<p>The “list” included some colleges that aren’t top schools and doesn’t include some colleges that are. </p>
<p>What is the source of this list? Does anyone have a link?</p>
<p>And, I’d like to see a list of what the mean LSAT was for accepted law school students on these campuses.</p>
<p>25th-75th percentile LSAT scores at top 14 law schools:</p>
<p>1.Yale 169-177
2. Harvard 170-176
3. Stanford 168-172
4. Columbia 170-175
5. NYU 169-173
6. UC Berkeley 164-170
7. Chicago 169-173
8. Penn 166-171
9. Michigan 166-170
10. Duke 165-170
10. Northwestern 166-170
10. UVA 166-171
13. Cornell 166-168
14. Georgetown 166-170</p>
<p>Essentially, then, the median law-school-aspiring Harvard undergrad is going to be shooting for the bottom quartile at every top law school in the country, and not even coming close at the top two, YLS and HLS. From there it gets worse. In other words, top law schools are going to be reaches to uber-reaches for the majority of LSAT-taking students at every single undergrad school on the OP’s list. I don’t find this all that impressive a statement about any undergrad school. </p>
<p>Generally it’s the 170+ LSATs that are going to be highly competitive for admission to top law schools—and well above that at Yale and Harvard. These will be scattered around the country, a little more concentrated at the most selective undergrad schools, but in absolute numbers there will likely be more of them at a UC Berkeley or a UVA than at a Swarthmore or a Pomona. The median undergrad at any college, no matter how selective, won’t be close to that level.</p>
<p>Bottom line: don’t think that admission to or attendance at ANY college—Harvard, Yale, Swat, Princeton, Pomona, whatever—puts you in line for admission to a top-tier law school. It doesn’t. They take only the cream of the crop, and the median undergrad at these colleges just doesn’t rise to that level.</p>
<p>The admissions process for law school is also a mystery, shrouded in gpa’s, classrank, major, LSAT score, what race or gender you are (or aren’t), what is your “agenda”? i.e. career choice or non profit passion, what is your undergrad school, how much work experience you have (if any), how old you are, whether you served in the military or not, geographical location, where you want to practice law, whether you are likely a bigtime donor in years to come, recommendations/legacy/connections etc etc. </p>
<p>Yes, gpa, classrank and/or prestige of your undergrad institution, and LSAT are the big three. But not the whole picture. </p>
<p>There is no such thing as a slam dunk in law school admissions.</p>
<p>And any of the top 50 and maybe even top 75 law schools (out of 170 total) are perfectly good “outstanding” institutions. So restricting your sights to the top 10-15 is silly.</p>
<p>I would bet the farm that the average LSAT scores of the students in each school is an aggregate of the student’s SAT scores. That said, it is hard to make cross school comparisons with large schools because there is a much wider range of student abilities taking the LSAT.</p>
<p>A more interesting and more accurate means of com[parison would be to look at the scores within different majors within a school. I would bet that the gap would close very quickly if one were to compare engineering LSAT scores at Penn State, Georgia Tech and Harvard.</p>
<p>Is there a source similar to this for the GRE, GMAT, MCAT, etc…</p>
<p>LSAT Mean = 151
SD = 7</p>
<p>The top 25 schools listed on the previous page range from 161-165. Let’s choose one for each score value.</p>
<p>Georgetown 161
Rice 162
Dartmouth 163
Williams 164
Yale 165</p>
<p>Let’s convert these scores into Z scores and compare them to SAT Math mean scores (estimated by using the median since I can’t find mean) and see if these schools are under/over/matching input variables of students.</p>
<p>School LSATZ SATMZ Difference</p>
<p>Georgetown 1.43 1.64 0.21
Rice 1.57 1.81 0.24
Dartmouth 1.71 1.81 0.1
Williams 1.86 1.68 -0.18
Yale 2 1.94 -0.06</p>
<p>Now, if you look at these numbers you can see they’re all over the place. In fact, a linear best-fit only gets an R^2= .39.</p>
<p>Now either SAT Math is a bad predictor of LSAT score (not sure, but if both/either are considered good predictors of intelligence they should correlate), or there is a difference in what school you attend, and it’s not always small.</p>
<p>Either some of these schools are attracting students that are smarter than other schools (by quite a bit) by some measure that’s not SAT math, something is going on different once you’re in each of these schools which has an effect, etc, etc.</p>
<p>Importantly, for the most part, schools that are annually at the top (Yale, Williams, Dartmouth) appear to be doing better at matching inputs than other schools. This at least suggests that no harm is being done and smart students are just as smart going in as out when it comes to standardized test ability.</p>
<p>Of course, this is all rudimentary stupid math done on five data points, but I’d say that LSAT scores this “low” are not really blemishes on top schools.</p>
<p>
</p>
<p>[Average</a> LSAT Scores for 29 Majors with over 400 Students Taking the Exam](<a href=“http://www.uic.edu/cba/cba-depts/economics/undergrad/table.htm]Average”>Business | University of Illinois Chicago)</p>
<p>Though it doesn’t break it down by school, it does show some real interesting stuff which may actually suggest that schools heavy in certain areas may have an “unfair” boost in LSAT score averages.</p>
<p>oohh…thanks for link…glad to see that math/physics/philosophy majors statistically do the best on the LSAT. My DS1 is a math and physics double major with a philosophy minor. LOL</p>
<p>^^ Hmmm. “Prelaw” majors rank #28 out of 29 majors listed in average LSAT scores. Avoid that one like the plague if you want to go to law school.</p>
<p>
</p>
<p>Every school includes the losers. That’s why the LSAT means are so low.
