How Horowitz "Proves" Campus Liberal Bias

What's Not To Like About The Academic Bill of Rights by Dr. Graham Larkin

Despite his claim to be a defender of freedom, David Horowitz reveals an unnerving lack of regard for the kind of ideological abstention that the Virginian judges were working to defend. This disregard is glaringly evident in the way he arrives at the "statistics" which he regularly evokes as the very reason for implementing the Academic Bill of Rights. In a recent response to the AAUP's condemnation of the Bill and the thinking behind it, Horowitz baldly asserts that

a series of recent studies by independent researchers has shown that on any given university faculty in America, professors to the left of the political center outnumber professors to the right of the political center by a factor of 10-1 and more. At some elite schools like Brown and Wesleyan the ratio rises to 28-1 and 30-1.

He goes on to contend that this "huge correlation between political categories and academic standing" amounts to a "corruption of academic integrity."

Because he doesn't resort to his opponents' tactic of supplying footnotes, I cannot be certain which "independent studies" produced the "10-1" left-right ratio, but all the circumstantial evidence points to two studies. These are the loopy 2001 "survey" by the Frank Luntz Research Center and the Horowitz-run Center for the Study of Popular Culture, and the complementary study, co-authored by Horowitz and Eli Lehrer, titled Political Bias in the Administrations and Faculties of 32 Elite Colleges and Universities. In a declaration very similar to the one in his retort to the AAUP, Horowitz contends in the latter study that "[t]he overall ratio of Democrats to Republicans we were able to identify at the 32 schools was more than 10 to 1." This also seems to be the source of the more extreme "statistics" for Brown and Wesleyan.

If these are indeed the "independent" studies Horowitz has in mind, then the "Democrats" and "Republicans" mentioned in Horowitz's AAUP retort are the 1,431 professors of Economics, English, History, Philosophy, Political Science and Sociology in various subjectively-selected "elite colleges and universities," mostly in the Northeast, whose names seem to match up with those of registered party members in voter records. Even if one were able to reasonably extend the resulting findings to represent the ratio of Democrats to Republicans "on any given university faculty in America," the question remains of how one could possibly use the exact same statistics to "show" just how much "professors to the left of the political center outnumber professors to the right." Easy! All you need to do is ignore the existence of the 1,891 professors in the same departments who you estimate to be "unaffiliated" in their party loyalty. [emphasis added]

I can think of only two ways of coherently defending such a move. On the one hand, one could argue that the unaffiliated majority simply doesn't matter, thereby leaving Horowitz free to concoct his 10-1 generalizations about all professors on the basis of less than half his dubious little data sample. On the other hand, one could simply assume that the unaffiliated majority must 'really' break down into exactly the same left/right proportions as the card-carrying Democrats and Republicans, leaving us with a 10-1 statistic that reasonably represents everyone.