The Ideological-Conflict Hypothesis

The Ideological-Conflict Hypothesis

Despite substantial support for the prejudice gap (Sibley & Duckitt, 2008), a small but growing body of evidence across disciplines has suggested that the prejudice gap may be overstated. For example, both liberals and conservatives make negative attributions for groups whose values are inconsistent with their own (e.g., Morgan, Mullen, & Skitka, 2010; Skitka, Mullen, Griffin, Hutchinson, & Chamberlin, 2002) and distance themselves from people who do not share their moral convictions (e.g., Skitka, Bauman, & Sargis, 2005). Furthermore, data from a variety of independent and diverse samples have revealed that both liberals and conservatives express intolerance toward groups with whom they disagree (Lambert & Chasteen, 1997; McClosky & Chong, 1985; Yancey, 2010). Research using the “least-liked-group” paradigm—a survey procedure in which participants first choose the group they like least from a list of often extreme activist groups and then complete a measure of political intolerance (e.g., application of free speech rights)—has yielded only weak effects of conservatism on political intolerance (e.g., Sullivan, Marcus, Feldman, & Piereson, 1981) and instead has suggested that both liberals and conservatives express intolerance. Finally, researchers using social identity theory have predicted and found that people, whether liberal or conservative, show intolerance of out-groups, even when groups are based on the most minimal of criteria (e.g., the groups were chosen at random; Reynolds et al., 2007).

These researchers have challenged the presumed asymmetrical relationship between ideology and intolerance and have suggested that intolerance can be prevalent across the ideological spectrum under certain circumstances. To organize, integrate, and expand on these previous findings, and to present a more complete picture of ideological intolerance, we proposed the ideological-conflict hypothesis. Our hypothesis predicts that conservatives and liberals will be similarly intolerant against social groups whose values and beliefs are inconsistent with their own (e.g., supporters of abortion rights for conservatives; opponents of abortion rights for liberals). In addition, the ideological-conflict hypothesis posits that both liberals and conservatives dislike ideas that conflict with their own and, thus, engage in a variety of strategies to maintain their worldview (for a review, see Proulx, Inzlicht, & Harmon-Jones, 2012), such as motivated information processing and defense against worldview-violating groups.

Defense against worldview-violating groups

The desire for a consistent worldview can also lead to intolerance against groups whose values conflict with, or threaten, one’s own values (Chambers & Melnyk, 2006; Henry & Reyna, 2007). Although research in this area has often been applied to conservatives and groups with whom they disagree (e.g., Reyna, Henry, Korfmacher, & Tucker, 2006), the inclination to defend one’s beliefs against worldview-violating groups applies to both liberals and conservatives. Liberals and conservatives differ in the moral values that form the bases of their worldviews (Graham, Haidt, & Nosek, 2009); however, because both liberals and conservatives endorse moral values with similar intensity (Skitka & Bauman, 2008), they should be equally likely to defend against violations and threats to their worldview. In sum, the ideological-conflict hypothesis predicts that both conservatives and liberals will display intolerance against groups whose values, beliefs, and ideas conflict with their own.

Our prediction that intolerance knows no ideological bounds sounds intuitive, yet decades of research have suggested that intolerance is primarily a conservative affliction (e.g., Hodson & Busseri, 2012; Jost et al., 2003; Sibley & Duckitt, 2008). Because an overwhelming majority of social and political psychologists identify as liberals (Inbar & Lammers, 2012), earlier research supporting the prejudice gap may reflect the tendency for many social and political psychologists to investigate and come to a consensus on topics of greater concern to liberals (e.g., prejudice against racial and other stigmatized groups) and to overlook topics that are of concern to conservatives (Mullen, Bauman, & Skitka, 2003). One of the major shortcomings in earlier research, in fact, is that researchers have confounded the target group with its political ideology (e.g., African Americans tend to be politically liberal), which makes it difficult to know what aspect of the target group (e.g., race or ideology) triggers conservative intolerance (see Chambers et al., 2013, for further elaboration).

