How Does Race Affect Law School Faculty Hiring?

Ming Zhu has a new paper out on SSRN - An Empirical Study of Race and Law School Hiring - that will be of interest to those who think about race in academia.  Zhu studied the 2004-05 hiring year and concluded that, holding factors like law school grades and law review membership constant, race had a positive effect on the odds of a candidate getting a law school job.  But, and it's a big but, she concludes that being a minority had a negative effect on the odds of a candidate getting a job at a high status school.  She notes:

As an anecdote, every single hire made by the top 16 schools from the FAR of 2004-2005 was of a white candidate; not a single minority candidate was hired by any of the top 16 law schools.

I haven't had a chance to read this piece through, but I suspect that it will raise a ton of issues - substantively and methodologically. 

Comments

[anon]

People get entry level jobs at top 16 law schools? Wow. This study IS shocking.

[anon]

Just a note - Ming Zhu is a she.

[Jacqueline Lipton]

People might also be interested in Angela Willig's recent article available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1557916

[Anon]

Any candidate, regardless of race, has a tough time breaking into the academy without connections.

[anon]

As I read the paper, the author assumes that there was no affirmative action that went into the credentials used to measure merit. This seems like a potentially serious methodological problem. For example, schools themselves have affirmative action policies for students, some law reviews have affirmative action policies for membership on law review, and most of not all judges consider diversity when it comes to picking law clerks (some very significantly). As a result, measuring the impact on race for getting a teaching job based on objective standards such as j.d. school, law review membership, and clerkships incorrectly assumes that these credentials are not also based in considerable part on race.

(Oh, and I'm in favor of affirmative action -- just pointing out a possible methodological problem with the paper.)

[anon latino]

I think the problem here is obvious. The top candidates have vaps and I don't think vap committees who usually are only picking one or two candidates try to pick diverse candidates.

[A. Anon.]

Anon latino,

Why do you think VAP comittees are less committed to diversity than appointments committies?

Also, it seems worth mentioning that there are VAP/fellowhip programs that specifically target minorities, such as the Reginald Lewis fellowships at Harvard.

[anon latino]

"Why do you think VAP comittees are less committed to diversity than appointments committies?"

Their results.

[A. Anon.]

"Their results."

But that argument is circular. If the results are explained by the lack of commitment by committees, and the lack of commitment by committees is proven by the results, then it's not an argument so much as an assertion.

[anon latino]

My argument isn't circular, just your construction of it. There are plenty of people of color that can write articles that no one will read. If they were interested in hiring minorities, they would. They aren't hired.

[A. Anon.]

anon latino,

My apologies if I misunderstood your perspective. My main experience on this issue is having been on a faculty myself for a good while, including serving on the appointments committee. In my experience, it was understood by everyone on the appointments committee and on the faculty at my school that hiring minorities was one of our biggest priorities. There was a quite significant boost for minority candidates, and it was understood that the selections of the faculty appointments committee would be scrutinized for its racial balance to ensure that the set of candidates were not all white. I've always thought my school was typical in this regard, and informal conversations have tended to support this, but maybe I am simply wrong about that.

[John Smith]

Race most certainly is a factor in law school faculty recruitment. I was told by a member of the recruiting committee on the drive back to the airport "well, good luck, you know these are tough positions to get as white males."

[anon2]

"In my experience, it was understood by everyone on the appointments committee and on the faculty at my school that hiring minorities was one of our biggest priorities. There was a quite significant boost for minority candidates, and it was understood that the selections of the faculty appointments committee would be scrutinized for its racial balance to ensure that the set of candidates were not all white."

Although those comments and the article suggest that even Asians get a bump in the hiring process, I think we should clarify that we are speaking of only certain types of minorities here.

[John Smith 2]

Agree that race is a *dominant* factor in law school faculty recruitment. I was on the market this year as well and was also told informally by an interviewer (at at least one callback) something along the lines of "it's probably going to be tough sledding for you as a white male."

Of course, that's fair. Or is it?

