The Liberal Skew in Higher Education--Posner
It is no secret that professors at American colleges and universities are much more liberal on average than the American people as a whole. A recent paper by two sociology professors contains a useful history of scholarship on the issue and, more important, reports the results of the most careful survey yet conducted of the ideology of American academics. See Neal Gross and Solon Simmons, "The Social and Political Views of American Professors," Sept. 24, 2007, available at http://www.wjh.harvard.edu/~ngross/lounsbery_9-25.pdf (visited Dec. 29. 2007); and for a useful summary, with comments, including some by Larry Summers, see "The Liberal (and Moderating) Professoriate," Inside Higher Ed, Oct. 8, 2007, available at www.insidehighered.com/news/2007/10/08/politics (visited Dec. 29. 2007).) More than 1,400 full-time professors at a wide variety of institutions of higher education, including community colleges, responded to the survey, representing a 51 percent response rate; and analysis of non-responders indicates that the responders were not a biased sample of the professors surveyed.
In the sample as a whole, 44 percent of professors are liberal, 46 percent moderate or centrist, and only 9 percent conservative. (These are self-descriptions.) The corresponding figures for the American population as a whole, according to public opinion polls, are 18 percent, 49 percent, and 33 percent, suggesting that professors are on average more than twice as liberal, and only half as conservative, as the average American. There are interesting differences within the professoriat, however. The most liberal disciplines are the humanities and the social sciences; only 6 percent of the social-science professors and 15 percent of the humanities professors in the survey voted for Bush in 2004. In contrast, business, medicine and other health sciences, and engineering are much less liberal, and the natural sciences somewhat less so, but they are still more liberal than the nation as a whole; only 32 percent of the business professors voted for Bush--though 52 percent of the health-sciences professors did. In the entire sample, 78 percent voted for Kerry and only 20 percent for Bush.
Liberal-arts colleges and elite universities are even more liberal than other types of institution of higher education. In liberal-arts colleges, the percentages liberal, conservative, and moderate are 62 percent, 4 percent, and 35 percent, respectively; and in elite universities the figures are 44 percent, 4 percent, and 52 percent. Professors in the 26 to 35 year-old age range are less liberal and more moderate (though not more conservative) than older professors, which I attribute to those youngsters' having reached maturity after the collapse of communism. It is thus no surprise that only 1 percent of the young professors describe themselves as "left radicals" or "left activists," compared to 17 percent of those aged 50 or older.
The summary in the Gross-Simmons paper of the previous literature on professorial political leanings finds that, at least since the 1950s, American college and university faculties have been more liberal than the nation as a whole, but that the liberal skew is more extreme today than it was in the 1950s. This is my experience. Between 1955 and 1962 I was a student at Yale College in the humanities and then at the Harvard Law School, and neither the humanities faculty at Yale nor the Harvard Law School faculty was noticeably liberal (the former was actually rather conservative), and I mean by the standards of that era, not by today's standards. Today both institutions are notably liberal, though the present dean of the Harvard Law School has been attempting with considerable success to make her faculty politically more diverse. The Gross-Simmons study notes that the liberal skew is not limited to the United States, but is found in Canada, Britain, and much of Continental Europe, as well.
The survey results raise two questions: What is the explanation for the results? And what are the consequences? I address only the first question.
There is nothing mysterious about the fact that the members of a particular occupational group should have a different political profile from that of the population as a whole. A 1999 survey of U.S. military officers found that 64 percent were Republican, 8 percent Democratic, and 17 percent independent. In contrast, a 2002 study found that 40 percent of journalists are liberal and 25 percent conservative--a breakdown similar to but much less extreme than that of professors.
The conservatism of military officers is easy to understand--conservatives are much more favorable to the use of military force, and to the values of honor, personal courage, discipline, hardiness, and obedience, which are highly prized by the military, than liberals are. And the liberalism of journalists probably reflects the tastes of their readers; in my 2001 book Public Intellectuals: A Study of Decline, I found that the liberal-conservative split among public intellectuals (roughly 2 to 1) corresponded to the ratio of the circulation of liberal newspapers and magazines to the circulation of conservative ones.
