Diversity? We don't need no stinkin' diversity!

On Monday February 9th, a conservative student organization at Duke University took out a full-page advertisement in The Chronicle, a Duke student newspaper, in the form of an open letter to Duke President Nan Keohane.   The ad was an indictment of Keohane and her faculty for hypocrisy and ideological bias: 

"According to your own words, "diversity" is of premium import to an individual's education.  You also claim that "diversity" does not only pertain to race; rather, intellectual diversity is a key element of any learning environment.

We would hope, then, that your earnest regard for "diversity" would be apparent on your own campus.  This does not, however, appear to be the case.  The following information regarding the political party registrations of Duke faculty members might be of interest in your pursuit of your noble goal of "diversity":

[The ad then detailed the voter registration data for faculty members in nine humanities departments showing that only 5% of the 168 faculty members were registered Republicans while 85% were Democrats and 11% were unaffiliated.   In six of the departments, no faculty members were registered Republicans.] 

Since many of the departments listed above have become increasingly politicized over the past few decades, it is disingenuous to argue that the marked disparities found in the data do not have a significant impact on the daily workings of their faculty members.

Our question is straightforward: Is this 'diversity'?"

Not surprisingly, the ad has kicked off a great deal of controversy.   Robert Brandon, chair of the philosophy department, added fuel to the fire by giving this ill-considered quote to a reporter from The Chronicle:

"We try to hire the best, smartest people available," Brandon said of his philosophy hires.  "If, as John Stuart Mill said, stupid people are generally conservative, then there are lots of conservatives we will never hire…

This tempest-in-a-teapot on the Duke campus can be seen as a isolated example, but it does reflect a broader problem.   Most dispassionate observers would agree that over the last several decades, the political center for most U.S. university faculties has move to the left; often far to the left of the younger generation that they are seeking to educate.   However, hard data on this issue is hard to find, unlike information on the racial or gender composition of faculty members, for example.  (For more on the Duke controversy, see this excellent post by Erin O'Connor at Critical Mass.)

David Horowitz and Eli Lehrer of the Center for the Study of Popular Culture did a study in 2002 analyzing the publicly available voter registration information for the faculty and staff of 32 elite US universities and colleges.   The study was able to identify the party registration for more than 1,500 faculty members and found that more than 91% were registered Democrats versus 9% who were registered as Republicans.  In four colleges (Williams, MIT, Oberlin and Haverford) the study was unable to identify any Republicans among the faculty.    Another study found similar ideological bias in terms of commencement speaker selection for these same educational institutions.

My attempt to identify the ideological composition of faculty and staff members at academic institutions used a different approach; tracking campaign contribution data filed with the Federal Elections Commission (FEC).   U.S. campaign finance law requires disclosure of all contributions in excess of $200 to the FEC including the name, address, occupation and employer for each contributor.   This information is published on the Internet by the FEC, and various organizations maintain web sites where anyone with internet access can search for the records of individual contributors or contributors with the same employer.   (The one I used is "The Political Moneyline" published by TRKC at http://www.tray.com.)

I examined political contributions made by staff and faculty members at US News & World Report's 15 top-ranked national universities for the period from January 2001 through the third quarter of 2003 (the most recent data available).   Over this nearly three-year period, faculty and staff at these 15 universities made 2,974 donations for a total of nearly $2.2 million to federal campaigns and political action committees.   Consistent with the results found by Horowitz and Lehrer's 2002 study, 84% of the funds contributed went to Democratic candidates or affiliated PACs while 12% went to Republicans and 4% went to non-partisan PACs that supported both Republican and Democratic candidates.   The chart below shows the distribution by university:

Faculty and Staff Political Contributions by Party:  2001 to 2003Q3
(Percent of funds contributed, excluding non-partisan PACs)

Overall, the ratio of Democratic contributions to Republican ranged from a high of nearly 18:1 at MIT to a little more than 2:1 at Duke and Washington University.   This compares with the nearly equal rates of party affiliation among the U.S. population at large.

Interestingly, the most "elite" and selective universities tended to be least diverse in terms of the political orientation of their faculties.   On the other hand, those schools located in the South or Midwest (which also tended to be less prestigious) had larger minorities of Republican contributors on their staff.   Duke, the university whose Conservative Union has done the most to draw public attention to the issue, is actually the most ideologically "diverse" of the 15 universities examined.

