« Chinese bishops and religious freedom | Main | Dellinger & Schroeder on USA & Executive Privilege »

Monday, March 26, 2007

"Fuck," Fairman, Leiter, and SSRN Rankings

Following a mini-controversy in the usual quarters, Brian Leiter writes about his decision not to include Chris Fairman's paper, Fuck, in his count of the 15 most downloaded law faculties for 2006.  Fairman himself, knowing a good thing when he sees it, has written an interesting short piece about which is available, naturally, on SSRN, here

Brian is dismissive of the fuss, calling it "amazing[ ]" that some people objected to the exclusion of Fairman's paper from his count.  He writes that his reason was obvious: "its unusually high download count was due to its provocative title, not its scholarly content."  He writes further:  "If Ohio State had been included in the list of the 15 most downloaded faculties for 2006, it would have been because 90% of its downloads were due to this one paper.  No other school's rank was so much a function of one paper by one person.  SSRN is a pretty weak measure of scholarly performance--as I made very clear--but it is just a joke if one paper by one person essentially determines a school's rank."  Finally, he denies that he excluded the paper for any reasons of bias or out of an antipathy to civil rights scholarship, and reiterates his reason for doing so: "the best explanation for why the paper was downloaded as much as it was had to do with its title, not its scholarly importance or impact."

In writing a short piece on law reviews, the online age, and law school reputation gatekeepers for an online law review supplement, I've been grappling with Brian's thoughts on some of these issues, which can be found at the above link and in this piece.  Although I am fairly sure I disagree with him (I'm still trying to discern precisely what he thinks, and to decide precisely what I think), his thoughts are well worth reading.  I surely think Leiter had no thoughts of bias when he excluded the paper.  My thoughts run, I guess, to the following sorts of questions: 

First, if the problem is one of download counts for schools being influenced by only one or two professors' work, why not just omit the top posters for all schools on a consistent basis, as Fairman argues in his response?  Or conversely, since Leiter already lists the top three posters for other schools, why not just do that here?  Second, if, as Leiter reasonably concludes, "SSRN is a pretty weak measure of scholarly performance," why take such care to carve out an exception here?  Third, why bother keeping such a measure at all?  Leiter has a perfectly reasonable response to this -- he maintains a law rankings site and wishes to be catholic in what he includes there -- but we might ask this question more broadly: Why do we care about such measures so much, even while we all question and dispute them so closely?  (See here for one answer.)  Finally, in what sense is SSRN a measure of scholarly performance at all?  If it's because people make an informed choice whether to download a paper after reading the abstract, why exclude Fairman?  Isn't it at least partially the case that downloads often reflect scholarly reputation, more or less crudely measured by the fame of the author or the prestige of his/her school?  Is this really a measure of performance?  Is it that much better a measure of performance than an author's ability to capture attention by writing a piece that deliberately reframes the debate on an interesting issue and, not incidentally, has a smashing, epater-les-bourgeois-type title?  Of course, perhaps that's not really what happened with Fairman's piece -- maybe it's all about the title -- but is that all that different from capturing attention because you teach at a fancy place and have a fancy title?  Maybe it's a little better -- but is the Emperor that much more dignified if he manages to throw on a pair of socks, and not much more?   

Let me add a slight update: A correspondent suggests that Leiter's post contains an answer at least to the questions posed above involving excluding or listing separately the top three downloadees -- namely, that the Fairman situation is different because it involves a single paper by one person skewing, rather than multiple papers from the top three downloaded profs.  Viz., from Leiter's post:  "If Ohio State had been included in the list of the 15 most downloaded faculties for 2006, it would have been because 90% of its downloads were due to this one paper.  No other school's rank was so much a function of one paper by one person."   

Posted by Paul Horwitz on March 26, 2007 at 11:49 AM in Life of Law Schools | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c6a7953ef00d834f0a7b453ef

Listed below are links to weblogs that reference "Fuck," Fairman, Leiter, and SSRN Rankings:

Comments

There's a lot more on these matters over at Law School Innovation:
http://lsi.typepad.com/lsi/2007/03/debating_ssrn_d.html

Posted by: Doug B. | Mar 26, 2007 1:25:13 PM

My favorite part of Brian's post is "No Ranking is Too Trivial to Spark Commentary from Folks with Time to Burn... especially if they work in Columbus, Ohio it seems!" If the ranking is so trivial, why would he bother collecting the data and posting it in the first place? This "catholicism" theory strikes me as very bizarre. Despite his protestations to the contrary, Brian must believe these idiotic SSRN rankings are of some value; why else bother with them? And why shouldn't they spark controversy if some disagree with his methodology? Doesn't he spend an awful lot of time debunking other rankings' methodologies? Time to burn, indeed.

