Friday, November 09, 2007


One of the people I like best in legal education had an experience that illustrates the craziness of the hiring process. He was in the market and waiting and hoping for calls to interview in DC. One day he came back to his apartment and his roommate had left a note, “Please return call to University of Jake.” He did and got a secretary who scheduled him for an interview. He went to the interview and was told that the school had never contacted him (he later reflected and realized that he had mixed up the school’s name) but, since he was there they would interview him. I don’t have to tell you the rest of the story, which incidentally is true. He got a callback, an offer, accepted the offer, and became one of the school’s best teachers and scholars. When I think of this story I wonder if random assignment of professors to schools is any less rational than the current process.

And then there is this. I have been complaining about relying on other schools' demand for a candidate as a means of influencing whether my school--or any school--should be interested in that candidate. Although I use my school as an example, I do not think it is atypical. The only difference between my school and many others is that it is saddled with someone who is outspoken. So here, depending on the day and who is doing the talking, one heavily promoted candidate, who had no part in this, was cited as having between 22 and 28 interviews (yes, we are pretty reckless). A Harvard reference said the candidate had 28 callbacks for interviews within three days of the conference. (As you know, Harvard faculty are evidently above the truth – they create truths.) But, I do not know if that was where the rumor started. Even a student on the appointments committee proudly told me that the Dean had said the candidate had over 20 callbacks.

When I approached a committee member about the issue of whether it is good policy to regard what is good for other schools as indicative of what is good for us, the answer was that it is human nature, and the colleague began a story about experiments with male (or was it female?) fish being attracted to fish of the opposite sex when they see other fish are attracted. I noted that whether we do actually think or not, unlike fish, we can engage in some reasoning. Please do not misunderstand. I have nothing against fish. I mean, not that there is anything wrong with being a fish.

Here too, you know where this is going. The 28 callbacks eventually shrunk to “more like 10,” thanks to the honesty of the candidate. The problem is this: If 28 was relevant (to me it never was), is it relevant that it is now 10 and, of course, stay tuned. Who knows what it will be tomorrow. If it goes to zero do we scratch the candidate off the list entirely even though it is exactly the same person?

My advice. Don't go down this path in the first place.


Blogger Ani Onomous said...

I know I'm overstaying my welcome here, but this is an interesting issue (and one that's almost driving me back to sports). I have two reactions that are in tension with one another.

1. If you're saying that nothing could be communicated by the interest of others, I think that's clearly wrong. Other schools may have scoured references, discovered that a candidate is completely unpleasant, can't answer a fundamental question that their local expert in the subject matter posed, dissed students, etc. There are lots of things, believe it or not, that aren't easily revealed by a publication or two. The point's the same as with Adrian Peterson, who apparently WAS thought to be at risk of injury (whether by teams drafting above, I don't know: (

Is heeding the interest of rival schools necessarily at the expense of flawless information? Doubt it. Just as 2000 yards one season doesn't guarantee 2000 yards in the next (see Sean Alexander), a couple of publications doesn't guarantee a stream of equal quality publications later on. Each of us has encountered one-trick ponies; each of us has mis-estimated the quality or significance of someone else's article. The worse or more subjective your data, the more significant secondary information. Assuming that, for whatever reason, you didn't trust your own judgment, or didn't trust your faculty's judgment, you might rationally find comfort in the judgment of others -- both higher and lower ranked, mind you.

2. On the other hand, supporting your view, there are important exceptions to the wisdom of crowds -- rational bubbles, information cascades, etc. I think you could sketch a case that recruitment exhibits some of those features. You would need to take account of certain rival tendencies -- e.g., the proven tendency to exaggerate our evaluative powers, the tendency to exaggerate even more the confidence we have in whatever judgments we reach, and the tendency toward, well, sour grapes.

3. My uninteresting bottom line: of course it's stupid to dismiss someone because other schools aren't interested; of course it's stupid to be interested in someone solely because other schools are. But it can be potentially probative, and a spur to investigation. It's messy to deal with, and hard to understand what weight it should be given, but that's true of every single factor.

4. However, if you're truly looking for MoneyBall, and want to regard this competitively -- and if your school is at a competitive disadvantage in some regard (e.g., you lack money, as did the Oakland A's) -- then BY ALL MEANS ignore what other schools are doing. Such candidates may be tipped, and mispriced, so you might as well look for diamonds in the rough.

11/09/2007 10:02 PM  
Anonymous Anonymous said...

Perhaps call backs are like prices in that they aggregate information for people who aren't in a position to effectively evaluate all relevant information. There's a danger of a "market" distorting feedback loop in something like the hiring process, but just like a higher price reflects relative scarcity or higher demand, and sends a signal to consumers that they should buy less of a good, a high call back number sends a signal to hiring committees that they should take a candidate more seriously.

Of course, hiring committees are not in the same position vis-a-vis a prospective candidate as, say, a Starbuck's consumer is to Brazil's coffee crop. The committee should have much more information, and more time to evaluate it according to criteria all their own. As such, relying on a call back number to determine a candidate's value seems to indicate laziness (free riding on the, presumably accurate, work of other schools' hiring committees) as much as it does bias.

11/09/2007 11:53 PM  
Anonymous Anonymous said...

I'm trying to understand the link between the two anecdotes in Jeff's post. One connection seems to be that, at least in this marketplace, signals of demand are unrelated to actual candidate quality or to positive outcomes. Does that presuppose that candidate quality is an objective property of candidates? Do people really think, as this implies, that the candidates on the market in any given year can be rank-ordered from the "best" candidate(s) on down? Or (more along the lines of Jeff's first anecdote and part 1 of ani onomous's post) do people think that candidate quality is actually not fully predictable or measurable at the time of hiring? Are proxies for quality just a way for hiring committees to reassure themselves that they are engaged in a rational process?

11/10/2007 5:48 PM  
Blogger Jeffrey Harrison said...

Dear Anon: I think there are two things going on. First I am suggesting that the process, as suggested by the first example, is far from a coherent way to identify the best professors. (Someone who was rejected even for DC turns out to be great.) Part of the reason is the reliance of surrogates (second example) that are largely the product of the recruiting myths created by candidates, references and hiring committees that become vested in the candidates.

It may be that the best explanation of why this goes on can be found in Marie Reilly's post

11/11/2007 9:55 AM  

Post a Comment

<< Home