When attempts have been made to broaden the list of target groups beyond traditionally liberal groups, methodological idiosyncrasies have limited the generalizability and interpretability of these results in terms of the prejudice gap. For example, the least-liked-group paradigm described previously has demonstrated intolerance toward a wide range of social groups. Although this approach is valuable, the most prevalent least-liked groups are extreme groups who hold minority political opinions dismissed by most Americans (e.g., the Ku Klux Klan in the United States; Sullivan et al., 1981; Sullivan, Shamir, Roberts, & Walsh, 1984), which makes it difficult to determine whether these results generalize to typically studied mainstream political and social groups. Moreover, these studies do not include all of the critical comparisons important for the ideological-conflict hypothesis (i.e., conservative and liberal judgments of ideologically similar and dissimilar groups) and often do not consider the role of political ideology in general (cf. Gibson, 2006).

To remedy these methodological shortcomings, our three independent labs conducted multiple studies of liberals’ and conservatives’ attitudes regarding various issues and groups that either conflicted or were consistent with their respective values and agendas (see Table 1 for brief summaries of each study’s theoretical background). Although in each study we used different terminology to describe the reason people are intolerant of ideologically dissimilar targets, all of the mechanisms converge and can fit comfortably under the conceptual umbrella of perceptions that the out-groups hold values, beliefs, and goals that conflict with one’s ideological worldview (see Table 1, column 5).

OPEN IN VIEWER

Table 1.

Summary of Theoretical Backgrounds of Studies on the Ideological-Conflict Hypothesis

Study	Measure of political ideology	Study-specific mechanism	People express intolerance against social groups who . . .	Common integrative mechanism	Primary measure of intolerance
Chambers, Schlenker, and Collisson (2013)	Self-rated liberalism/conservatism	Value conflict	. . . have dissimilar values.		Dislike
Crawford and Pilanski (in press)	Self-rated liberalism/conservatism	Threat	. . . threaten their security and safety.	Conflicting worldview	Political intolerance
Wetherell, Brandt, and Reyna (2013)	Self-rated social and economic liberalism/conservatism	Value violation	. . . violate their moral values.		Willingness to discriminate

The research designs differed among our three labs, but the basic outcomes were the same. Figure 1 shows that consistent with decades of research in political and social psychology, results indicated that conservatives expressed more intolerance of liberal groups (e.g., pro-choice advocates and people receiving welfare benefits) than did their liberal counterparts. However, Figure 1 also shows that inconsistent with the traditional prejudice-gap conclusion, but consistent with the ideological-conflict hypothesis, results indicated that liberals expressed more intolerance of conservative groups (e.g., pro-life advocates and Tea Party supporters) than did their conservative counterparts (Chambers et al., 2013; Crawford & Pilanski, in press; Wetherell, Brandt, & Reyna, 2013). Consistent with the concept of ideological conflict, results showed that the size of the ideology-intolerance relationship for each target group was strongly positively correlated with perceptions that the target group supported an opposing ideology (see Fig. 2; Chambers et al., 2013). Furthermore, the ideology-intolerance relationships were accounted for by the perception that the groups violated or threatened peoples’ values and worldviews (Crawford & Pilanski, in press; Wetherell et al., 2013); that is, both liberals and conservatives were intolerant of groups with conflicting worldviews.

OPEN IN VIEWER

Fig. 1.

Although the overall absolute levels of political intolerance and the willingness to discriminate are not high, these measures reflect extreme and blatant acts of intolerance, such as denying rights expressed in the U.S. Constitution. Error bars represent 95% confidence intervals.

OPEN IN VIEWER

Fig. 2.

Liberals’ intolerance of groups perceived as strongly conservative (red data points) and conservatives’ intolerance of groups perceived as strongly liberal (blue data points; see Chambers et al., 2013, Study 1, Sample 1, Table 1). The x-axis is perceived political ideology of the groups measured on a scale from 1 (strongly liberal) to 5 (strongly conservative); the vertical dashed line is the midpoint of this axis, which indicates political moderates. The y-axis is the difference between liberal and conservative perceptions of the groups measured on a scale from 1 (strongly dislike) to 5 (strongly like); the horizontal dashed line is the zero point of this axis, which indicates that liberals and conservatives do not differ in their impression of the group (r = –.95, p < .001).