Law school applicants are weeded out when pre-law advisers convince them that law is not a suitable or realistic field to pursue, or when they realize there are greener pastures in education, finance, medicine, etc. Thus, an equal share of applicants with both high scores and low scores will be eliminated by this natural process of self-realization.</p>
<p>
</p>
<p>Is that what the committee told you? ;)</p>
<p>
</p>
<p>Woman, we’re talking about admission to ANY medical school, admission to ANY law school. Not restricted to the top schools. Thus, a 95%+ chance of being admitted to ANY graduate school is a credible estimate.</p>
<p>
equal share? 
Hah, failure becomes me.<br>
Does it make you feel better knowing you have 95% chance of getting into a TTT? Those stats are meaningless. YLS or die.</p>
<p>
</p>
<p>Well, here’s one possible difference: the most selective schools (Yale, Williams, Dartmouth) are able to attract more students with top SAT M <em>and</em> top SAT CR scores. Most schools just a rung down (e.g., Georgetown) are going to attract a lot of students who are strong in either CR or M, but not both. Notice that this won’t necessarily show up in their US News-reported stats (or in the Common Data Set from which the US News data is drawn). That’s because they don’t actually ask for or report on the 25th/75th percentiles of combined (CR + M) scores. Instead, they report the 25th/75th percentile CR scores, and independently report the 25th/75th percentile M scores, then simply add the reported 25th percentile CR to the 25th percentile M, and the 75th percentile CR to the 75th percentile M. Which tells you — nothing, because a school’s top quartile CR scorers could all be below that school’s 75th percentile in M, and vice versa. In fact, it’s possible in principle that a school could have not a single student in its entering class whose actual combined (CR + M) score is above the number reported by US News as that school’s “75th percentile CR + M.” My guess, then, is that Yale has a lot more kids in the double-750 to double-800 range than a school like Georgetown, and that the difference is actually larger than one might guess from the US News-reported stats. </p>
<p>I can’t prove that’s what’s going on; just a hunch. But if it is, then yes, the students at Yale, Williams et al are indeed “smarter” (in a relevant sense) than those at Georgetown, Rice et al.</p>
<p>Why does this matter? I believe both math-like analytical and deductive reasoning ability and critical reading skills are called upon pretty heavily in the LSAT exam. That’s why philosophy majors do so well; the field demands both math-like analytical and deductive reasoning ability, and the ability to parse difficult texts. Thus it both attracts people with a combination of those abilities, and hones their skills while in college—preparing them well not only for the LSAT but also for law school.</p>
<p>Well, I could factor in CR but what you’re saying is that some schools may have an average CR+M score of individuals that’s lower than the average CR + average M score?</p>
<p>^ Well, I guess I’m not sure how this affects the “average” scores. That’s because you probably get a similar distorting effect at the 25th percentile level: US News gets its “25th percentile CR + M” score by adding the reported 25th percentile CR score to the reported 25th percentile M score. But very likely many of the low CR-scorers at any given school did significantly better in M, and vice-versa; most schools are not going to admit that many people who are low in both. (Thought experiment: do you really think Harvard admitted 1670 people—1/4 of its entering class—with actual (CR+M) scores below 1390, when they had thousands and thousands of stronger candidates to choose from? I don’t). Bottom line there may be nowhere near 25% of the class whose combined CR + M scores fall below the figure US News reports as the school’s 25th percentile CR + M. The US News-reported 75th percentile is too high, and the US News-reported 25th percentile is too low. These are really just garbage numbers that don’t tell you anything, and may be downright misleading.</p>
<p>On the other hand, when it comes to averages maybe it’s just a wash; maybe the actual average CR + M score is somewhere near the figure you’d get by averaging the US News-reported 25th and 75th percentiles—too high on the high end, too low on the low end, the errors cancel out when you average them and you might come out pretty close to what you’d get by averaging the real numbers. Maybe. But there’s just no way to know. </p>
<p>So go ahead & try it, see what happens.</p>
<p>^^ Also notice that even at the HYP level, the US News-reported “75th percentile CR + M” scores are probably too high—and for similar reasons, their reproted 25th percentile scores are probably too low.</p>
