<p>I'm entering college next year at a top state school, and I really want to go to a top law school, but I'm concerned that I won't be able to do well on the LSAT. I got a 1450 (750M, 700V) on the SAT, which is at the 98th percentile, but apparently the LSAT is a significantly harder test on which to score at or above the 98th percentile (about a 170). For instance, the average score for Harvard graduates is 166, and the average SAT score for Harvard students is around 1490. </p>
<p>Does the fact that I got a 1450 mean that I am unlikely to score above 166? I know that the LSAT and the SAT are different, but they seem to test very similar concepts. If there is any long term LSAT preparation I can do over the next four years, please let me know! My biggest fear is that I work my butt off to get a 4.0 but only score in the low 160s on the LSAT, dooming my chances at a top law school.</p>
<p>Also, I would really appreciate it if people would be willing to share their SAT and LSAT scores.</p>
<p>In the aggregate, LSAT scores and SAT scores are very tightly linked. At an individual level, however, standard deviations are noticeable and substantial.</p>
<p>In any case the formula I’ve seen – which is a generalization and subject to deviation – is SAT/21 + 101. That would project you out to about a 170 or so.</p>
<p>While I do know that correlations usually only apply to groups, it still seems to me that very high LSAT scores require sheer aptitude and that LSAT scores should be quite similar to SAT scores, even at the individual level, since the SAT and LSAT are basically an IQ tests. </p>
<p>The formula does seem a bit off, however – it would predict that the average LSAT for Harvard graduates would be 172, when the actually average is only around 167. So I guess 165 would be a pretty good estimate for me. </p>
<p>What do you think would be the best way to maximize my potential during the next four years? What sort of classes help students prepare for the LSAT?</p>
<p>One more totally unrelated question: Do Hispanics with South American ancestry (e.g. Columbia) get any special treatment in law school admissions? Or is affirmative action limited to blacks, Native Americans, Mexicans, and Puerto Ricans?</p>
<p>Or it’s explained by Harvard undergrads systematically regressing to the mean a little bit compared to comparable students with similar scores, which I think is very plausible.</p>
<p>As you said, the LSAT is basically an IQ test. There’s no long-range planning that I’d bother with.</p>
<p>Colombians don’t get AA benefits, ostensibly. Not sure how it works in practice.</p>
I think this is a little bit off. It’s definitely easier to get a 1450 on the SATs than a 170 on the LSATs. A lot of kids could get a near perfect score on the SATs without studying at all, but the highest start point for LSATs is usually 165-170 at best.</p>
<p>
That’s not true at all. The general consensus is that it takes about 6 months of hardcore (4-5+ hours a day) studying to peak for the LSAT, subject to diminishing return. There are sooooo many people who improved their scores by as much as 15-20 points.</p>
<p>1.) Even if you’re right, the OP means “long term” in years, not months.</p>
<p>2.) Six months is way too much. I don’t know what kind of consensus you’re referring to; certainly no consensus this board would remotely agree with that. Six months at five hours a day is a surefire recipe for burnout and not just diminishing returns but actually a diminishing score.</p>
<p>I’ve been saying that it takes six to eight weeks; some folks on here have been arguing with me that it takes as much as three months. I don’t know anybody who would advocate six months. That’s just insane.</p>
<p>
</p>
<p>Look, my score went up nine points in three weeks. I know people improve (although 20 point spikes are extremely rare, there’s a high <em>number</em> of people who see that). But the magnitude of the expected spike doesn’t mean you should kill yourself for six months on what is essentially a six-week test.</p>
<p>And it certainly doesn’t mean the OP should be prepping years in advance, which was what he was <em>actually</em> asking.</p>
<p>Also I said it takes 6 month to peak, subject to diminishing returns. That means maximizing your average score by 1-2 points or less after taking it infinite amount of times. 2-3 month of hardcore studying will be enough and practical for most, which translate into a semester+ a summer of semi-hardcore studying.</p>
<p>Okay, I figured out the problem with bluedevilmike’s formula. It’s supposed to use pre-1995 SAT scores, so you basically need to subtract 100 points from your SAT score for the equation to work.</p>
<p>It works perfectly for Harvard students: (1490-100)/21 + 101 = 167.19
The actual average is 167, so it’s damn near perfect, at least for large numbers of students.</p>
<p>Individuals obviously vary more than groups, but I’m sure that the formula is a pretty accurate predictor of LSAT performance.</p>
<p>Mine would be about ~165: (1450-100)/21 + 101 = 165.29</p>
<p>I would guess that the overwhelming majority of people would fall within a +/-5 point range of the score predicted by this formula.</p>
<p>Hm, interesting. I’ve always found that the formula in present form worked pretty well among my friends; it would logically have to be deflated to correct for kids who studied very hard for the SAT (e.g. Harvard types).</p>
<p>Again, I’m not sure if these are pre-1995 SAT scores; my guess is that they are.</p>
<p>However, the math scoring changed very little in the 1995 overhaul, so I think we can use the math section as a fairly accurate predictor of LSAT performance. Interestingly enough, the math SAT section correlates better than the LSAT than does the verbal section.</p>
<p>Yeah, I guess that makes sense. The LR and AR sections seem to require skills more commonly associated with math than with reading. It’s funny though, the Verbal section of the SAT is considered the more g-loaded (i.e. it correlates more strongly with IQ) part of the SAT.</p>
<p>I will point out that browsing the rest of the geocities site shows that it includes lots of data from around 2004-ish. I suspect the SAT scores used are post-1995, but not sure.</p>
<p>I did pretty crappy on the writing portion of the SAT (660), but I got a 1520/1600 besides that. And that was with no studying courses, just looking over a few ‘Crack the SAT’ books (I only took it once, early junior year).</p>
<p>That range puts me in the 171 (169-173) range. But hopefully I can study a lot this time, and improve by a few points.</p>