While I know of no source of historical data on the ideological composition of university faculties, my own sense is that ideological conformity has increased over time at U.S. universities.   As a student at Reed College, Harvard University and Oxford University during the late 1970s and early 1980s I remember that many faculty members were more politically conservative than their students.   (And certainly more conservative than I was, at least in my youth.)   Today, it would appear the reverse is true and that faculty members are considerably more liberal than their students.   According to the University of Michigan's National Election Studies data, congressional voting patterns among the youngest age cohort in their study sample (those born after 1975, which would include nearly all college age voters) were highly volatile and not overwhelmingly aligned with any one party.    For example, in 2002, 61% of this age group voted for Republican congressional candidates vs. 39% who voted for Democrats.   In 2000, the vote was nearly reversed with 65% voting for Democrats vs. 35% voting Republican.  By contrast, the overwhelming majority of university faculty and staff appear to be firmly aligned with the Democratic Party over this period. 

Clearly, if discrimination in employment on the basis of political affiliation were illegal, the "statistical disparities" between the ideological composition of faculty and staff at elite universities and the population at large would provide prima facie evidence of discriminatory behavior.   However, this form of discrimination is not only widespread, it is also legal.

In the recent US Supreme Court decision regarding the University of Michigan's use of race as a factor in admissions, the Court endorsed the precedent established in the 1978 Bakke decision authored by Justice Powell that the state had a compelling interesting in ensuring educational diversity.   Writing for the majority in Grutter v.Bollinger et al., Justice O'Conner quoted extensively from Powell in his Bakke decision:

Justice Powell grounded his analysis in the academic freedom that "long has been viewed as a special concern of the First Amendment."    Justice Powell emphasized that nothing less than the "'nation's future depends upon leaders trained through wide exposure' to the ideas and mores of students as diverse as this Nation of many peoples."  In seeking the "right to select those students who will contribute the most to the 'robust exchange of ideas,' " a university seeks "to achieve a goal that is of paramount importance in the fulfillment of its mission."  Both "tradition and experience lend support to the view that the contribution of diversity is substantial." [Citations omitted]

Of course, the Court was considering racial diversity, not intellectual diversity, and student admissions, not faculty appointments.   However, it seems clear that the underlying logic behind these decisions is that diverse perspectives and ideas are an essential part of an educational system in a free society.

What is to be done?

If we accept the premise that politically conservative thinkers are under-represented on the faculties of most institutions of higher learning in the U.S., the question then becomes what (if anything) should be done about it?

I would suggest that the same approach taken in the past to remedy racial and gender discrimination in employment (commonly known as "affirmative action") would be appropriate with regard to political ideology as well.   (This would be particularly relevant in the social sciences and humanities where the political beliefs and world view of the teacher would have a greater impact upon the classroom than it would in fields like mathematics or engineering.) That is not to say that political philosophy or ideology should be a "litmus test" for faculty hiring decisions or that quotas should be used to insure that university faculties accurately reflect the political beliefs of the population at large.   Rather, ideological balance should be taken into consideration along with all of the other relevant factors.   If two equally qualified candidates were competing for a position in a department that was under-represented by ideological conservatives, than the more conservative candidate should be given preference.   In other words, providing politically under-represented groups (in this case conservatives and Republicans) the same preferences currently granted to minority or female applicants. Given the significant role played by the federal government in financing higher education (both directly though Pell Grants and research contracts and indirectly through guaranteed student loan programs), the federal government clearly has the power to compel academic institutions to end discrimination against politically conservative teachers.   Of course, new legislation would be needed requiring schools that receive government support to have faculties that reflect the ideological diversity of the country at large. 

However, one would hope that university faculties, administrators and trustees would recognize that campus ideological conformity is inconsistent with our pluralistic traditions and the ideals of a liberal arts education and voluntarily undertake reforms to insure greater representation of conservative views.   After all, institutions that do not reflect the ideological diversity of our society cannot adequately discharge their responsibilities to educate future leaders in a democratic system of government.   It remains to be seen, however, whether or not the academic community will recognize the need for greater balance in their ranks without the threat of legislation requiring ideological diversity. 

Note on methodology

The data on individual contributions for the 2002 and 2004 election cycles as reported by the FEC database as of mid-January 2004 were copied from The Political Moneyline. However, many other sources are available, including the FEC's own website.   More precisely, I queried the employer/occupation field for matches with several alternate forms for a university name; for example "California Institute of Technology" as well as "Cal Tech", "California Tech" and "CIT".

The data were then manually scrubbed to remove non-University affiliated entries like "Stanford Financial" or "Yale Materials Handling". Student contributions were also manually deleted by searching for the term "student".

The recipients of the contributions were then classified as either Democratic, Republican or Non-partisan by searching for the names of the candidates or PACs on the Political Moneyline to determine their party affiliation or the party distribution of their further contributions.    PACs that gave 80% or more of their contributions to a single party's candidates were considered to be partisan.  "Independents" like Jeffords or Sanders who caucus with the Democrats were considered Democrats.   Republican leaning groups like the Libertarian Party were considered Republican.

The Microsoft^® Excel file containing the data and charts can be downloaded here.

You can also see my earlier post on this subject here.