<p>Anyways I'm a senior in high school who plans on going into law school after college... I know it's very early to start thinking about the LSAT but I am curious. Is there a strong correlation between what a kid scores on the SAT and what he scores on the LSAT? If there is a correlation what would a person who scored in the 95th percentile on the SAT (1380 math and reading) score on the LSAT? thanks for your time...</p>
<p>Your two questions do not make sense as a series.</p>
<p>A correlation simply means that a strong SAT score is one of many factors that are involved in predicting an LSAT score - as a trend, kids who tend to do better on the SATs tend to do better on the LSATs all else being equal, but all else is NEVER equal.</p>
<p>To be able to generate a complete prediction from an SAT score, however, it would have to be the ONLY factor.</p>
<p>Well since my d's SAT scores were just at the average range as per Mr. Payne's chart (710 & 708 avr SAT/ per chart), we'll see if she gets that avr. range LSAT of 168. I think she'd be very content with that score. I'll know more in a few weeks as she is planning to take the exam in Dec.</p>
<p>I just don't see the connection between the SAT and the LSAT: both tests test different aspects of one's mental aptitude. Any website that purports to give an analysis of the relation between the LSAT and SAT will find no such thing; correlation does not mean that there is a relation, it just means that for some reason or another, XYZ LSAT score seems to be concomitant with WXYZ SAT score.</p>
<p>This should be fairly obvious; I mean, philosophers knew about it back in the 18th century:
<a href="http://www.amazon.com/Treatise-Human-Nature-Oxford-Philosophical/dp/0198751729/sr=8-1/qid=1163790189/ref=pd_bbs_1/103-7080357-5903027?ie=UTF8&s=books%5B/url%5D">http://www.amazon.com/Treatise-Human-Nature-Oxford-Philosophical/dp/0198751729/sr=8-1/qid=1163790189/ref=pd_bbs_1/103-7080357-5903027?ie=UTF8&s=books</a></p>
<p>(Another reason why philosophy majors tend to score higher: they don't look for garbage like correlation to feed false hope. Heck, we learn the informal fallacies and the problem of inductioin during our first year.)</p>
<p>Edit: As Gregory House said to a doctor who claimed that his medicine prevents migranes:
House: M'am, do you have cancer?
Random Lady: Uhhh... no.
House: (Points at the lady's orange juice bottle) Ladies and Gentlemen, orange juice prevents cancer!</p>
<p>Correlation may not imply causation, but the original question had to do with correlation.</p>
<p>And correlation does imply correlation.</p>
<p>I agree wit BDM, as individual mileage may vary.</p>
<h1>17 got a 1200 on the SAT and a 170 on the LSAT where #S 18 & 19 had a 1180/170 SAT and both scored 162 # 35 scores in the 1300s and got a 154 lsat</h1>
<p>
[quote]
but the original question had to do with correlation.</p>
<p>And correlation does imply correlation.
[/quote]
</p>
<p>No, duh.</p>
<p>But reread my post: (I think common people just have a tendency to think that when someone argues against correlation, it must have something to do with causation. It sounds cool, I know) Not once did I mention that there is no basis for causation - that much is obvious. What I did argue is that there is no basis for any relationship between the SAT and the LSAT, thus making any correlation between the two completely otiose. You are answering the OP's question, sure, but I am unsure what anyone in this thread has to gain from it, including the OP.</p>
<p>Oh, and before someone argues that "correlation is a relation so there must be a relation," I am arguing that there is not even any correlation, if you use proper reasoning. Used loosely, the "correlation" relationship can be applied to anything. </p>
<p>If any decent correlation is to be ascribed, there must be more credible data. Some random geocities site, unless the standards of scholarship are being defecated on, just won't cut it. And even if you do prove a decent correlation, it means nothing, unless you want superficial analysis to help massage your ego.</p>
<p>The LSAT tests for a lot of the same things the SAT does; which is mainly how smart you are on a certain set of desired metrics versus everyone else who takes the test.</p>
<p>Since the LSAT selects for college grads, getting a higher percentile score is more difficult than on the SAT.</p>
<p>Since the LSAT is also designed to create a normal distribution of the students who take it, it is "more" difficult in order to differentiate the stronger students at the top.</p>
<p>A lot of the test-taking skills which cause one to perform well on the SAT are similar to those on the LSAT. So you can use the correlation between scores to give yourself an idea of your LSAT score given what you got on the SAT. Holding all else equal, your percentile score on the LSAT should be a little lower than on the SAT.</p>
<p>Nspeds must've bombed his SAT or something to be so defensive about any correlation between the LSAT and SAT.</p>
<p>
[quote]
Nspeds must've bombed his SAT or something to be so defensive about any correlation between the LSAT and SAT.
