Researchers have found evidence they think explains why university professors tend to lean liberal, rejecting the notion that liberal bias in higher education informs their political views. Two studies conducted by Prof. Neil Gross of the University of British Columbia and Ethan Fosse of Harvard University released Monday attribute this disconnect primarily to self-selection on the part of scholars.
In their first study, Gross and Fosse posed as potential graduate students sending e-mail inquiries to the directors of graduate studies across the nation. A third of the e-mails mentioned the applicant had worked on the Obama campaign, a third mentioned having worked on the McCain campaign and the other third gave no hint of political leanings. They then interpreted the received responses.
"In terms of the amount of information DGSs provide to students and the emotional warmth of their e-mails, there is no discernible bias," Gross and Fosse concluded in their paper. "Although liberal applicants have a slight advantage when it comes to whether or not a response will be received, how quickly, and how enthusiastic that response will be, such an advantage is neither substantively nor statistically significant."
In their second study, Gross and Fosse used a longitudinal database which had information on the political thoughts of thousands of individuals at the beginning of their careers. The study found that those who pursued academic careers were far more likely to have been liberal from the outset. This shows that the disproportionate number of liberals in higher education does not stem from conservatives being turned away from doctoral programs or pressured to change their political beliefs during the pursuit of their degrees.
In a paper released January last year, Gross and Fosse had argued that self-selection, rather than interpersonal selection, was the reason behind the discrepancy in ideological representation.\nUniversity Psychology Prof. Jonathan Haidt, however, argued at the Society for Personality and Social Psychology conference in late January that the bias is real and present. During his speech, he polled the audience of about 1,000 psychologists and found that nearly 80 percent of them identified themselves as politically liberal. Fewer than three dozen identified as centrist or libertarian and only three individuals identified as conservative.
Haidt noted how vastly overrepresented liberals were considering that nearly 40 percent of the American population identifies as conservative.
"Since we're also human beings, we get bound to some extent into a team that shares values," Haidt said. "This puts us at risk of creating a hostile climate for conservative students and it makes it harder for us to look on all sides of scientific questions that relate to politics or more generally liberal sacred values."
Haidt, who identified himself as a liberal-turned-centrist, used correspondence from non-liberal graduate students as examples of the developing hostile community. In several e-mails he has received from current graduate students, the moderate or conservative students described their hesitation to participate in political discussions at work and their uneasiness when confronted with jokes designed for and catering to the liberal-minded. Some have expressed their reluctance to conduct research in political psychology which may provide contrary findings.
"[Fosse and Gross] did a very good experiment," Haidt said when asked about the results of the recently released study. "Almost all the work on this topic is correlational, and that has its limits. They did an actual experiment, a field experiment, in which bias could have shown through, and it did not. That is an encouraging sign, but I would point out that it was a very unpromising place to look for bias." He said the failure to find bias in that place does not indicate a lack of discrimination.
"My main concern has been not whether we discriminate but whether our science suffers from a lack of diversity," Haidt said.