Intellectually Consistent Discrimination

A new study in the BePress journal The Forum (subs. required) verifies that college faculties are overwhelmingly liberal (see also Dan Klein’s research here and here).  In addition to political attitudes, the authors also have information on the respondent’s academic achievement (number of articles published in the past 5 years, presence on editorial boards, percent of time spent in research etc. – see further below).  Finally, they also have a measure of university quality (based on the Carnegie Foundation and US News and World Report rankings).  Put it all together and what do they find?

…being a Republican or conservative significantly reduces the predicted quality of the college or university where he or she teaches, after taking scholarly achievement into account.

I must admit that for a moment I enjoyed basking in my own victim hood.  My failings are not my own but are due to discrimination!  Ahhh, that feels good.

Much as I would like to lay my failings at the feet of the system, however, I cannot do so.  In truth, the system has treated me well and I think myself lucky (no doubt others say the same! 🙂 ).  Perhaps that is one reason why I reject studies that show bias against conservatives in academia.  But it is not the only reason.  I also reject these studies out of intellectual consistency.

When I am confronted with evidence that women earn less than men I do not take this as a sign of discrimination.  Instead, I point out that if you add to the regression a variable measuring continuous attachment to the work-force the difference in wages fades away.  In fact, most "discriminatory" findings fade away when variables like work force attachment, ability, and IQ are included – and don’t forget that there are many unobserved variables of importance and these can also differ systematically across race, sex, and ethnicity. 

I also point out that for whatever reason, men and women prefer different professions and thus make different career choices, all perfectly legitimate.  Different choices lead to different wage distributions.

Taking all of this together along with my understanding of competitive markets and I find it difficult to believe that there is much discrimination in wages. 

I think the same arguments apply to conservatives.  Consider the variables for academic qualifications used in the aforementioned study, number of published articles etc., these are all self-reported with no control whatsoever for quality!  The authors measure quality just like your Dean (according to the old saw, he can count but not read).  Include some better measured variables and I bet the results will fade, and don’t forget that many variables are not observed by the econometrician.

Conservatives and liberals prefer different professions and thus make different
career choices, all perfectly legitimate.  Different choices lead to
different wage distributions.

I am skeptical about the old-boys club.  I am skeptical about the old-liberals club.  I don’t say that these clubs don’t exist.  I just don’t believe that these old clubs are preventing much accomplishment by either women or conservatives.

I think that my positions are consistent.  Wrong, perhaps, but consistent.  Now consider that the same study which finds bias against conservatives also finds bias against women.  I reject the study in its entirety.  Where stands the liberal-left?