Aeroplane wrote:Edit: just realized Leiter's figures were from 2006, so the 5 year period he's talking about must be 2001-2005 or 2002-2006. So that would explain there being a difference. In any case, his numbers also show Michigan as a clear winner over Penn.
Well, it's easily remedied with some math. The size of each class-year was computed generally by dividing the student body by 3; not ultra-precise, but any tiny fluctuation is probably being rounded off anyway. I'm guessing the Georgetown numbers I found on their wiki include the night program. I haven't adjusted for that, nor have I adjusted for Michigan's summer starters class.
Original Numbers wrote:1. TIE: HARVARD (26), YALE (26)
3. TIE: CALIFORNIA – Berkeley (11), MICHIGAN (11)
5. TIE: COLUMBIA (10), NEW YORK UNIVERSITY (10), STANFORD (10)
8. CHICAGO (6)
9. TIE: GEORGETOWN (3), MINNESOTA (3)
Per Capita Numbers wrote:1. Yale: 13.8%
2. Harvard: 4.7%
3. Stanford: 5.6%
4. Berkeley: 3.9%
5. Chicago: 3.1%
6. Michigan: 2.9%
7. Columbia: 2.4%
8. NYU: 1.8%
9. Minnesota: 1.1%
10. Georgetown: .4%
and without Yale:
Per capita numbers probably make more sense if you're trying to compute "what are my odds of getting a position". Then again, even at top schools which are not Yale you're talking about a tiny, tiny fraction of students. (Also: I figure a good part of this is the inclinations of the student body. For instance, Columbia is an outstanding school, but I guess a lot of its students get sucked into the NY biglaw market or something.) I'd at a glance break it down into these groups:
1. Yale (14%)
2. Stanford, Harvard (within 1% of 5%)
3. Berkeley, Chicago, Michigan, Columbia (within 1% of 3%)
(there is a good argument for putting Berkeley up in 2)
I suppose you could add
4. NYU, Minnesota, U Penn. (hovering around 1%; NYU arguably belongs in category 3 above)
After that, you're talking about schools that send substantially less than 1%. If comparing 3% to 5% is splitting hairs and making meaningless distinctions about a very small population of geniuses, comparing .5% to .7% is really crazy.
I do feel compelled to add that I think Michigan in general takes some hits for being a bigger school.
Only about 1,410 students with a top 1% LSAT score are generated each year. It's hard for bigger schools like NYU, Harvard, Columbia, and Michigan to pull in a large share of these, which drags average LSAT down. I guess it comes down to what meaningful trait average LSAT measures. It might signal to employers that you probably did very well on the LSAT, but I have a hard time seeing why they'd care about a couple logic games problems when they have your law school grades. If professors teach towards the median student, a higher average LSAT might mean a higher level of instruction. Then again, if the main benefit of having a lot of smart people (yes, with due reservations about equating LSAT scores with intelligence) is having smart fellow students to study with and so on, the average would seem less important. There are also some ways in which a large class provides benefits that aren't measured by US News and the like: for instance, being more likely to find a graduate of your law school at a given firm. (There's a study http://leiterreports.typepad.com/blog/2005/03/new_study_of_na.html
which is based on 40% per capita, 40% number of firms with at least 1 graduate, and 20% number of firms with at least 5 graduates.)