Better 25th% at CLS or 75th% at Umich Forum

[gmreplay]

According to the LSAT scores of admitted students, Michigan's 75th percentile is the same as Columbia's 25th percentile. Given that the LSAT is (erroneously?) considered to be a good predictor of law school performance, should I presume that my 170 means I would do quite well relative to my fellow Michigan students, and quite poorly compared to my CLS peers?

Regardless, is it better to be at the 75th percentile grades wise at Michigan, or the 25th percentile at Columbia?

[jcunni5]

No, the schools are so close in student quality and a couple points on the LSAT is not going to help you outperform your peers. So if you choose Mich, don't go b/c you think you will automatically be ranked higher in the class. With that being said 75 percentile at UM >>> 25 percentile at CLS

[Other25BeforeYou]

According to the LSAT scores of admitted students, Michigan's 75th percentile is the same as Columbia's 25th percentile. Given that the LSAT is (erroneously?) considered to be a good predictor of law school performance, should I presume that my 170 means I would do quite well relative to my fellow Michigan students, and quite poorly compared to my CLS peers?

Regardless, is it better to be at the 75th percentile grades wise at Michigan, or the 25th percentile at Columbia?

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Assuming you will be in any particular percentage of your class based on your LSAT score is extremely unwise when considering which school to attend (even if you were a 180 planning to attend a T4). Writing a law school exam has very little in common with taking the LSAT.

[d34d9823]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

[Alex-Trof]

I actually had a similar question for current students. Do you find that people who got higher LSAT scores typically perform better than those that got lower scores? Is a person with below medium LSAT score more likely to have below medium class rank?

[Patriot1208]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

No, it's not.

[BruceWayne]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

Learn common sense. Yes that's a relevant correlation in the world of statistics; it is not good in a real world predictive sense (If I told you had a 20 percent chance of living if I pushed you in front of a bus, would you call that "pretty good"?). Besides 0.4 is for LSAT and GPA combined--not LSAT alone. If you think getting a 170 means you'll do well at Michigan and you'll do poorly at CLS (because there median is a 172 or whatever) you need to get off autoadmit and TLS for a while.

[thesealocust]

I actually had a similar question for current students. Do you find that people who got higher LSAT scores typically perform better than those that got lower scores? Is a person with below medium LSAT score more likely to have below medium class rank?

lol no. It's all over the board.

Especially at top schools where everyone (a) is within a narrow band of LSAT scores and (b) that narrow band is already missing just a handful of question per section, there is a barely perceptible difference. Of the people that I know both LSAT and law school GPA information about, there is no correlation whatsoever.

And the '.4 correlation' number thrown about is for uGPA + LSAT, and it's looking at all law schools. Many schools have very wide LSAT bands. If you get a full ride to cooley, I'd bet my socks you'll do better. But at most law schools in the country, within the population at the school, there will be no worthwhile difference in performance based on individuals.

[BruceWayne]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

No, it's not.

It's been a while but in his defense I took an upper level stats class in undergrad and I do remember that that's the level at which they say "statistically significant" or something of that nature. I hated that class.....the point though is that in the world of statistics they aren't looking for a huge one to one ratio when they say "statistically significant". That's what he's trying to pass it off as.

[Patriot1208]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

No, it's not.

It's been a while but in his defense I took an upper level stats class in undergrad and I do remember that that's the level at which they say "statistically significant" or something of that nature. I hated that class.....the point though is that in the world of statistics they aren't looking for a huge one to one ratio when they say "statistically significant". That's what he's trying to pass it off as.

I've taken three statistics classes, I understand it would be considered relevant. But, a regression analysis that gives us a .4 correlation is one that i'm not using as evidence for, pretty much anything other than I need more variables.

[Alex-Trof]

I actually had a similar question for current students. Do you find that people who got higher LSAT scores typically perform better than those that got lower scores? Is a person with below medium LSAT score more likely to have below medium class rank?

lol no. It's all over the board.

Especially at top schools where everyone (a) is within a narrow band of LSAT scores and (b) that narrow band is already missing just a handful of question per section, there is a barely perceptible difference. Of the people that I know both LSAT and law school GPA information about, there is no correlation whatsoever.

And the '.4 correlation' number thrown about is for uGPA + LSAT, and it's looking at all law schools. Many schools have very wide LSAT bands. If you get a full ride to cooley, I'd bet my socks you'll do better. But at most law schools in the country, within the population at the school, there will be no worthwhile difference in performance based on individuals.

Thanks!

[bender18]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

It's actually more like .35

Going off of what other people have already said, the .4 must be LSAT and GPA combined

[Patriot1208]

The LSAT isn't considered to be a good predictor of law school performance. It's the best predictor, but it's still a really shitty predictor.

Learn 2 statistics. 0.4 is pretty good.

It's actually more like .35

Actually, I believe the median correlation (it varies by school) is .33 for the LSAT alone and something like .45 for UGPA + LSAT.