Our studies, which included a diverse array of liberal and conservative target groups, eliminated the traditional prejudice gap and revealed that conservatives and liberals can both be intolerant. Associations between conservatism and racism were also eliminated when a Black person was portrayed as endorsing conservative values (e.g., by opposing welfare; Chambers et al., 2013). This finding suggests that the conservatism-racism association (Federico & Sidanius, 2002) may be due, in part, to the assumption that Blacks endorse values that conflict with those of conservatives (cf. Reyna et al., 2006). In sum, liberals and conservatives are both intolerant of groups that they see as endorsing values and worldviews different from their own.

There are several reasons we can be confident in our conclusions:

Support for the ideological-conflict hypothesis was found by three research teams working independently and converging on the same basic results. These results are also consistent with past work that has revealed both liberal and conservative intolerance (e.g., McClosky & Chong, 1985; Yancey, 2010).

The hypothesis was supported across student samples (Chambers et al., 2013; Wetherell et al., 2013; see Fig. 1a), community samples (Chambers et al., 2013; Crawford & Pilanski, in press; Wetherell et al., 2013; see Fig. 1b), and a representative sample of the United States spanning nearly three decades (Chambers et al., 2013; see Fig. 1c).

Each study used different measures of intolerance, including measures of like-dislike (Chambers et al., 2013), feeling thermometers (Chambers et al., 2013; Crawford & Pilanski, in press), measures of political intolerance (i.e., willingness to deny basic constitutional rights to others; Crawford & Pilanski, in press), and measures of the willingness to discriminate (Wetherell et al., 2013; see Table 1, column 6).

The effects are unlikely to be the result of the idiosyncratic responses to any one group because each study used a different set of groups as experimental stimuli with only some overlap (e.g., see Fig. 2 for a list of some of the groups used).

We are not suggesting that previous work on the prejudice gap is wrong or mistaken; rather, we are arguing that it is incomplete. Although we replicated the usual prejudice gap when conservatives and liberals evaluated liberal groups, we found precisely the opposite effect when we assessed attitudes toward conservative groups. Our studies suggest the intriguing possibility that if researchers had spent the past six decades studying intolerance toward conservative instead of liberal groups, the field would have a much different view of the tolerant liberal.

Future Directions

The ideological-conflict hypothesis incorporates past findings and makes new predictions, thereby offering a new way for social psychologists to think about ideology and intolerance. These studies suggest that researchers interested in the relationship between ideology and intolerance should measure intolerance toward an ideologically diverse set of groups. Although our studies show that liberals and conservatives are intolerant of individuals who hold conflicting worldviews, our work does not tell the entire story about political ideology and intolerance, and additional work is necessary to fully understand the implications of the ideological-conflict hypothesis and its integration with previous models of political ideology (e.g., Hibbing, Smith, & Alford, in press; Jost et al., 2003).

One issue that needs further exploration is the underlying causes of political ideology. Although we argued that the psychological underpinnings of liberals’ and conservatives’ intolerance toward ideologically inconsistent groups are the same, the antecedents of political ideologies themselves likely differ and help determine the groups that liberals and conservatives view as ideologically inconsistent. For example, some research has suggested that conservatism develops as a result of greater sensitivity to negativity (Hibbing et al., in press; Jost et al., 2003), which might make some groups seem more threatening to conservatives than to liberals.

Moreover, because of the different values, traits, and cognitive styles that underlie conservative and liberal political ideologies, the consequences and scope of intolerance may differ among liberals and conservatives. For example, the different abstract values and principles endorsed by liberals and conservatives may change the way intolerance is expressed. It could be that the abstract values endorsed by conservatives (e.g., traditionalism) increase discrimination, whereas the abstract values endorsed by liberals (e.g., universalism) may prevent discrimination in some circumstances (cf. Wetherell et al., 2013), but it is also possible that these differences in abstract values do not carry over to groups one dislikes (cf. Crawford & Pilanski, in press). In other words, one open question is whether an ideological commitment to tolerance in the abstract translates into actual expressions of tolerance toward ideological out-groups.