[Jon]

Based on my experiences serving on recruitment, I agree that everyone talks about hiring a minority and may even mean it, but ususally no serious minority is every really considered. And it's not because of credentials of the minority candidates. There are plenty with good backgrounds and they get an initial interview. But, by the time it comes to decide call backs or go even further in the process, most recruitment committees just go with the person they identify with and feel most comfortable with and that's usually bad news for minority applicants.

[GE]

Whether or not race plays a role in law school hiring remains to be decided empirically, or at least is not answered with this paper. After spending about seven minutes reading the paper, it seems pretty clear to me that this paper would get torn apart if it were peer reviewed in, say, a law and economics journal. Race aside, the methods need to be improved.

For the first model, probit or logit would be more appropriate--or at very least the results should be robust to probit and/or logit estimation. In addition, robust standard errors (which would be appropriate) would likely absorb any statistical significance. More appropriate would be clustering the standard errors at the school level--this makes sense if you think that there might be some sort of unobserved synergy created within applicants all coming from the same school--a likely scenerio.

The second model has less than two hundred observations. That alone should raise all sorts of warning about the validity of the results. Not to mention, many of the estimated coefficients don't make sense (i.e., should a federal clerkship be a signal of a good hire? Aren't federal clerkships reserved for the best?)

Extend the data to multiple hiring years, give more thought to modeling, and better identify the variance/covariance matrix, and this paper might retain some meaning.

[anon]

GE:

The first model is a logit -- and the model is already reporting robust standard errors (look at the Appendix tables and the end).

On your other point, I agree that the second model would benefit from more years -- the paper itself acknowledges this problem. Hopefully the AALS releases more data and someone else runs a broader study.

I think the federal clerkship results make sense though. The study has a positive coefficient for appellate level and a negative coefficient for district level - I'm pretty sure the federal circuit clerkships are still more prestigious, so top schools are less interested in you if you've only got a district level clerkship. Do you disagree?

[GE]

Let me clarify a little (I was trying to settle my newborn son down when I wrote the first comment):
A logistic regression is not quite logit. Logistic regressions report odds ratios, while logit reports log odds ratios. Reporting Logit coefficients is much more common (I feel, I could be wrong though) and regardless of the method, marginal effects really need to be reported.
I missed the part about the errors being robust in the first model--that's my bad, but, I feel in my heart that the errors really should have been clustered at the school level. It's really pretty common in the education literature where there are multiple schools, in say, a district, to cluster on the school level. Given the sensitivity of the results on race, I think it would be useful to report them at the very least and try to argue that they shouldn't be (if the results become insignificant).
As for the clerkships: The paper is unclear about what is being reported in the appendices. I'm guessing they are marginal effects since they differ from the odds ratios reported in the body of the text and report negative coefficients. Let's assume that Appendix D reports marginal effects. The only significant and positive clerkship is a federal appellate clerkship. Doesn't a supreme court clerkship send the strongest signal of potential success? I would expect a supreme court clerkship to have a positive and significant coefficient in both equations--more so than a federal appellate clerkship. This would suggest to me too much noise in too little data.
All in all, a good idea, but nothing too conclusive. The only thing I think that seems pretty robust across specifications is that women fair better than men on the meat market all else equal which is good . . . for women.

[anon3]

A. Anon seems to be walking on eggshells with anon latino, who's got a bee in his bonnet on this issue. But a real conversation about race wouldn't start with apologies and deference, but with frank and open dialogue...No one should be intimidated in talking about this stuff.

So, anon latino, you claim: "My argument isn't circular, just your construction of it. There are plenty of people of color that can write articles that no one will read. If they were interested in hiring minorities, they would. They aren't hired."

Here's what I say: there are tons of people, Latino and otherwise, who "can write articles that no one will read." I think it's absurd to imply the reason no one will read them is because they're written by Latinos. I am at a top school, and I think the faculty would JUMP at the chance to hire (more) Latinos (and African-Americans, and women, for that matter.)

So, instead of insinuating a racial bias, which goes 180 degrees against everything I've seen (which is affirmative action IN FAVOR of minority candidates and against white males), why don't you give us some names. Who are the hot-shot Latino professors or candidates you think have been passed over unjustly by schools that don't want to hire them, putatively on racial grounds?

By the way, I'm a minority myself, fwiw, but not Latino.