Monday, May 22, 2006
Why Statistics Should be Mandatory for Law Students
I think statistics should be a mandatory class for law students. Why? Because there are entire fields of law that are nearly impossible to fully grasp without a basic knowledge of statistics. To name a few, I would put employment discrimination, products liability, and much of torts and evidence in this category. In addition, statistical evidence or thinking plays (or at least should play) some roll in almost every area of law. The point is not to enable lawyers to produce competent statistical evidence, but rather to allow lawyers to be educated consumers of statistics.
I also think that making statistics mandatory would help address another problem I see with law-- the relative absence of statistical evidence in many cases where it would be useful. To use but one example that I've seen repeatedly while clerking, courts are often called upon to determine the constitutionality of profiles used by law enforcement officers to target certain individuals for extra scrutiny. Similar cases come up again and again, and yet neither the government nor defense attorneys ever present evidence of the relative accuracy of a given profile. There are probably lots of reasons why this is true, but I think one cause is that lawyers and judges untrained in statistics don't demand or even seek out statistical evidence because they are at a disadvantage while evaluating it. I believe that a bar more adept in statistics would lead to better outcomes in many cases because more sophisticated and objective evidence could be used.
I don't doubt that mandatory statistics wouldn't be popular amongst law students. But part of this stems from math phobia, which should not be encouraged given the inevitability of statistical evidence in the professional life of lawyers. Statistics is a core skill for lawyers, and law schools should be teaching it, even if its rather far afield from the traditional curriculum.
Another alternative, of course, is to make a course in statistics a prerequisite for law school. This would adress all the concerns I have raised without requiring law schools to offer courses outside their core competencies. In many ways this would be preferable, but I can understand why law schools, which currently have no course prerequisites, would be hesitant to open that pandora's box.
TrackBack URL for this entry:
Listed below are links to weblogs that reference Why Statistics Should be Mandatory for Law Students:
» What (Pre) Law Students Should Know from Discourse.net
It isn't going to win me many friend with students, but I have to say that Greg Listokin is basically right: Statistics Should be Mandatory for Law Students. I've long believed that statistics should be a near-prerequisite for law school, and if you ha... [Read More]
Tracked on May 21, 2006 11:26:12 PM
Stats should be mandatory in all US high schools, because you can't be an engaged citizen without understanding the basics of stats. It should also be mandatory for many of the liberal arts, such as poli-sci, government, and so on.
Posted by: Maddy Albright | May 21, 2006 10:32:10 PM
Yiar, not Greg. I know that. And it's fixed on the blog.
Posted by: Michael Froomkin | May 21, 2006 11:53:03 PM
Even though I agree with all of Yiar's good points I am not yet persuaded that they make the case for *requiring* law students to take a stats course (or, in the alternative, making "a course in statistics a prerequisite for law school"). More law schools adding a stats course tailored to law and legal issues (or, perhaps more broadly, an empirical methods course) to their offerings as an elective (upper-level or first-year) would represent helpful and significant progress in this regard.
Again, although I am as empirical as any law prof and firmly believe that, as a normative matter, law students *should* become familiar with basic statistical concepts, my default position (to be sure, rebuttable) tilts away from required courses (beyond, of course, the standard or "traditional" set of required law courses).
Posted by: Michael Heise | May 22, 2006 9:50:10 AM
I agree with Maddy that stats should be part of the high school math curriculum. Teaching stats in law school is very much a second best solution.
I share Michael's general distaste for mandatory courses. If stats weren't mandatory, however, then I would predict that the more numerate students would take the class and the less numerate ones would avoid it. Only if it were mandatory would the students who need it most take it.
Posted by: Yair Listokin | May 22, 2006 10:00:44 AM
Don't most social science majors in most colleges already require stats? And don't most law students come from social science majors? (That's my impression on both of the above assertions, I welcome refutation.) Albeit this isn't at a "producer" level, which would require, e.g., calculus, but I'll hold you to your claim that all they need is a "consumer" level.
Posted by: Paul Gowder | May 22, 2006 10:03:15 AM
I once again agree with Paul Gowder's evil twin. Shall we also require basic economics courses? A writing course that focuses wholly on basic style rather than legal writing per se? Both of these would be useful, and both are brought in by only significant minorities of the class.
Posted by: Dylan | May 22, 2006 12:20:40 PM
"Oh, people can come up with statistics to prove anything. 14% of people know that." -H. Simpson
I think a stats requirement for law students would be *great*. Yes, most people would hate every useful minute of it. But you can deny the utility.
Posted by: Dave! | May 22, 2006 3:48:21 PM
Dylan and Paul:
It is true that a significant minority of law school students have taken statistics and that there are other classes that would be helpful but: 1. The purpose of the statistics requirement is to target the statistics-illiterate-- those other than the minority who have already taken the subject. 2. While its possible to get a sense of economic or philisophical reasoning "on the fly", many basic points of statistics are not particularly intuitive (e.g. Bayes' Rule). A stats class is therefore more urgently needed than these other courses, though I won't deny that these other classes are also a good idea.
Posted by: Yair Listokin | May 22, 2006 4:18:01 PM
One implication of requiring the stats class is that you would have to always have someone available to teach it (assuming that it couldn't be "farmed out" to another department). This means that recruiting might lean toward someone who could teach classes X, Y, & Z and ... a course in stats, thus restricting the pool of potential candidates (to a degree) - and some faculty may not like this idea.
Jeff Yates - J.D., Ph.D.
Department of Political Science
University of Georgia
SSRN page: http://ssrn.com/author=454290
Posted by: Jeff Yates | May 22, 2006 4:30:20 PM
I guess the issue is that there's a lot of things that in an ideal world, people would have before becoming lawyers. I'd like to see law students have advanced math + formal logic (to learn what a really rigorous argument would be like), and, now that I think about it, it kind of bothers me that it's possible to go to law school without having the sort of reverence-generating experiences that can be generated by, e.g., reading Pablo Neruda and J.M. Coetzee and being disturbed by Kafka, or the empathy-generating set of experiences that can be created by some combination of foreign travel (esp. to poor countries) and volunteer work.
Unless there's something special about stats, I don't see why stats should get privileged by being required, as opposed to just strongly recommended. It's not immediately convincing to assert that philosophy and economics are more intuitive than statistics. Bayes' Rule is probably about as intuitive as the concept of opportunity costs, and both are much more intuitive than revealed preference theory and deflationary theories of truth.
Posted by: Paul Gowder | May 22, 2006 6:15:35 PM
I couldn't agree more. There was an editorial today in NY Times on students' aversion to science and this is doubly true for statistics. Even people who take statistics courses often do not have a clue as to statistical thinking.
Statistical thinking is needed for operational definitions (what does safe mean, what does on time mean), prediction, understanding variation and its forms (see Shewhart) etc. Statistical thinking is fundamental to prediction and there are virtually no acitivities that humans engage in that don't involved prediction.
Posted by: John Dowd | May 25, 2006 2:04:26 PM
In my law school class at UT Austin, only 5 of some 140 students had taken degrees in the hard sciences or math. While all were taught to "think like a lawyer," few, unfortunately, were trained to "think like a scientist."
Lawyers, like politicians, are smart people who can't do math, and we are all paying for it.
Posted by: jimbino | May 28, 2006 2:34:05 PM
Right on. I was a Poly Sci/French major, but I am always saying it was my math and economics courses (in which I was only a couple of credits short of a major) that helped me the most in law school and in my practice. Math has to be one of the best pre-law majors....
Posted by: China Law Blog | Sep 3, 2006 1:37:24 AM