THE UNIVERSITY, IDEOLOGY, AND POLITICAL SCIENCE

As in academic fields in general and the social sciences in particular, American political science professors lean well to the left of citizens. Summarizing numerous surveys, Klein and Stern (Reference Klein, Stern, Maranto, Redding and Hess2009) reported that among political science professors, Democrats outnumber Republicans by at least six to one. This may reflect our field’s roots in early twentieth-century Progressive thought, when leaders including Woodrow Wilson and Frank Goodnow sought a scholarly discipline that would promote a strong state able to reshape society. This contrasted with the laissez-faire American Economic Association and the then-reactionary and politically uninvolved American Historical Association (Dryzek Reference Dryzek2006; Gunnell Reference Gunnell2006; Seidelman and Harpham Reference Seidelman and Harpham1985).

Despite our field’s statist beginning, Klein and Stern (Reference Klein, Stern, Maranto, Redding and Hess2009) reported that whereas professors in other social science and humanities disciplines moved well to the left in the 1960–2007 period, political science professors shifted only slightly, with conservative political scientists nearly as common in 2007 as a half-century before. Except for economics, political science has the most ideologically diverse faculty in the social sciences and humanities.Footnote ¹ Although to our knowledge, no American Political Science Association (APSA) president since the late Elinor Ostrom (1996–1997) might regularly vote Republican, a conservative or libertarian Republican well might serve on the APSA Council. As Paul (Reference Paul2018) and Yancey (Reference Yancey2012) described, social conservatives face greater barriers than economic conservatives in academic and corporate institutions. Our field’s relative ideological diversity likely means that most political science departments of any size have a Republican professor, giving political scientists personal contact with conservatives that professors in other fields lack. Such exposure encourages dialogue and discourages stereotyping (Lukianoff and Haidt Reference Lukianoff and Haidt2018).Footnote ²

Ceaser and Maranto (Reference Ceaser, Maranto, Maranto, Redding and Hess2009) argued that, in part, the relative ideological stability of political science reflects the import of ideas and personnel from law, economics, and philosophy, as well as foci on such real-world matters as war and elections in an era when (pre-Trump) conservatism had some intellectual impact and, arguably, some policy success. Furthermore, they cited the professional ideology of American political science. As Merelman (Reference Merelman2003) described in a fascinating study of the leading political science department in the 1955–1970 period (i.e., Yale), a commitment to pluralism as both an empirical and normative theory made American political science relatively likely to tolerate dissenters from both left and right. Our field also embraces relative pluralism regarding what and how we research (see King, Schlozman, and Nie Reference King, Schlozman and Nie2009).

An example of this commitment to pluralism came after Middlebury College students and nonstudent activists attempted to assault conservative intellectual Charles Murray. Murray’s host, left-leaning Middlebury political science professor Allison Stanger, suffered injuries while defending the would-be speaker. Although numerous academic organizations (quite rightly) have denounced reactionary violence, only APSA (2017) criticized the leftist behavior at Middlebury, declaring “the violence surrounding the talk undermined the ability of faculty and students to engage in the free exchange of ideas and debate, thereby impeding academic freedom on the Middlebury campus.”

Lukianoff and Haidt (Reference Lukianoff and Haidt2018) posited that numerous factors make disruption of (conservative) speakers such as recently occurred at Middlebury, Claremont, and Berkeley more likely in the near future. These factors include increasingly protective child-rearing practices, social media promoting polarization, and campus “safety bureaucracies,” which reflect the redefinition of students as consumers. Indeed, whereas past demands for collegiate censorship came mainly from administrators, since 2013, students increasingly have pushed to ban emotionally disturbing speech and symbols. In part, as Whittington (Reference Whittington2018) suggested, this may reflect inadequate civic education: young people simply do not understand the importance of free speech to democracy, even when speech causes discomfort. Of course, not all risks to free speech come from the campus left. As Lukianoff and Haidt (Reference Lukianoff and Haidt2018) described, from Charlottesville to Evergreen State off-campus, reactionary elements have threatened or even perpetrated violence. On both the campus left (Rothman Reference Rothman2016) and the populist right (Galston Reference Galston2018), many now question the need for deliberative processes, free speech, and pluralism in general.

We made a prima facie case that, compared to most other academic disciplines, American political science facilitates ideological diversity and dialogue. Does the relative pluralism of our field trickle down to its undergraduates? Relative to their peers, are political science undergraduate majors more open to or at least tolerant of ideas they oppose? In this respect, are they better prepared for liberal democracy?

