Wednesday, October 29, 2008

Average and Marginal

I think being a scholar and being a good teacher overlap. I am not convinced that writing and being a good teacher overlap. I have a small sample on this but some of the brightest and most insightful law professors I have known do not write very much. (Maybe they actually know something the rest of us have missed.) On the other hand, some of the least interesting have a list of publications a mile a long.

Even though being a scholar and good teacher may overlap I am not sure they are commensurates. If they were, we could come up with a measure of law teacher performance that accounts for both. I don't have a good label for that like economists have for all things good -- utility. But let's say it is called "condex" (as in contribution + index) and each faculty member would have a condex number based on scholarship and teaching. (I leave out service because at many schools service is directly related to what the dean asks you to do. If you are among the "ins" or really annoying, the dean appoints you and then awards you for doing what ever it was.)

So everyone has a condex number and there is a faculty average condex. This would nice because we could then get into the process of raising the average by working at the margin. By margin here I mean any change. For example if the average height of a basketball team is 6' 5" we raise the average by adding a player who is taller than the average. The new player is at the margin.

In faculty tenure decisions, the only way to raise the average condex is to tenure people who are above the current average. In effect, this means saying no to people who are better than nearly half of the existing faculty. It's hard to do but existing tenured faculty who are unwilling to do it (because, let's say and as I have heard, it would upset things socially) are shirkers. In fact, a claim that someone should be tenured because he or she did more that an already tenured person -- or even half of the tenured faculty -- is a non starter and a sure way to stagnate a faculty. Existing tenured faculty, above or below average, are sunk costs.

But what if a school really has aspirations? Does the average/marginal analysis work then? Going back to the basketball analogy. Suppose the team average is 6'5" but your goal is to have a team that averages 6'8". Now the strategy changes and the players in the 6'5" to 6'8" range should be skipped. Sure, adding a player who is 6'7" gets you closer to where you want to go but at the same time increases the size of the base you are attempting to improve on and uses up a roster spot for a 6'11" player.

So too with law faculties. A faculty dedicated to improving itself should understand and be willing to reject people who are actually better than half or even most of the incumbents. Adding someone near the average does little and actually hurts in the long run.

4 Comments:

Anonymous Matt said...

"if the average height of a basketball team is 6' 5" we raise the average by adding a player who is taller than the average."

Isn't this false, at least for a basketball team or any other group with a fixed size? I guess this is pedantic and only comparable to a law school faculty if we take a faculty to have a fixed size, but on a basket ball team, one that has a fixed number of players, if you replace any player with one who is taller, even if both players are below the average height, you'd increase the average. Obviously faculties don't have set sizes in the same way, but often enough they only get new hires when someone retires, moves, or doesn't get tenure. So long as the new hire is better than the old the average will then go up. (I'm a bit skeptical that this is the right way to look at the point anyway, but it seems that the math only works if you take certain assumptions that are not obviously right [and are wrong for things like most organized sports teams].)

10/30/2008 3:40 PM  
Blogger Jeffrey Harrison said...

And your point is?

10/31/2008 11:27 AM  
Anonymous Matt said...

That the account given here isn't right, and that we might do better at getting what we want if we have a correct account?

10/31/2008 4:31 PM  
Blogger Jeffrey Harrison said...

I am still missing your point on what is false about the statement. I think what you mean to say is not that the statement is false but that in some circumstances there are other ways to raise the average.

And, I take it your advice to a faculty making a tenure decision for the purpose of replacing someone else (I've never actually seen this.) is that they should simply strive to just make sure the newly tenured person is better than the newly departed person even though both are not very good.

Hopefully no school does this and I would not say it in part because it is not what I think "we want."

10/31/2008 8:00 PM  

Post a Comment

Links to this post:

Create a Link

<< Home