[thesealocust]

Google: Know it, use it, love it.

http://www.lsac.org/jd/pdfs/LSAT-Score- ... rmance.pdf

[Knock]

.35 correlation:

--LinkRemoved--

[Patriot1208]

Google: Know it, use it, love it.

http://www.lsac.org/jd/pdfs/LSAT-Score- ... rmance.pdf

Looks like my numbers were right, using the 2007 data.

[Knock]

.35 correlation:

--LinkRemoved--

.45 correlation:

--LinkRemoved--

[Cupidity]

I answered 3 more multiple choice questions correctly --> Therefore, I shall be better than 50% of the students at Columbia

[JG Hall]

I actually had a similar question for current students. Do you find that people who got higher LSAT scores typically perform better than those that got lower scores? Is a person with below medium LSAT score more likely to have below medium class rank?

Or, just don't talk about your LSAT scores with others. Seriously.

Edit: and asking other people their grades is pretty fucking awk too.

[Cupidity]

I actually had a similar question for current students. Do you find that people who got higher LSAT scores typically perform better than those that got lower scores? Is a person with below medium LSAT score more likely to have below medium class rank?

Or, just don't talk about your LSAT scores with others. Seriously.

Edit: and asking other people their grades is pretty fucking awk too.

Kid I know with a 170+ on our T-30 w/ full scholarship scored far lower than my 166

[d34d9823]

I've taken three statistics classes, I understand it would be considered relevant. But, a regression analysis that gives us a .4 correlation is one that i'm not using as evidence for, pretty much anything other than I need more variables.

Actually, in a complex system like this, a correlation that strong is a huge predictor (since even with all possible input variables included, you typically have 0.3-0.6 of the regression going to random noise). If this was my job, and we could control an input that I showed to be correlated at 0.3, I would get a raise. That said, I don't really have an issue with people who understand statistics and don't find this very meaningful. If we had to, we could calculate confidence intervals for 1L grades given an LSAT score to quantify exactly how significant the effect is.

My issue is with the people who don't have a statistics background and dismiss it as a significant correlation based on some nebulous reason. They seem to be in love with anecdotes about people who performed differently than their LSAT score. No shit, look at what a 0.3-0.4 looks like. See all the outliers? That says nothing about the strength of the correlation.

[helloperson]

Between Michigan and CLS the answer is that there isn't a significant difference in the caliber of student that you could predict such a huge difference. I think it would more closely resemble this: at michigan you have a 75% chance to fall between the 15th and 90th percentiles, whereas at Columbia you might have a 75% chance to fall between the 5th and 80th percentile. Certainly nothing you should be picking your schools with. That said, if you scored a 175 and you're attending a school with a 163 median LSAT score, I think your odds improve significantly and maybe you would have a 90% chance of being in the top 30%.

All of this is pure conjecture, though.

[Patriot1208]

I've taken three statistics classes, I understand it would be considered relevant. But, a regression analysis that gives us a .4 correlation is one that i'm not using as evidence for, pretty much anything other than I need more variables.

Actually, in a complex system like this, a correlation that strong is a huge predictor (since even with all possible input variables included, you typically have 0.3-0.6 of the regression going to random noise).

This is where your assumption dies. This isn't a complex system. It's a fairly simple analyiss based on 1 or 2 variables. If you were an economist that was studying all the variables in consumer preferences, i'd agree with you. Since you are taking one to two variables and simply comparing them to class rank, a .33 correlation (according to the LSAC article) is not very good.

[dabbadon8]

Think of it in terms of percentiles, when your talking about LSAT scores that high you are talking about the 99th percentile. I don't think the LSAT can be that predictive when your already so far right on the bell curve. Also, Imagine an applicant who took the LSAT 3 time after a full year of studying and got a 162, 167, 174, and then some one who took it after a month of prep got a 169 and called it a day. Both had a 3.8, one is going to UMich another to columbia. I couldn't tell you who I'd rather compete against. I can tell you that I am going to umich and am 2 points over their 75th from last year and I am worried about competing against such a qualified pool.

[d34d9823]

I've taken three statistics classes, I understand it would be considered relevant. But, a regression analysis that gives us a .4 correlation is one that i'm not using as evidence for, pretty much anything other than I need more variables.

Actually, in a complex system like this, a correlation that strong is a huge predictor (since even with all possible input variables included, you typically have 0.3-0.6 of the regression going to random noise).

This is where your assumption dies. This isn't a complex system. It's a fairly simple analyiss based on 1 or 2 variables. If you were an economist that was studying all the variables in consumer preferences, i'd agree with you. Since you are taking one to two variables and simply comparing them to class rank, a .33 correlation (according to the LSAC article) is not very good.

No, the system of 1L grades is complex. Obviously there are more inputs than we can account for and the human factor involved in the performance.