Another possibility stems from work that has indicated that conservatives are less open to diverse experiences (Jost et al., 2003; Sibley & Duckitt, 2008) and prioritize a more diverse array of moral values (Graham et al., 2009). Because people who are not open to new experiences express intolerance against groups who represent new, potentially threatening views (Sibley & Duckitt, 2008), and people are intolerant of those individuals who violate their moral values (Wetherell et al., 2013), the absolute number of social groups that conservatives see as potential targets of intolerance would be higher than it is for liberals. Conversely, given that many social groups with conservative values tend to be larger in terms of population than are social groups with liberal values (e.g., Evangelical Christians outnumber atheists/agnostics; Pew Research Center, 2012), liberals may be intolerant toward a larger absolute number of individuals. A test of this “absolute-value” hypothesis requires that researchers choose target groups at random from a population of social groups without the a priori consideration of their consistency with liberal or conservative values that we made in our work. By randomly choosing the target groups, future researchers could draw firmer conclusions about the true size and variability of the ideological-conflict effects in the real world (Fiedler, 2011).

Finally, in the three sets of studies described in detail here, we treated political ideology as a unidimensional construct ranging from liberal/left to conservative/right (see Table 1, column 2). Political ideology, however, is at least a two-dimensional construct with one dimension focusing on conserving or challenging traditions and the other focusing on supporting or opposing inequality (Jost et al., 2003; Sibley & Duckitt, 2008). We suspect that the use of a multidimensional model of ideology will further confirm the ideological-conflict hypothesis but that the specific groups that are assumed to be in conflict will be specific to the relevant ideological dimension (cf. Crawford & Xhambazi, in press; Sibley & Duckitt, 2008).

Implications: Theory and Discourse

The ideological-conflict hypothesis has implications for psychological theory and political discourse. With regard to theory, the hypothesis connects ideological-intolerance research with broader theories about preferences for consistency (Proulx et al., 2012). Specifically, it suggests that by treating conservatism as an ideology uniquely characterized by intolerance, social scientists have almost entirely overlooked the phenomenon of liberal intolerance and, indeed, the psychology of liberalism broadly (for some exceptions, see Lambert & Chasteen, 1997; McClosky & Chong, 1985; Skitka & Bauman, 2008). At best, this result provides us with a view of ideology that is half-complete. Only by broadening our focus can we capture the full psychological complexities of political ideology and intolerance, which means not only asking when and why conservatives are intolerant but also asking the same about liberals. Accomplishment of this broader focus may require collaboration with researchers from different ideological and theoretical perspectives.

With regard to discourse, these studies indicate that opposition to particular policies or issues (e.g., health care, affirmative action) should not be taken as clear evidence of the inherent intolerant nature of one ideology or another. Instead, this opposition may stem from fundamental psychological processes that humans all share. Our studies showed that people across the political spectrum think their positions on issues of the day are correct, denigrate people who disagree with them, and express revulsion toward people who violate their moral values. This insight will probably not solve deep political divides; however, it may help people better understand the underlying similarities among ideologically disparate groups rather than dwell on the differences. Moreover, we hope that this work gives ideological rivals pause to first consider their own biases, prejudices, and expressions of intolerance before dismissing alternative points of view.

Recommended Reading

Chambers, J. R., Schlenker, B. R., & Collisson, B. (2013). (See References). A representative study of the ideologicalconflict hypothesis.

Crawford, J. T. (2012). The ideologically objectionable premise model: Predicting biased political judgments on the left and right. Journal of Experimental Social Psychology, 48, 138–151. A more thorough examination and discussion of ideological similarities and differences in motivated reasoning.

Jost, J. T., Federico, C. M., & Napier, J. L. (2009). Political ideology: Its structure, functions, and elective affinities. Annual Review of Psychology, 60, 307–333. A recent comprehensive overview of what is known about political ideology.

Sibley, C., & Duckitt, J. (2008). (See References). A recent metaanalysis on the associations among political ideology, personality, and prejudice.