[/quote]
</p>
<p>Yes... a 1490 is really low.</p>
<p>
[quote]
The LSAT tests for a lot of the same things the SAT does; which is mainly how smart you are on a certain set of desired metrics versus everyone else who takes the test.
[/quote]
</p>
<p>...too bad the metrics for each test are different.</p>
<p>nspeds,</p>
<p>The OP wants to know two things. The second question, I agree, is a strange question to ask and relies on a misunderstanding of statistics.</p>
<p>The first question, however, is not only very reasonable but very easy to answer.</p>
<p>Your discussion has me completely bewildered at this point. So here's what I'll say:</p>
<p>1.) I concede that the link above does not answer the question - i.e. it does not prove correlation.
2.) I also concede that the SAT does not "cause" anybody to do well on the LSAT.
3.) Remember, correlation simply means: "If I have some information about X, do I have some information about Y?"
4.) The answer in this case is obviously yes. If we know that somebody scored very highly on the SAT, we can say that all else being equal, he is likely to have a higher score on the LSAT than somebody with a lower SAT score.
5.) Notice the explicit qualifiers in the above sentence: certeris paribus, likely, etc.
6.) I agree that this population correlation is not established by the geocities link, since he has not calculated a p-value for us.
7.) The correlation within his data set, however, is clearly present.
8.) I will argue, however, that the correlation is absurdly obvious. Basic multiple choice and reasoning skills are highly involved on all standardized tests, not to mention scholastic achievement in general. I would go so far as to state that SATs are strongly correlated in MCAT scores, GMAT scores, etc. (The link to GRE scores is obvious enough.) If only because multiple choice tests involve process of elimination, quick reading comprehension, bubble filling, stress management, time management, etc., the correlation between the tests is obvious.
9.) If you want to suggest that the metrics between the SAT and the LSAT are totally different and don't overlap at all, then the burden of proof is clearly on you to show this.</p>
<p>
[quote]
9.) If you want to suggest that the metrics between the SAT and the LSAT are totally different and don't overlap at all, then the burden of proof is clearly on you to show this.
[/quote]
</p>
<p>That is certainly not the case. There are metrics, which you mentioned, that both tests utilize in order to measure aptitude; however, those metrics form the penumbra of the system on which the tests operate. The core is entirely different; indeed, the difference should be salient merely because the they test two different things: the SAT tests how a student will likely perform in his first year of college, while the LSAT tests law school performance. </p>
<p>Being more specific, though, I think the LSAT uses the principles and problems of logic to measure one's analytical abilities. This aspect is simply not the case with the SAT: for the SAT, you can simply memorize words and know some basic mathematical formulas in order to master the test. It is a process of memorization and practice. There is no such memorization for the LSAT, which should be indicative of something; indeed, it is indicative of the sheer mental prowess required to perform well on the test. The test does not ask you to fill a blank or connect two words or comprehend some passage (though the latter is definitely involved): rather, the LSAT wants students to use correct principles of logical reasoning in order to arrive at a conclusion about a given state of affairs whether the state of affairs is defined via rules and atomic variables in logic games, or a given scenario in the logical reasoning, or through some scholarly excerpt in the long passages.</p>
<p>The SAT doesn't even come close to measuring one's analytical abilities. If there is any valuable correlation between tests that could be drawn, it would be between, in my opinion, the SATs and the GREs. Do you want to compare the LSAT to another test? Try IQ tests. They, if I am not mistaken, utilize questions that put one's analytic abilities to the test. </p>
<p>It is 1 in the morning here. My point is that because the testing methodologies of both tests are different, no substantive correlation can be drawn. Sure, there are time-constraints, ability to operate under pressure, and so forth, but if we want to draw correlations based only on those metrics, then why not argue for a similar relation between the LSAT and driving tests? It is obvious that while such metrics constitute necessary conditions to the test, they do not in any way embody the principles of each test.</p>
