Perceptions of Gender, Race, and Anti-Conservative Discrimination on Campus

Hal R. Arkes

Scholarly research has adduced various levels of racism, sexism, and anti-conservative discrimination on college campuses, though there is considerable debate regarding the frequency and intensity of each transgression type.1 Although there are theoretical papers and qualitative research on each category of discrimination, we could find no empirical studies comparing the pervasiveness of discrimination against individuals who possess all three targeted demographic characteristics (non-White, female, conservative). The modest number of studies of universities that have compared perceived discrimination against race/ethnicity and gender have generally found that racial discrimination is the more virulent form of discrimination, confirming the “ethnic-prominence hypothesis” that racial or ethnic discrimination is more prevalent and insidious than gender discrimination. But no research has been conducted in which the relative intensities of gender, racial, and ideological discrimination against conservative, female racial minorities is gauged.2 This study hopes to fill this important research gap.

Comparison of Discrimination: Race, Gender, and Political Viewpoint

Perceived discrimination based on political viewpoint differs from racial and gender discrimination in a number of ways. First, among the possible targets of discrimination, one’s political views are generally much easier to hide than one’s gender and race. Second, there are stark differences in the acceptability of expressing negative views toward conservatives versus women or African Americans. For example, colleges regularly announce that they are disinviting a conservative speaker. But it is inconceivable that a college would publicly announce that it was disinviting a speaker due to that person’s race or gender. This contrast might provide even more motivation for persons to conceal their political viewpoint, particularly in a college setting in which differential acceptance toward the three bases of discrimination might be particularly marked. Third, race and gender are relatively fixed, whereas political viewpoint is voluntarily chosen. Hence, observers might be more willing to “blame” persons for adopting an unfavorable political viewpoint, but have little basis for blaming them for their race or gender. Given that ideology can be hidden more easily than race or gender, and that ideological discrimination is likely more socially acceptable, it is possible that the true extent of anti-conservative prejudice on campus is not reflected in reported data: furtive campus conservatives may reduce the frequency of instances in which discrimination becomes manifest.

Study 1

To accomplish our goal of examining relative rates of discrimination among women of multiple demographic groups, we surveyed women who were either African American conservatives or White conservatives. To assess whether current exposure to academia fostered the three potential types of discrimination, we sought survey respondents who were either currently in college or within a year of graduation, comparing them to college graduates who were between one and ten years past their graduation. Thus all respondents had college experience, but for one group immersion in the college experience was current or very recent; for the other group it was somewhat more remote.

Subjects

All 181 women were recruited from Qualtrics3 panels which participants had previously consented to join. Participation in the experiment was done on a computer. We divided the subjects into (a) 111 women who were beyond their freshman year in college but had not graduated more than a year earlier, and (b) 70 women who had graduated more than a year earlier, but not more than ten years ago. We assumed that any bias or discrimination experienced in the college environment would be more salient for the former group than for the latter.

Method

As is the case with many national surveys (e.g., American National Election Study, 1972 - 2012), we used one political viewpoint screening question. Ours had three options: “liberal,” “conservative,” and “middle of the road.” Only those selecting the second option were entered into the experiment. Then African American women read the following:

In this experiment we are interested in whether a person might have a perception that others might not like them based on who they are. For example, under some circumstances some people might not like your political views. Under some circumstances some people might be prejudiced against your race. Under some circumstances some people might be prejudiced against females. On the other hand, perhaps you might feel that you have not been discriminated against or disliked based on any of these factors. By “discrimination” we don’t necessarily mean that you’ve been denied a job, for example. Instead we mean that you may have felt reluctant to express your political views because you’ve been afraid that if you did express your political views you might be disliked by others, or when writing an application you didn’t want to reveal your race or gender because you thought it might negatively influence your chances. On the other hand, perhaps you’ve never had any of these concerns.

On the following three scales please indicate the extent to which you have been concerned that each of the following three characteristics might have led to a negative perception of you by others. Please drag the slider on each scale to indicate your answer. “0” means no concern at all. “100” means very great concern. Any answer between 0 and 100 is acceptable.

(The scale was a line marked with 11 deciles (0 – 100) and with an arrow situated at 0. The participant could click the mouse on the arrow and then drag it to the desired rating.)

The text for White women was identical except that all reference to race was omitted. All subjects then saw either three scales, which were gender, race, and political viewpoint in the case of the African Americans, or two scales, which were gender and political viewpoint in the case of the Whites, one scale for each of the relevant demographic variables.

