The point is, the fucking sat is not a predictor of ones ability to score on the LSAT. Break down my individual points to whatever degree you want the reality is each introduced a variable in support of that conclusion. Very few people take the sat and the LSAT in the same year, thus it is a variable that effects the predictive ability if the test. The LSAT does not test the same skills as the sat, again, weakening the predictive value of the sat to the LSAT.

If you agree with the above, then your attempts to invalidate those points based on my incorrectly wording the evidence is rather pointless correct?

You're totally right, neither of those statements is correct, I should have been more clear in the point I was getting at; that these differences effect the vast majority of test takers. I believe the number of people in defiance of one or both of my misstatements is sufficiently small to allow my conclusion to stand even though I over generalized or misrepresented the point. For example, my incorrect statement implies that everyone who takes the Lsat has completed or will complete college.... In reality, certainly some people take the last, then fail out, I'm merely claiming that the number is small enough to not completely invalidate the argument as a whole... I've gotta stop with this thread. We all seem to agree on the OPs question, this has gotten way out of hand over my poorly worded evidence that I believe even after adjusting for my over generalizing still stands strong enough to support my conclusion. No hard feelings here, let's agree to disagree on this one.

User avatar
Noblesse_Oblige
Posts: 44
Joined: Mon Feb 06, 2012 7:41 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby Noblesse_Oblige » Wed Jun 27, 2012 9:30 am

longagofaraway30 wrote:I know there's varying opinions on the correlation between SAT and LSAT scores, I've seen at least 5 different equations that claim to predict LSAT score based on your SAT (+/- 3-5 points), and people claim for the most part that they seem to work. I've also heard a few arguments against them - one is that it's just correlational and may just prove that some students do consistently well on standardized tests, while another is that such equations may not work well for students who prepared well for one test but took the other one cold, etc.

I haven't even taken the LSAT yet, and my undergrad gpa is "OK" - a 3.5, yet I don't think going to law school in this economy is even worthwhile unless I get into at least a top 30 with some scholarship money.

I took the SAT years ago when it was out of 1600 (the last year before the change, 2005), and I received a mediocre 650 verbal, 600 math. Did I study? Not really. I DID take the PSAT sophomore year before taking the SAT junior year just once. But I think altogether I only devoted about 2 hours to studying the SAT in all of high school (and even that amount of studying was done kind of carelessly).

Based on one equation:

1250/21 + 101 = 160.52

I need at least a 167 to have a safe shot at getting into a top 30 school. But based on those equations..I have no chance. My question is for the people who do buy into such equations - is the LSAT learnable, at least to a certain extent? Is there any realistic chance at all I could get a 167, if I devote 2 months to studying the LSAT and considering that my SAT score was achieved after little preparation? People seem to imply as much, but I want to be realistic here.


The SAT is learnable too. What this tells me is that you probably Diagnosed around 160

I'm sure if you studied for the SAT you would have done better. Just like if you study for the LSAT you will do better.

woeisme
Posts: 781
Joined: Mon Jan 07, 2008 8:39 pm

Re: LSAT/SAT correlation + realistic potential score?

Postby woeisme » Wed Jun 27, 2012 9:51 am

longagofaraway30 wrote:Based on one equation:

1250/21 + 101 = 160.52


This has to be affected my study prep. If you got a 1250 with minimal study and are suggesting you'll get a 160-61 with minimal study, then you're in a really good place. Just study and you should be able to get your 167.




Return to “LSAT Prep and Discussion Forum”