It is tempting to conclude that the liberal bias of journalists and professors (especially in the humanities and social sciences) is the same phenomenon--the liberalism of the "intelligentsia," usefully defined by the Merriam-Webster Online Dictionary as "intellectuals who form an artistic, social, or political vanguard or elite." But that just pushes the question back one step: why should an intelligentsia be liberal? Because intellectuals are naturally critical of their society, which in the case of the United States is rather conservative, or at least not "liberal" as academic liberals understand the word? That is not a satisfactory explanation, because a society can be attacked from the Right just as easily as from the Left. Some of the most distinguished intellectuals of the twentieth century attacked social, cultural, political, or economic features of their societies from the Right--think of Martin Heidegger, William Butler Yeats, T. S. Eliot, Friedrich Hayek, and Milton Friedman. Today, in fields such as law, political theory, and economics, there is a vibrant conservative movment--the puzzle is why it is so distinctly a minority movement in the university world. Moreover, our college and university professors, especially those whose interests and background overlap most closely with those of the majority of journalists, appear to be markedly more liberal than journalists, the other major division of the intelligentsia.
One explanatory factor may be that colleges and universities select for people who are comfortable in a quasi-socialistic working environment. Virtually all colleges and universities in the United States are either public or nonprofit, there is usually salary compression within fields, tenure shields professors from the rigors of labor-market competition, and professorial compensation substitutes fringe benefits (such as tenure), leisure, and other nonpecuniary income for high salaries. The ablest academics generally have the highest opportunity costs--the brilliant chemist could get a high-paying job in the private sector, the brilliant law professor could make a lot of money as a practicing lawyer, and so forth--which suggests that the ablest academics attach especially great value to nonpecuniary relative to pecuniary income and hence are likely to feel especially alienated from a capitalist economy.
This may be one reason why elite universities are more liberal than nonelite ones. (The greater liberalism of liberal-arts colleges may just reflect the fact that such colleges employ fewer scientists and engineers, who are less liberal on average than professors in the humanities and the social sciences.) In addition, there is the curious but well-documented fact that Jews are far more liberal than their socio-economic standing would predict; they are also disproportionately found in the faculties of elite colleges and universities. Furthermore, conservatism is associated in many people's minds with religiosity, and faculty in nontechnical fields in elite universities are rarely religious. Catholics and evangelical Christians are underrepresented in such universities. Professors who are conservative in matters of economics, crime control, and national security but liberal with regard to social issues such as abortion rights, homosexual marriage, and separation of church and state would hesitate to describe themselves as conservatives, and many would not vote Republican.
Another factor that may explain the liberal skew in the academy is political discrimination. Academics pick their colleagues, so once a department or school is dominated by liberals, it may discriminate against conservatives and thus increase the percentage of liberals. There is a good deal of anecdotal evidence of such discrimination, but the best test (though hard to "grade" in soft fields) would be whether conservative academics are abler on average than liberal ones. If conservatives are disfavored, they need to be better than liberals to be hired. Political discrimination is less likely to be prevalent in fields in which there are objective performance criteria, which may be why there is a smaller preponderance of liberals in scientific and technical fields.
Related to discrimination is herd behavior, or conformism. Despite their formal commitment to open debate, academics, like other people, do not like to be criticized or otherwise challenged. The sciences, well aware of this tendency, have institutionalized practices, such as peer review, insistence that findings be replicated, and high standards of logical and empirical rigor, that are designed to foster healthy disagreement. These practices are much less common in the humanities and the soft social sciences.
One response to discrimination or herd behavior favoring liberals in academic has been the formation of conservative think tanks; if their professional staffs were added to college and university faculties, the liberal skew would be less extreme, though the difference would not be great.
A further point also related to both discrimination and conformity bias is that once a field acquires a political cast, it will tend henceforth to attract as graduate students and thus as future professors students who share its politics, as otherwise (as Louis Menand pointed out in a comment on the Gross-Simmons study) the students may have difficulty surviving graduate school, obtaining a good starting job, and finally obtaining tenure.
My last point is what might be called the institutionalization of liberal skew by virtue of affirmative action in college admissions. Affirmative action brings in its train political correctness, sensitivity training, multiculturalism, and other attitudes or practices that make a college an uncongenial environment for many conservatives.
For all these reasons, although the weakening of left extremism in college and university faculties can be expected to continue, the liberal skew is unlikely to disappear in the foreseeable future.
