Early next year I am going to attend a symposium on Neil Gross’s book Why are Professors Liberal and Why do Conservatives Care? (2013). So last month I read the book and have been working on my assessment of it. This book tackles the important academic issue of the political makeup of academia within our current political economy since the disproportionate politically progressive nature of academia is well established in previous work. Gross uses data to theorize why academics are so politically progressive and also why the political nature of academia has captured the concern of political conservatives. When I present my talk next year I will address the way he answers both of these research questions. But for this blog I will only concentrate on his explanation of why academia is so politically progressive.
Before I get into the critique of his work I first want to state that I respect Gross’s attention to data as he attempts to answer these questions. I have seen efforts on academic bias that are basically thinly veiled attempts to push for a certain social or political agenda. I never received that impression from Gross’s work. I am going to disagree with him as it concerns his conclusions but I do not want my disagreement to be interpreted as disrespect for his work.
When the issue of the political makeup of academia is discussed, the explanations roughly come down to two major reasons. One is that there is a self-selection mechanism that encourages political progressives to take academic jobs at a higher rate than political conservatives. The other is that political conservatives face discrimination and prejudice hindering their ability to succeed in academia. Gross postulates a variation of self-selection that relies on the notion of the political typing of occupations. He argues that academia has been “politically typed” so that progressives feel more comfortable making a commitment to academic study than conservatives. One way to think about this is to consider how jobs are often sex-typed. The job of elementary school teacher is generally sex-typed for women while the job of fork lift driver is generally sex-typed for men. Men are allowed to teach elementary school and women are allowed to be fort lift drivers but we generally do not expect them to be in such occupations. Likewise, both conservative and progressive students will do well in college. But conservatives who do well are generally expected to go into certain types of occupations such as business or law enforcement. Progressives who do well are generally expected to go into other types of occupations, academia being one of them. According to Gross, these expectations naturally sort out progressive students into academic, scientific careers and sort conservative students into other careers.
An alternative explanation is that political conservatives do not simply choose to avoid academia, they face barriers to entry that dissuade them from entering academia. I do not come into this debate unbiased since I have published on the subject of anti-conservative prejudice in academia. But I do not assert that it is only bias keeping conservatives out of science. I argue that it is both self-selection, and Gross’s version of self-selection is more convincing than other variations of this theory, and barriers of discrimination and bias playing important roles in producing a progressive academic institution.
But to get to my argument we should first look at why Gross argues that discrimination is not a major factor in the political makeup of academia. In his book, Gross reports on his audit study with directors of graduate studies (DGS). He sent out an email from a fictitious graduate student seeking more information about the program. Sometimes the email contained no political information. This was the control group. Sometimes the emails contained information suggesting that the student worked on the presidential campaign of Barak Obama. Sometimes the email contained information suggesting that the student worked on the presidential campaign of John McCain. He found that the differences in how the DGSs reacted to the three types of letters were not significant. With such results of his study he argues that while there may be isolated cases of discrimination or bias that these are not major factors in determining the political makeup in academia.
I respect this study. However, it is only part of what we know about academia bias. Putting this study in the context of the other work on this topic allows us to gain a more holistic picture of this research question. Previous research, such as that in my previous book mentioned above, has confirmed that academics state a relative unwillingness to hire political conservatives. Empirical work (by Rothman and Lichter in 2009) has also documented that social conservatives tend to wind up in positions that are of lower status than is warranted by their professional accomplishments. Gross’s study has to be understood in light of these results.
Beyond having published in this area I have another advantage to understanding Gross’s study. I have the position of graduate advisor at my own school and so I understand this position he is studying. It is one of the places in academia where there are powerful institutional interests that work against personal and social prejudices. It is in the interest of DGSs to maximize the number of students applying to a given program. So we are more likely to overlook potential political, religious and social incompatibilities with incoming candidates than other academics. But it is important to remember that there are several steps in the process of becoming an established professor. One must obtain an undergraduate degree, contact graduate programs, be accepted into a graduate program, complete the requirements of the doctorate, find an academic job, obtain tenure and then finally obtain full professorship. Gross’s work may have caught academics at the stage of making contact with graduate programs where there are institutional pressures for acceptance. Thus while his work informs us on the issue of potential discrimination, it clearly is not the last word on this subject.
Furthermore, I admit that I often skim emails from prospective students. It is in my interest to persuade them to apply for the program and so I look at information that will help me to construct a response to meet that interest. So I wonder how much I would pay attention to the political activities of the emailing student. I also wonder if I would have even noticed such an activity. Audit studies can be useful by supplying subtle cues respondents may react to, but there is the danger of the cues being too subtle to activate the potential prejudice of the respondent. I am not certain that this is the case with this particular study, but my experience as a graduate advisor suggests that this may be a problem.
A final issue should be brought up concerning Gross’s study. Although he does not state so in the book, in his paper with this research he comments that he chose to use John McCain as the representative of conservatives instead of Sarah Palin because he wanted the email to be believable. That very statement is an indication of a larger atmosphere of discrimination. If supporting Palin potentially disqualifies a graduate student from a program then political discrimination is at play. McCain has a reputation, whether deserved or not, as a rebel or maverick to his own Republican party. While most academics clearly are not supportive of Republicans, a Republican who often goes against the policies of that party, as McCain is willing to do, is likely to be more acceptable than other Republicans such as Sarah Palin or George Bush. (Some may say that the mere act of supporting Palin indicates that a student is not ready for graduate school. I find such an attitude highly prejudicial and tapping into a stereotype that conservatives are dumb. It is a stereotype that has flourished in a media that emphasis that Palin is dumb instead of the Democrat Hank Johnson). The audit study would still have the weaknesses I outlined above if a more conservative Republican was used instead of McCain, but it would have been a stronger finding if DGSs showed little or no prejudice even when a candidate worked for a Palin election team.
At best Gross’s study indicates that at a key point of the process – when the prospective student contacts a graduate program – it is fairly likely that politically conservative student will not run into a great deal of discrimination. However, discrimination is more likely at other points of the process. As I have argued elsewhere academic bias is not equally likely to show up at every stage of the process or for every type of conservative. Research suggests that social and religious conservatism is more likely to be stigmatized than economic or foreign policy conservatism. Thus the weakness of choosing McCain, who is not known for social conservatism, instead of Palin or Bush adds more questions about the accuracy of Gross’s conclusions.
All of this is not to say that discrimination is the only factor in the political makeup of academia. My argument is that it is an important factor. Is discrimination more important than self-selection in determining this political makeup? To date we have not come up with the proper methodological techniques that adequately compare these two potential effects. My inclination is that instead of arguing whether discrimination matters, that we should concede that it matters and focus on whether discrimination matters more or less than self-selection. Hopefully future research will explore such a question.