Results

The average age of the 70 post-college women was approximately what we expected: 32.9 years, with a median of 31. The average age of the 111 college women was surprisingly high at 33.9 years, with a median of 25 years, the result of there being 11 women 60 years of age or older. For the analysis we decided to remove the 12 women who were 60 years of age or older (one was in the post-college group), because they were more than a generation older than the mean age of the total sample (33.6). Their racial, gender, and political concerns and experience might therefore be different from those of the other women in the sample.

These deletions left 69 African American women and 100 White women in the study. In this reduced sample the average age was 31.1. The average age of the college group was 30.2, that of the post-college group was 32.6. Table 1 contains information about all participants.

Participants in each age Group-Study 1

Group

Age

Total

Mean

Median

<30

30-39

40-49

50-59

60-69

>70

In College-1 year past

 African American

20

9

7

2

1

2

41

33.5

30

 White

42

6

6

8

6

2

70

34.2

25

1-10 years Post College

 African American

13

18

0

0

0

0

31

30.3

30

 White

12

17

5

4

1

0

39

35.1

31

White conservative women rated perceived discrimination toward their political viewpoint 46.8 and their gender 33.6 on their respective scales. An analysis of variance revealed that the difference between these two sources of perceived discrimination was significant (p < .01). The correlation between the two sources of perceived discrimination was also significant (+.36, p < .01).

African American conservative women rated perceived discrimination toward their race, political viewpoint, and gender 74.8, 54.4, and 51.8 respectively. An analysis of the data revealed that the perceived target of discrimination was statistically significant (p < .01), obviously due to the higher level of perceived discrimination due to race. Perceived political viewpoint and gender discrimination did not differ significantly. The correlations among the three types of perceived discrimination among African American women were all highly significant, all ps < .001 (race - political = +.51, political–gender = +.57, race–gender = +.70). The average of the three correlations among the possible bases of discrimination perceived by African American women (+ .60)4 accounted for almost three times the variance (35%) compared to the correlation among the two possible bases of discrimination perceived by White women (13%). Table 2 contains the ratings given by all groups in this experiment and in Study 2.

Ratings of each possible source of discrimination

Source

Study

1

2

White

African American

White

African American

Political viewpoint

46.8

54.4

56.8

51.9

Gender

33.6

51.8

32.9

57.6

Race

74.8

73.1

To examine the full range of ages in the following analysis we re-inserted the women who were 60 years of age or older. Within each race we examined the relation between age and perceived discrimination toward each of the perceived targets of discrimination. The regression among the African American women revealed that age was negatively related to perceived racial discrimination (p < .01). This relation was no longer significant when only African American women less than 60 years of age were included in the sample, although the coefficient was still negative. It is interesting that all four of the women who were 70 years of age or older rated all of their respective perceived sources of discrimination as zero. There were two women of this age in each racial group.

Discussion

Among conservative White women we found that their perceived discrimination based on their political viewpoint significantly exceeded their perceived discrimination based on their gender. This result was obtained both for the college and post-college groups, but we acknowledge that the age difference between the two groups was very small. This may have vitiated any college versus post-college comparison. A regression comparing age and perceived discrimination based on political viewpoint was not significant for White women. This result would be expected if anti-conservative discrimination on campuses began ten or more years before this study was undertaken.

Among conservative African American women their perceived discrimination based on their race exceeded perceived discrimination based on the other two possible sources of perceived discrimination. The greater magnitude of perceived racial as opposed to gender discrimination in this group is not a new finding.5 However it is a new finding that the magnitude of perceived discrimination based on political viewpoint is similar to that based on gender in this group.

The mean correlation among the perceived sources of discrimination was high among African American females (+.59). One possible reason for this high correlation is that if an African American conservative woman perceives that she has been discriminated against, it may not be apparent in some instances what the basis of the discrimination might be. As a result the discrimination might be attributed at least partially to all of the possible sources, resulting in the three ratings moving together as a group. We note that the White female conservatives’ correlation was substantially lower; yet they might be expected to have had a similar attributional ambiguity problem as would the African Americans.

We offer a speculation concerning the finding that all four women in their 70s rated all sources of perceived discrimination as zero. First, these women had experienced the civil rights revolution of the 1960s and also the subsequent feminist movement. They also had been exposed to conditions that preceded these movements when they were young adults. Perhaps they—unlike the younger women—were able to compare the current environment with the earlier environment and found the current one to be far less discriminatory than the prior one. The younger women would not have been able to make that comparison based on personal experience.