AN EMPIRICAL TEST

To study whether political science undergraduates are prepared for pluralism, we used HERI data collected from a range of US colleges and universities, a nationally representative sample of students. The longitudinal student survey data that tracked respondents’ attitudes during their four years is from the 2009 Cooperative Institutional Research Program (CIRP) freshman survey (TFS) and the 2013 CIRP College Senior Survey (CSS). The combined TFS–CSS data include 17,667 students at 156 campuses; regressions include about half of this number because we excluded respondents who failed to answer one or more questions. The combined TFS–CSS data include students who completed both the freshman and senior surveys at the same institution. The data also include additional variables about each respondent’s college, such as size and Carnegie classification.

Regarding ideological self-placement, the TFS–CSS survey shows that political science majors lean left relative to their peers, with 46% self-identifying as “liberal” or “far left” compared to 32% of undergraduates generally, 38% of other social science majors, and 41% of arts and humanities majors. Whereas political science professors tend to be more centrist than faculty in other disciplines, the field’s undergraduate majors are as liberal as their peers in other social science and humanities majors when they begin their college career.Footnote ³

What impact might political science have on the development of student views regarding social policies and free speech? Table 1 shows shifts in student views from their first through fourth year in college measured on a four-point scale. The four responses listed at the top are policy/political judgments: belief that abortion should be legal, belief that gay marriage should be legal, support for affirmative action, and agreement that racism is a serious problem. For these policy/political views, negative values indicate movement toward the more left/liberal position from the first to fourth year. The responses listed at the bottom are three measures of support for freedom of speech: belief that colleges can ban extreme speakers, belief that colleges should ban extreme speakers, and belief that dissent is critical to political processes. Although perhaps not ideal, we think that these free-speech items gauge key aspects of support for pluralism. For these free-speech views, positive values indicate movement toward agreement from the first to fourth year.

Table 1 Student Change in Views from First to Fourth Year

Consistent with prior scholarship on student ideological drift in college (Rothman, Kelly-Woessner, and Woessner Reference Rothman, Kelly-Woessner and Woessner2011; Woessner and Kelly-Woessner Reference Woessner and Kelly-Woessner2009b), we found that students tend to drift left on most policy/political judgments, particularly on social issues. It is not surprising that the most significant shift occurs on gay marriage, which is consistent with changes in elite and public opinion in the 2009–2013 period (Paul Reference Paul2018; Pew Research Center 2017). Of greater significance, we see a statistically substantial shift in student views on abortion: a mean of about one third of a point to the left on the four-point scale. Unlike gay marriage, during this period, Pew Research Center (2018) reported stable public support for abortion. Students become slightly more supportive of affirmative action, shifting 0.15 point to the left in four years. The view that “racism is a serious problem” remains stable.

For the most part, opinion change among political science undergraduates resembles that for their peers in other social science disciplines, the humanities, and the hard sciences, which suggests that shifting views do not reflect field of study as much as other aspects of the collegiate experience. The one notable exception concerns affirmative action. Political science majors move 0.36 point to the left in four years compared to other social science majors (0.18 point left) and students as a whole (0.15 point left). Relying on a difference-of-means test, we cannot determine whether this shift is a function of being a political science student or that it reflects other underlying relationships. To examine whether the political science major likely causes this shift in political views, we used a regression model with other personal and institutional variables.

Regarding free-speech survey items, students increasingly agree over time that colleges can ban extreme speakers but that they should not do so. Additionally, we found that, overall, students become slightly more likely to agree that dissent is an important part of the political process. Concerning whether colleges have the legal authority to ban extreme speakers, political science students’ views change in ways consistent with their peers in other disciplines. Yet, regarding whether colleges should ban extreme speakers, political science students become more supportive of free speech than their peers in any other major fields, including the social sciences. Similarly, political science students are far more likely to agree that dissent is critical to the political process. Political science students increase support for the dissent item at a rate four times faster than their peers in other majors.

To consider the underlying cause for student shifts supporting affirmative action, we constructed a series of regression models to determine whether, controlling for other factors, being a political science major correlates with moving left on an item. Table 2 includes four models: (1) the “Field Model Simple,” with variables for whether the respondent is a social science or a humanities major; (2) the “Model w/Poli Sci,” which adds a variable accounting for whether the respondent is a political science major; (3) the “Model w/Controls,” which adds controls for the respondent’s sex, race, and institutional factors; and (4) the “Model w/ SAT Scores,” which adds the respondent’s standardized testing scores.

Table 2 Regression Models Predicting Movement on Affirmative Action (First to Fourth Year)

*p<=0.05; **p<=0.01; ***p<=0.001

First, it is noteworthy that all four models have a low R²; therefore, although a number of the independent variables attain statistical significance, most variance remains unexplained. Second, the first model suggests that social science students move left on affirmative action; however, when the political science variable is added, the social science variable ceases to be statistically significant. Third, whereas a respondent’s race and sex are not significant predictors, both college ranking and size of the undergraduate enrollment correlate with moving left on affirmative action; that is, students from higher-ranked and larger schools are more likely to shift left. Fourth, when we accounted for student verbal SAT scores, both the political science and the humanities variables ceased to be statistically significant.