<p>Consider this: H, Y, and P all show up in US News with “75th percentile CR + M” scores of 1580. But the College Board says that among 2009 HS graduates there were only 2146 students in the entire country with actual (CR + M) scores that high. If H, Y, and P each have 1/4 of their freshman class at that level, that means almost half (49.3%) of all the entering freshmen in the country with scores in the 1580+ range are at HYP. That leaves Stanford, MIT, Caltech, the rest of the Ivies, Duke, the top LACs, the top publics, etc, etc. to fight over the other half. I have no doubt HYP have a lot of the top scorers, but to me it’s just not plausible that they have as many as half. No, that 1580 “75th percentile” figure is just a made-up number, the artifact of the goofball way US News calculates these scores.</p>
<p>You’re definitely right, bck, just not sure I have a way to aggregate the scores in a true sense. I’m not that good with statistics. This problem may be one reason why people almost always use one score as predictor. A quick way to confirm what you’re talking about is if there are statistics about the score of one section and how strongly correlated it is to the score in other sections. My guess is there is a weak correlation, meaning that many people do significantly better on 1 or 2 of 3 sections than the do on the remaining sections.</p>
<p>If I have time this weekend (and that’s a pretty big if), I’ll do my best to take as many of the LSAT scores on page one and see how strongly correlated they are to those schools’ SAT scores in each section. Then we can use far more data to get a far more accurate picture of how much, on the whole, LSAT is a measure of input variables, and if LSAT is a measure of input variables, which SAT section has the strongest relationship.</p>
<p>If the relationship is strong with any section, we can also look at outliers and speculate why these anomalies occur.</p>
<p>Well…comparing SAT math scores to LSAT scores is like comparing prunes to figs. I mean really. Neither is a complete measure of intelligence but only a measure of success on that particular exam on that particular day with those particular questions. Nor is LSAT necessarily a measure of success in law school or success in the practice of law, let alone how ethical you might be (or not be.) Its an imperfect instrument at best. </p>
<p>There are kids in law school with all sorts of majors, just like kids in medical school. Aptitude for what you are studying is very important, but that doesnt necessarily correlate to your undergraduate major or even the LSAT, GRE or GMAT or MCAT or whatever. </p>
<p>Just like the people who finish at the top of their law school class arent necessarily the best or smartest or most successful lawyers. </p>
<p>Admissions committees have to use some objective measurement, imperfect as they are, to decide who gets in and who doesnt. </p>
<p>Its amazing the people who are so obsessed with proving that the Ivy League is full of the best students and have the best programs and so forth and so on. If they do that in the real world, in real offices they will be in for a rude awakening. They are fine schools to be sure and I hope kids who want to go there get in. But I just don’t buy this attempt to measure their alleged superiority with SAT/LSAT/GRE/MCAT scores etc.</p>
<p>So what did you guys score in the LSATs??</p>
<p>ghostbuster, I think you misunderstand the exercise.</p>
<p>1) Of course they’re not complete measures of intelligence. However, they’re both tests which try to measure intelligence with roughly normal distributions. One previous comment made was that because the average scores listed for top schools would not necessarily get you into a top law school, these scores were unimpressive. My goal was to prove that due to the distribution of LSAT scores, these LSAT scores at different institutions were actually quite similar to the SAT scores of students at said institution when you compare both on a percentile-basis. The “z-score” is essentially how many standard deviation units students scored above the mean. If at Yale, students scored about 2 SD above the mean on one intelligent measure, assuming that this measure across thousands of random people is generally correlated with base intelligence even if for no individual it has meaning, then we’d expect on another exam with similar conditions (random sample, meaning particulars of individuals average out) the sample mean should be 2SD above population mean. What I found was a much weaker correlation than I thought, however, the differences between these measures are not super high and I need to get more data to determine if I happened to choose a couple of outliers by chance.</p>
<p>2) I don’t think this says you’re full of the best students or best future lawyers or anything. I just think that these average scores are still actually quite impressive. Hell, 165 is a full 2SD above the mean. To give a sense of how unlike it is to produce that result randomly, that’s like the 97th percentile of all scorers, assuming a normal distribution. Since when is 97th percentile unimpressive?</p>
<p>I am questioning because of 3 reasons:</p>
<ol>
<li>I remember seeing a very similar list few years ago.</li>
<li>Most research docs on the website are fairly old.</li>
<li>If these data are, say, more than a decade-old, it would explain why WashU has “underperformed”</li>
</ol>