Study 2

We decided to attempt a replication of Study 1 for three reasons. First, that survey included women who were much older than we had intended. This had a few unanticipated consequences, such as a college group that was older than the usual undergraduate population to which we wanted to generalize and a post-college group that included women who were up to two generations older than the college group. In this second study we only used women who were less than 25 years of age in the college group and no older than 40 in the post-college group.

Second, we wanted to ensure that our sample of women was geographically distributed. Therefore one-fourth of each sample (college and post-college, White women and African American women) came from each of the four sectors of the country (Northeast, South, Midwest, and West).

Third, one commentator suggested that we explicitly mention prejudicial behavior in our request to subjects to rate each of the multiple possible bases for discrimination. We modified our questionnaire accordingly.

Subjects

We divided the subjects into two groups, the latter of which was a modification of a group used in Study 1: (a) those who were in college beyond their freshman year or who had graduated not more than a year ago, and (b) those who had graduated more than 5 years ago but not more than 10 years ago. We limited persons in the former group to be younger than 25 and in the latter group to be no older than 40. Post-graduate persons within 5 years of graduation were not used in order to achieve more separation in age between the college and post-college groups. Eighty-eight females were White, and 88 were African American. One-quarter of each of the four groups (race x college status) were from each of the four regions of the United States. All participants were recruited from Survey Sampling International (SSI) panels.

Method

The political viewpoint screening question again had three options: “liberal,” “conservative,” and “middle of the road.” Only those selecting the second option were entered into the research. Then African American females read the following:

In this research we are interested in whether a person might have a perception that others might not like them based on who they are. For example, under some circumstances some people might exhibit negative behavior toward you due to your political views. Under some circumstances some people might exhibit negative behavior toward you due to your race. Under some circumstances some people might exhibit negative behavior toward you due to your being a female. On the other hand, perhaps you might feel that you have not been discriminated against or disliked based on any of these factors. By “discrimination,” we don’t necessarily mean that you’ve been denied a job, for example. Instead we mean that you may have felt reluctant to express your political views because you’ve been afraid that if you did express your political views you’d be disliked by others, or when writing an application you didn’t want to reveal your race or gender because you thought it might negatively influence your chances. On the other hand, perhaps you’ve never had any of these concerns.

As was the case in Study 1, all mention of race was deleted from the materials provided to White women. The scales were the same as in Study 1.

Results

There were 44 women in each of the four groups (college, post-college x 2 races). The average age of the college group was 21.3, that of the post-college group was 29.6. We note that 20 of the 88 women who were in the 5-10 years post-graduation group were age 24. We presume that they graduated with associate degrees rather than baccalaureate degrees. Pre-registration of the study and all data can be found at this website: osf.io/qm8nb.

White conservative women rated perceived discrimination toward their political viewpoint 56.8 and their gender 32.9 on their respective scales. Our analysis revealed that the difference between the two bases of perceived discrimination was statistically significant (p < .01). The correlation between the two sources of perceived discrimination was also significant (+ .32, p < .01).

African American conservative women rated perceived discrimination toward their race, gender, and political viewpoint 73.1, 57.6, and 51.9 respectively. Our analysis revealed that the basis of perceived discrimination was significant (p < .01). The rating based upon race exceeded that based upon gender, again replicating Levin et al. Perceived political viewpoint discrimination and gender discrimination did not differ significantly (p > .05). The correlations among the three types of perceived discrimination among African American women were all highly significant, all ps < .001 (race - political = + .59, political – gender = + .59, race – gender = + .60). The average of the three correlations among the possible bases of discrimination perceived by African American women (+ .59) accounted for more than twice the variance (35%) compared to the correlation among the two possible bases of discrimination perceived by White women (14%).

Discussion

Study 2 closely replicated all of the results of Study 1: (1) White conservative women perceived discrimination based on their political view to be significantly greater than that based on their gender; (2) African American women perceived discrimination based on their race to be significantly greater than that based on either their gender or political view, which did not differ significantly from each other; (3) college versus post-college comparisons never approached significance; (4) correlations within each racial group among the perceived sources of discrimination were significant; they were substantially higher among African American women.