Political science students have higher verbal SAT scores than students in general. Because the addition of this variable renders the political science variable statistically insignificant, we concluded that leftward movement on affirmative action is not a function of the major. Rather, in the second and third models, the political science major acts as a proxy for language skills. Why high verbal SAT scores are modestly correlated with affirmative action is an interesting question but beyond the scope of this research. The fact that it eliminates the importance of the political science variable suggests that being in the major alone does not increase support for affirmative action. We can predict only a small fraction of the overall variance (with an R² of only 2%); however, it nevertheless is noteworthy that when accounting for verbal SAT scores, being a political science major is unrelated to shifts in affirmative-action views.

When modeling shifts in student attitudes concerning free speech, we again found low explained variance (table 3). The one variable that predicted increased support for freedom of speech is whether a respondent is a political science major. Political science majors become less supportive of speaker bans and more inclined to view dissent as critical to the political process. Although we cannot prove causation, the regressions in table 3 controlling for demographics, SAT scores, school size, rankings, and Carnegie classification provide evidence that being a political science student increases undergraduate commitment to free speech.

Table 3 Regression Models Predicting Movement on Free Speech

*p<=0.05; **p<=0.01; ***p<=0.001

CONCLUSION AND DISCUSSION

Consistent with prior research that found that students move ideologically left during their four years of college (Woessner and Kelly-Woessner Reference Woessner and Kelly-Woessner2009b), we found that undergraduates become somewhat more liberal on social issues such as abortion and gay marriage, with political science majors moving left at roughly the same rate as other social science majors. Although a simple difference-of-means test shows that political science majors move farther left on affirmative action, they are indistinguishable from students in other fields when controlling for verbal SAT scores.

Student support for freedom of speech is less predictable because regression models of attitude change have limited explanatory power. That said, majoring in political science is associated with increased support for free speech. Over time, political science majors become less likely to support speaker bans and more likely to regard dissent as critical to the political process.

The potential impact of ideology on education would not be limited to policy questions in American government.Footnote ⁴ It would be useful to know more about students’ evolving views of the use of military force, support for immigration, and perceptions of organizations such as the International Criminal Courts. Shifts in these views would help us understand how evolving public perceptions alter congressional support for foreign aid, international trade deals, and participation in transnational political organizations. This has implications for other subfields in political science including international relations and comparative and political theory. Our examination of the relationship between policy positions and grades is limited by the questions asked regularly by UCLA’s HERI. In its current form, the data only allow an assessment of changing views in the realm of American politics.

For those who believe political science as a field should promote civic values such as free speech without staking out ideological positions on policy controversies (i.e., abortion, gay rights, and affirmative action), these results are encouraging. Furthermore, given the relatively small magnitude of shifts to the left even on social issues, the results undermine the political narrative that American colleges and universities are “indoctrination machines” transforming conservatives into progressives. At least concerning inculcating free speech—a key aspect of liberal democracy—political scientists can be reassured by these findings.

Although these results are encouraging, indicating that our field may have positive (albeit very modest) impacts, we should do more to leverage the discipline’s strength to promote pluralistic values. First, in our prior work, we argued that political science should take a greater role in the development of state-level civic-education standards, which currently have limited content regarding pluralistic processes. Political science is traditionally a pre-law major; however, we should lobby to make it the dominant major for future social studies teachers. APSA is uniquely positioned to promote nonpartisan civics standards for public schools (Maranto Reference Maranto, Gooch and Rogers2015; Maranto and Woessner Reference Maranto and Woessner2012).

Second, recognizing that our field (like most academic disciplines) leans left, APSA should model pluralism by seeking to represent the ideological diversity of the political science professoriate. Klein and Stern (Reference Klein, Stern, Maranto, Redding and Hess2009) provided evidence that a sizable minority of political science faculty identify as Republican. Yet, as Ceaser and Maranto (Reference Ceaser, Maranto, Maranto, Redding and Hess2009, 216–17) showed, without external intervention from sponsors such as the American Enterprise Institute, APSA’s annual conference programs would have far less ideological diversity than exists among political science professors and (in particular) undergraduates. More recently, in the 2018 election of APSA Council members, seven of 12 nominees endorsed social-justice–related themes in their statements or research summaries whereas none mentioned liberty-related themes or national security. To maintain our tradition of modeling civil engagement across partisan divides, APSA should make good-faith efforts to render its own elite more ideologically diverse, representing its professors and students.

Political science can model civil, thoughtful policy disagreement. That we teach students about political processes without substantially altering their core beliefs is a testament to our relative professionalism and objectivity. We also should model inclusion of political minorities within our field. Doing so will improve research, build public confidence, and maintain democratic traditions. The future of our field—and of liberal democracy—requires no less.Footnote ⁵