Posted by Richard Posner at 05:09 PM | Comments (32) | TrackBack (0)
Events, Field, and the Liberal Skew in Higher Education-Becker
The study by Gross and Simmons discussed by Posner in part confirms what has been found in earlier studies about the greater liberalism of American professors than of the American population as a whole. Their study goes further than previous ones by having an apparently representative sample of professors in all types of colleges and universities, and by giving nuanced and detailed information about attitudes and voting of professors by field of expertise, age, gender, type of college or university, and other useful characteristics. I will try to add to Posner's valuable discussion by concentrating on the effects on academic political attitudes of events in the world, and of their fields of specialization. I also consider whether college teachers have long-lasting influences on the views of their students.
As Posner indicates, the type of persons who go into different fields varies by the characteristics of the field, so that students who become sociologists tend to be more liberal, while those who enter accounting tend to be more conservative-see Table 8 of the Gross-Simmons study on political identification of professors by field. It is also true, however, that the nature of the material analyzed in a field affects the political identification of persons in that field. The late eminent economist George J. Stigler claimed in an article many years ago that the study of economics tends to make the student more conservative because economics emphasizes that the hidden longer run effects of many government policies have much more negative consequences than the initial direct effects. Economists also show how decentralized competitive markets contribute to the general welfare. Similarly, the study of sociology emphasizes the oppressive effects of certain social forces on particular groups, like the less educated and minorities, which influence the attitudes of sociologists toward the prevailing capitalist economic system.
Admittedly, it is difficult to see the connection between the political attitudes of professors in various other fields and the nature of these fields. For example, why do less than 4 percent of historian, according to Gross and Simmons, consider themselves Republicans, whereas 23 percent of nurses do? Perhaps one important factor is that teachers in practical fields, like engineering, nursing, and medicine, see the limitations of what can be accomplished by various types of interventions, whereas those in theoretical fields, like mathematics and literature, can dream of more utopian solutions. Still, the dichotomy between the theoretical and the practical has trouble explaining why a field like history has such liberal academics since many historians deal with various disasters brought about by government ventures.
The differences in political views by age are informative. Generally, younger men and women are more liberal than older ones since age brings experience with the limitations of what can be achieved by grandiose programs. This is captured in the old adage that goes something like "if you are not a socialist when young you have no heart, but if you remain one when you get older you have no brains". Yet Table 15 in Gross and Simmons shows that academics aged 26-35 are significantly less liberal than those aged 50 and older. I suggest that events of the past 30 years are a major reason for this age-reversal on liberal tendencies. The collapse of communism, the growth of the Asian tigers that have emphasized private enterprise and export-oriented policies, the rapid development of China and India after abandoning communism and socialism, respectively, all reduced the attractiveness of Marxist, socialist, and communist ideologies. These events had less effect on the views of older academics since their views were largely determined when older academics were young, but these events had a great influence on attitudes of younger academics since their beliefs were formed while these transforming events were occurring.
Even economists, traditionally more conservative than those in other social sciences, are now much more market oriented and less sympathetic to various forms of government intervention than they were when I was a student many years ago. During the interim, not only did communism, etc collapse, but Keynesian interventionist attitudes also lost favor, and many more studies have shown the harmful effects of different attempts at government interventions in labor and other markets. The retreat among economists from interventionist policies is found not only among American academic economists, but also among younger economists in Europe and Asia, and also to some extent in Latin America. The reason is that the same forces affected economists elsewhere as affected American academic economists. I suspect, but do not have the evidence, that younger academics in other countries are also decidedly less liberal than older ones in other fields as well.
Given the indisputable evidence that professors are liberal, how much influence does that have on the long run attitudes of college students? This is especially relevant since some of the most liberal academic disciplines, like the social sciences and English, have close contact with younger undergraduates. The evidence strongly indicates that whatever the short-term effects of college teachers on the opinions of their students, the long run influence appears to be modest. For example, college graduates, like the rest of the voting population, split their voting evenly between Bush and Kerry. The influence of high incomes (college graduates earn on average much more than others), the more conservative family backgrounds of the typical college student (but less conservative for students at elite colleges), and other life experiences far dominate the mainly forgotten influence of their college teachers.
This evidence does not mean that the liberal bias of professors is of no concern, but rather that professors are much less important in influencing opinions than they like to believe, or then is apparently believed by the many critics on the right of the liberality of professors.
Press release