<p>Edit: When I finished the SAT, I was not tired because the test was hard. I was tired because of the stress. Whenever I finish an LSAT, though, I am not tired because of stress: the test is fun, for me. Rather, I am exhausted because of the sheer mental energy involved in doing the test. In fact, I found that with the SAT, I had to reason a little more stupidly (akin to shiboing boing) in order to answer the questions correctly. With the LSAT, you need every ounce of mental power. The LSAT is demanding, and it gets the machine pumping.</p>
<p>lol ok we don't have to get so enthusiastic in explaining the difference</p>
<p>As someone who did well on both, and thus "mastered" what each was testing for, I found both of them to be remarkably similar. The LSAT was actually a little easier to me, because I didn't have to memorize anything and only had to increase my problem-solving speed. I can see why the LSAT would be hard for other people because the games-section would take people, that aren't very good at math or logic, longer to finish. Plus, reading speed is very much dependent on inherent ability to be able to process words and ideas quickly. These things are not easily studied for, unlike the SAT where the problems were relatively simple in comparison and a dedicated work ethic could allow you to memorize all the vocabulary and math properties you needed to know.</p>
<p>So the LSAT and SAT are different in no more than a few ways, but on balance, people tend to score a little lower in percentile on the LSAT than the SAT. That's my intuitive feeling and anecdotal observation from all my 165+ scoring friends.</p>
<p>I can barely read Nspeds posts. It's so obtuse and circulatory that its very hard to suspend your disbelief. Using Ocham's Razor, I don't think any of his claims hold a lot of merit.</p>
<p>
[quote]
I can barely read
[/quote]
</p>
<p>I think everyone on this forum can just leave your statement at this.</p>
<p>
[quote]
So the LSAT and SAT are different in no more than a few ways
[/quote]
</p>
<p>That doesn't refute my argument. My argument was that those few ways preclude any correlativity. </p>
<p>
[quote]
Using Ocham's Razor, I don't think any of his claims hold a lot of merit.
[/quote]
</p>
<p>That doesn't make sense: you probably don't know what parsimony is and how it works, and you don't even know if Occam's razor is true. You're a bloody inferior looking to insult people. You're not offending me: as I have said to many others on this forum, intellectual inferiors cannot insult me. What you are doing is humoring me, as well as exasperating me because of your stupidity. Don't get us both banned because of your inability to read or form a coherent argument.</p>
<p>For your disabuse on Occam's razor, I highly recommend the following article:
Robert Nozick, "Simplicity as Fall Out," *Socratic Puzzles<a href="Cambridge:%20Harvard%20University%20Press,%201997">/i</a>.</p>
<p>Take this from a guy who has done more research in the area and is more intelligent than you: you don't want to argue with me on this matter.</p>
<p>
[quote]
lol ok we don't have to get so enthusiastic in explaining the difference
[/quote]
</p>
<p>LIKE, LOL, HAHAHAHa, OMG, LIKE !!! LOL!, YOU"RE SO RIGHT, LOL!</p>
<p>OP,
Unless you are curious for some reason other than predicting your own score, you can just download the free practice test and see how you do. Or you can buy a book of 10 old tests and do several.</p>
<p>
[quote]
Unless you are curious for some reason other than predicting your own score, you can just download the free practice test and see how you do. Or you can buy a book of 10 old tests and do several.
[/quote]
</p>
<p>Yes, exactly. That is far more useful than looking for some dubious relation.</p>
<p>The answer to the OP's question can be found in the graphs that Mr. Payne posted. I'm no statistician, so maybe someone who is can explain in more detail, but the R-squared figures of .36, .29, and .25 tell you the corollation between SAT and LSAT. While there is a corrolation, those figures are statistically relatively low.</p>
<p>I'd be more interested in seeing the data run as logarithmic data anyway. It might compress the issues of scale a bit.</p>
<p>Either way, as MarathonMan said, the r^2 values are just far far far too low to be of any value. The site just shows how poor the correlation is.</p>