Tuesday, March 13, 2007

Top 50 Law Schools by SSRN Downloads

Brian Leiter's recent ranking of the Most Downloaded Law Faculties, 2006 got me thinking about how to use SSRN downloads to rank law faculties with greater reliability and validity. As he points out, provocatively-entitled articles or corporate finance postings, inter alia, can significantly skew such rankings. He also notes that the top 3 faculty members at each school account for 50%, on average, of that school's downloads, and that that percentage varies significantly from school to school.

I offer a simple solution: omit downloads of the three most-downloaded faculty members at each school when ranking law schools by SSRN downloads. Odd articles (like mine) and potentially unrepresentative faculty members are thereby removed. What is left is a fairly reliable indicator of the depth of each faculty's current articles productivity and visibility. I have computed "recent downloads less top 3" for the 75 top schools by recent downloads as of March 1, 2007. Here are the Top 50 Law Schools, ranked by "recent downloads less top 3."

RankSchoolDownloads
1Harvard26,1900
2Columbia19,2650
3Yale15,6455
4UCLA15,1201
5Vanderbilt14,0638
6NYU13,8308
7U Chicago13,665-3
8Stanford11,760-2
9U Pennsylvania11,6688
10Georgetown11,4742
11George Washington11,107-1
12U Illinois10,444-1
13UC Berkeley9,9183
14Duke8,7205
15U Minnesota8,5300
16U Texas7,801-13
17Northwestern7,5141
18U San Diego6,7084
19George Mason6,4942
20U Michigan6,2354
21Yeshiva6,0645
22Fordham6,0516
23U Virginia5,6074
24Florida State5,307-1
25U Southern California5,202-5
26Cornell5,1414
27Emory5,140-20
28Boston College5,069-3
29Loyola Los Angeles4,4350
30Temple University4,3816
31Indiana Bloomington4,1333
32Boston University3,934-1
33Chapman3,8330
34Ohio State University3,463-25
35Notre Dame3,3904
36New York Law School3,3609
37U Arizona3,2255
38Case Western Reserve2,820-1
39U Maryland2,71011
40Seton Hall2,6847
41Hofstra2,5098
42U Houston2,506-4
43Washington U St. Louis2,5011
44Brooklyn2,45310
45U Wisconsin2,3951
46UC Davis2,3876
47U Iowa2,3564
48Michigan State2,083-13
49U North Carolina2,00015
50Rutgers Camden1,927-10

The right-most column indicates the ranking shift each school experiences by reason of the fact that each school's three most downloaded authors have been omitted. One might think of it as a measure of how much those authors' downloads distort the school's rankings. Thus, for example, Yale, Vanderbilt, NYU, and the University of Pennsylvania fare much better when the top 3 authors at all schools are removed. (Yale, ranked 8th on the basis of all recent downloads, ranks 3rd in "recent downloads less top 3," moving up 5 ranks.) This presumably reflects the depth of those faculties' scholarly output in article form.

4 Comments:

Blogger Alfred Brophy said...

Thanks for this, Ted. Very interesting stuff. Am I right in thinking that this doesn't correct for faculty size? I wonder if we should we be dealing with the average downloads per faculty member, rather than total downloads? Seems to me that the smaller schools are at a huge disadvantage in this kind of ranking scheme.

3/13/2007 2:18 PM  
Blogger Unknown said...

Hi, Ted & Al! I'd sure be interested in the follow-up question that Al asked about average downloads per faculty member--also, is there a way to see if there are folks using BEPRESS instead of (or in addition to) SSRN?

3/13/2007 3:21 PM  
Blogger Unknown said...

I just did a quick search at SSRN re UH, and we went from 38 to 42 in Ted's ordering after removing Lonny Hoffman, Doug Moll, & me from the list of downloads. (I note that Chris Fairman of OSU is down one rank to #2, behind Lucian Bebchuk. Sorry, Chris; time to do one of 6 more articles on related words. (grin))

For some of the UH folks, it's clear that they haven't posted any of their articles on SSRN. For others, it's probably that they do more books than articles, although I assume that they could post abstracts of their books/chapters if they so chose.

I then did a check on UNLV (Boyd School of Law). Same problem. So part of the difficulty in comparing SSRN downloads lies in the fact that not everyone's decided to participate in posting on SSRN. Still, SOME measure of usefulness of articles is nice. I'm wondering if there's a way to add in things like court cases relying on someone's scholarship, etc. Any thoughts?

3/13/2007 3:36 PM  
Blogger Jeffrey Harrison said...

Thanks for this. I think it makes sense and look forward to the per faculty member ranking. I do have one concern. Like Nancy I wonder how widespread SSRN posting is. A number of people at Florida do not post.

3/13/2007 6:29 PM  

Post a Comment

<< Home