It is noteworthy that among the White women in these two studies perceived discrimination due to their conservative political view averaged over 50% higher than their perceived discrimination due to their gender. Yancey found that many conservative faculty members felt it very unwise to reveal their conservatism to their peers.6 Our data lead us to conjecture that many college and post-college students might also choose to hide their political views due to the discrimination they believe that they would suffer if such views were divulged. We note that in neither of our two studies would it have been possible to ascertain to what extent subjects’ perceptions were justified. It may frequently be difficult to discern whether race, gender, or political discrimination is an accurate construal of any questionable behavior directed toward oneself.7

We made no attempt to ascertain the accuracy of any of the perceptions of discrimination. Instead we were concerned with the perceptions themselves. However Iyengar and Westwood provided a comparison of actual racial and political viewpoint discrimination which may serve as a guide as to whether the perceptions of our White participants were based in fact.8 In their second experiment Iyengar and Westwood asked Republicans and Democrats to select a job candidate whose name indicated the person was European American or African American and whose party affiliation was either Republican or Democrat. Iyengar and Westwood found that discrimination by both Republicans and Democrats based on the candidate’s political affiliation dwarfed any discrimination based on the race of the candidate.

In their third experiment Iyengar and Westwood had participants play one of two economic games, either the trust game or the dictator game. In the former, Person 1 is endowed with $10, any fraction of which can be given to Person 2. The amount transferred is then tripled, and Person 2 can then return any of the tripled amount to Person 1. This economic game assesses to what extent Person 1 trusts Person 2 to reciprocate any generosity on the part of Person 1. In the dictator game, no tripling occurs, and Person 2 has no opportunity to return any money to Person 1. This is a strict test of generosity. All participants were assigned the role of Person 1. Person 2 was identified either as a Democrat or Republican and as either a European American or African American. Iyengar and Westwood found that the race of Player 2 made no difference to Player 1, but both Democrats and Republicans were significantly more trusting and generous toward persons who shared their own political affiliation.

Iyengar and Westwood pointed out that there are social norms against discrimination based on race, but there are no such restrictions placed on discrimination based on political viewpoint. In fact, harsh criticism of one’s political opponents seems to be encouraged in current American culture. If we assume that social norms against discrimination based on race also apply to gender, then the perceptions of the White women in our study have some support for their view that they face more discrimination based upon their political views than their gender. The African American women, however did not rate perceived discrimination based on political viewpoint higher than that based upon race. To the contrary, they rated perceived racial discrimination highest, which is congruent with the ethnic-prominence hypothesis.

We note that in both of our surveys the perceived discrimination based on gender was higher among African American women than among White women. This may be due to the attributional ambiguity of any perceived discrimination among African American women. We hypothesize that “leakage” from perceived racial to political sources of discrimination is less likely than spillover to gender, because of the efforts many individuals make to conceal their political viewpoint. Such concealment is becoming more difficult for conservatives. For example, the American Psychological Association’s flagship journal published an article in which the assertion that “I believe the most qualified person should get the job” was deemed a “microaggression” by the authors.9 If a conservative expresses this viewpoint on campus, a “bias response team” may take note. Our data suggest that perceived discrimination against a conservative viewpoint may be more than a perception.

Summary

  • African American conservative women perceive race as the most powerful basis of discrimination compared to discrimination based upon gender or political viewpoint, which do not differ significantly from each other.

  • White conservative women perceive conservative political ideology as a more powerful basis of discrimination compared to gender.

  • Perceived discrimination has not changed among those currently in college beyond their freshmen year and those who have graduated within the last ten years. However the four oldest women perceived no discrimination whatsoever.

  • Share
Most Commented

April 16, 2021

1.

Social Justice 101: Intro. to Cancel Culture

Understanding the illogical origin of cancel culture, we can more easily accept mistakes, flaws, and errors in history, and in ourselves, as part of our fallen nature....

April 19, 2021

2.

Critical Race Theory and the Will to Power

A review of "1620: A Critical Response to the 1619 Project" by NAS President Peter W. Wood....

May 30, 2018

3.

The Case for Colonialism

From the summer issue of Academic Questions, we reprint the controversial article, "The Case for Colonialism." ...

Most Read

August 23, 2021

1.

Testing the Tests for Racism

What is the veracity of "audit" studies, conducted primarily by sociologists, that appear to demonstrate that people of color confront intense bias at every level of society?...

May 30, 2018

2.

The Case for Colonialism

From the summer issue of Academic Questions, we reprint the controversial article, "The Case for Colonialism." ...

March 29, 2019

3.

Homogenous: The Political Affiliations of Elite Liberal Arts College Faculty

A study on the partisanship of liberal arts professors at America's top universities. ...