Thursday, October 30, 2008

Distributive Justice in Law Schools

Sometimes I feel like it's Bill Henderson's world, and I'm just living in it and trying to help connect the dots. So when John McCain talks about "spreading the wealth," I start thinking about distributive justice, specifically who gets what in the law schools that employ some of us, collect tuition from others, and ask still others for money every once in a while.

Our current system of no competition on educational quality among law schools, which I'm trying to help address with the Race to the Top project (download the Voter's Guide to the U.S. News survey here), has serious consequences for distributive justice in law schools.

The first is how we allocate the scarce resources of admission slots and financial aid. I talked a bit about this yesterday, but the basic answer is LSAT scores, as Henderson recently demonstrated. Financial aid and fundraising priorities goes to buying LSAT scores to move up in the rankings, when it could be going to expanding loan repayment programs for public interest or government jobs, or any number of other priorities.

Now, competing for the best students through merit-based aid doesn't sound so bad -- a bit of a waste of money from a public-good perspective -- but not terrible. Until you think about how merit is defined: test-taking speed, wihch is what the LSAT is about in large part.

And, as Bill Henderson has demonstrated in what has to be one of the most important law review article of the past twenty years, the only reason why the LSAT is a good predictor of law school grades is because most law school grades are determined by these time-pressured exams, having little to do with analytic ability or other skills relevant to quality lawyering, and everything to do with speed.

Which brings us to distributive justice problem #2: the next scarce resource we allocate is access to top jobs -- at most law schools, they're only accessible to the top of the class.

And what Henderson and others have demonstrated is that who falls where on the curve is different depending on the assessment method professors choose. That is, if you use a few short memo assignments instead of a time-pressured final exam to determine the grades, different people will get "As" and access to the top jobs. Doing memo assignments -- or final exams with word limits and no heavy time pressure (take-home, 6-8 hr, etc.)-- does a much better job of sorting people by analytic ability and work ethic than the time-pressured final exam. Mike Madison (Pittsburgh) and others have ably discussed the virtue of memo assignments -- one benefit for professors is that doing most of the grading during the semester frees up valuable time for uninterrupted writing at the end. I've found this to be a huge benefit, and on take-home exams or memo assignments, I've never had a problem doing a curve.

So the question for those law professors (until this year, myself included) who continue to use time-pressured exams to determine most of the grades is: why are you choosing speedocracy over meritocracy?

In the status quo, the speedy high-scoring LSAT folks get the best grades on the first-year time-pressured exams that determine grades, giving them the access to the top jobs that generally pay the most, making them the least in need of significant financial aid, while continuing to receive the most aid.

I'm not the first to criticize legal education on such grounds, I realize, but anyone else think it's time for a change? Law professors can act locally, of course, but competition on quality, through the U.S. News rankings, is the easiest way to do this globally and bring about the change we need. In the Voters' Guide we published earlier this week, we highlighted a set of schools that use "best practices" in legal education such as multiple assessments, feedback during the semester, and less reliance on time-pressured exams -- if U.S. News voters would award these schools high marks, we could have a race to the top that would help students learn more and better, and make law schools more meritocratic.

You can contact us at if you want to help; we could use it. In the meantime, I'll shut up for a while until after this other election. Thanks for listening.



Anonymous Anonymous said...

"next scarce resource we allocate is access to top jobs "

Isn't a better idea to expand the number of top jobs rather than change the composition of your top 10% of the law review?

10/30/2008 2:30 PM  
Anonymous Anonymous said...

I don't know about your exams, but mine test something other than speed. I try not to put the students under time pressure, and most students do report having enough time. The ones that don't are probably struggling because they don't understand the material (or can't answer the question). Even time-pressured exams test the students' ability to identify the most important issues and prioritize, things they will need to do as lawyers. Finally, to the extent that the LSAT and law school exams are both testing (in part) pure intelligence as well as legal aptitude, I have a vague recollection of some study that correlated pure speed (reaction time) with intelligence (as measured by IQ). Indeed, if time pressure is a factor on the LSAT, it may be a good measure of the ability to comprehend, analyze, and decide quickly -- which I would expect to correlate with intelligence with subsequent performance in legal work.

10/30/2008 3:12 PM  
Blogger Jason Solomon said...

Anon frustrated, sounds good to me! Suzanna, thanks for your comment. That's good that they're not time-pressured in your exams, maybe I just need to make mine less complicated. Although I think you still need to make an argument for why a 4-hr exam with no word limit is better than a 6-hr with word limits (assuming same level of difficulty, etc.)-- maybe a debate in front of our 1Ls? :)

I don't know much more than what's in Henderson's paper, but what Henderson's paper indicates -- inded this is the main point of his paper -- is that the study you have in mind is false or outdated. That is, speed and intelligence are not correlated. And that's the problem with both LSAT, which obviously measures intelligence to a certain degree, and time-pressured exams. On ability to prioritize, etc. a memo or exam with less time pressure tests that as well. as long as you have space limits.

10/30/2008 6:29 PM  
Blogger Jason Solomon said...

I should have known better than to try to win an argument with Suzanna Sherry, who points out this line in Henderson's paper: "In general, the results of a speeded power test will be very strongly correlated with the same test given under unspeeded conditions." Not quite sure how this squares with the psychometric evidence he cites earlier in the paper that I thought cut the other way, but maybe I'll ask him, or someone else knows.

10/31/2008 9:09 PM  
Blogger Unknown said...

For many types of lawyering, speed of thought is crucial. Appellate argument is maybe the best example, but also trial or deposition work where objections must be immediately thought through.
But on the issue of whether the LSAT's predict anything much: I doubt there's much difference within a similar band, such that at any top law school, there probably isn't much difference in legal skills between the 85th percentile and the 80th, and that difference is probably washed out by other factors, but has anyone compare those who scored in the 95th percentile or above with those in the, say, 30th and below. I'll bet there's a hell of a difference in legal skills.

3/13/2009 9:02 PM  

Post a Comment

<< Home