A final point: If Brian is so sure SSRN rankings are not useful measures of anything meaningful, it would strike me as all the more reason to include "Fuck" to make that very point. He can't be trying to perfect the rankings and claim they are silly all at once. Or maybe he can. But he can't be so dismissive of those questioning his method; his claim to fame is that his ranking methods are superior to others.

Posted by: Ethan Leib | Mar 26, 2007 1:36:15 PM

It seems to me that SSRN rankings are a measure of how interesting a certain community (some combination of "people on the internet" and "the legal community" and "the legal academic community") finds the scholarship posted there.

An article can be interesting because its author has pre-existing prestige that doesn't relate at all to the particular article being downloaded. Is it fair to include these articles in the rankings since they don't measure the value of the scholarship itself?

An article can be interesting because it is concerns a relatable subject, regardless of the subject's academic value (e.g., people may want to read what Neal Katyal has to say about the students who worked on Hamdan but that doesn't mean the article constitutes groundbreaking legal scholarship). Is it fair to include these articles in the rankings?

Why are these categories of scholarship more worthy of inclusion in the rankings than an article like "Fuck," which people find interesting because of its title (and, I would actually propose, because it's an interesting topic separate and apart from the title)?

Posted by: Lindsay | Mar 26, 2007 4:26:19 PM

Let's all agree to call this the "'Fuck'-off controversy".

(As in, off the list.....in case that needed saying.)

:-)

Posted by: jonah gelbach | Mar 27, 2007 9:10:46 AM

In brief reply to Professor Leib:

I "bothered" to post the SSRN download data because people I respect, like Professors Black and Caron, have made a case for the value (subject to various caveats) of this data. I am not persuaded by their case, but I take their opinion seriously.

The "controversy" has not been about the methodology generally--for example, about the meaning of SSRN downloads, or the kinds of issues I raised as caveats--but about one outlier data point that was excluded. Someone who wanted to discuss the methodology would do well to consider in some detail the arguments of Professors Black and Caron.

I do not spend "an awful lot of time debunking other ranking methodologies"; indeed, I spend almost no time on the subject, since it is boring. Several years ago I wrote something about US News, which I link to whenever the subject comes up. Professors Bell, Henderson, Morriss, and Stake have written interesting pieces on rankings, far more interesting than anything I've written, and certainly more intellectually substantial than the feeble objections being raised about my exclusion of one outlier data point.

I am glad that Professor Horwitz's update notes that my original post already contains responses to several of his questions.

Posted by: Brian | Mar 27, 2007 10:39:16 AM

My thoughts on this matter are explained in my SSRN essay, "Fuck and Law Faculty Rankings." I comment here on a single inaccuracy. Professor Leiter continues to state as fact that "90% of [Ohio State's] downloads were due to this one paper." Unfortunately, this has led others (like Professor Horowitz) to repeat this as a potential justification for exclusion. This is, however, false. For 2006, SSRN reports that Ohio State had 22,311 total downloads. In 2006, I had 17,459 total downloads; 35 or so were from an article on collaborative law I also posted in 2006. Some small number may also have been due to downloads of older articles previously posted. Even if one deliberately overstates Fuck’s downloads by including all of my SSRN downloads for 2006 for all articles, I account for only 78.2% of Ohio State’s total (17,459/22,311) and only 72.6% of Emory’s total downloads (17,459/24,044). Indeed, as I point out in more detail in my essay, when you look at the currently reported data (I didn’t preserve all of the year-end data for Emory and Texas authors), the “Top Three” percentages are: Ohio State 85.0%, Emory 79.6%, and Texas 75.1%. I highlight this point so that (1) Professor Leiter can correct the “90%-due-to-one-paper” error, or (2) provide the data he used for his calculation if it was not the publicly available SSRN download data. I offer the “Top Three” calculation merely to put Professor Leiter’s explanation for exclusion into perspective.

Posted by: Chris Fairman | Mar 28, 2007 1:37:19 PM

The comments to this entry are closed.