Taking Race Off the Table

Abstract

Whites are theorized to support color-blind policies as an act of racial agenda setting—an attempt to defend the existing hierarchy by excluding race from public and institutional discourse. The present analysis leverages work distinguishing between two forms of social dominance orientation (SDO): passive opposition to equality (SDO-E) and active desire for dominance (SDO-D). We hypothesized that agenda setting, as a subtle hierarchy-maintenance strategy, would be uniquely tied to high levels of SDO-E. When made to believe that the hierarchy was under threat, Whites high in SDO-E increased their endorsement of color-blind policy (Study 1), particularly when the racial hierarchy was framed as ingroup advantage (Study 2), and became less willing to include race as a topic in a hypothetical presidential debate (Study 3). Across studies, Whites high in SDO-D showed no affinity for agenda setting as a hierarchy-maintenance strategy.

Keywords

color-blind policy hierarchy maintenance social dominance orientation agenda setting

In October 2003, the State of California put the following question to voters: “Should state and local governments be prohibited from classifying any person by race, ethnicity, color, or national origin?” (“Proposition 54,” 2003). Proposition 54, also known as the “Racial Privacy Initiative,” would have banned the collection of racial data by state and municipal employers, California’s university systems, and, with the exception of law enforcement and corrections departments, all other government entities. Though it was defeated, Proposition 54 garnered over 3 million “yes” votes. Exit polling revealed a racially divided electorate, with Whites voting in favor of Proposition 54 almost 3 times as often as Blacks (HoSang, 2010).

Proposition 54 was a vivid manifestation of many Americans’ opposition to the governmental practice of racial categorization. Press reports warn that the Obama administration is amassing an “‘Orwellian-style’ stockpile of statistics” comparing the life circumstances of Whites and racial minorities (Sperry, 2015). The Chief Justice of the U.S. Supreme Court, John Roberts, has lamented the “sordid business” of “divvying up” citizens by race (Parents Involved in Community Schools v. Seattle School District No. 1, 2007), and radio personality Glenn Beck has urged his listeners not to answer the U.S. Census’s race query (Seitz-Wald, 2010). Some commentators (Robinson, Connerly, & Douglass, 2002; Whitehead, 2010) even argue that the U.S. Census Bureau should jettison race and ethnicity questions from the decennial census, the American Community Survey, the Economic Census, and the Bureau’s other data collection activities. Because the Census Bureau is a primary source of comprehensive demographic information for policy-makers, analysts, and social scientists (Anderson, 1988), the abolition of racial data would render the government and large swaths of academia essentially “color-blind.”

What drives people to support policies that would curtail not only the government’s use but also its very knowledge of racial facts? We argue that, for some members of the dominant racial group, support for color-blind policy reflects a hierarchy-maintenance tactic that has thus far received little attention: agenda setting (Bacharach & Baratz, 1962). Although some Whites may support color-blind policies because they believe, as Chief Justice John Roberts put it in Parents Involved in Community Schools v. Seattle School District No. 1 (2007), that “the way to stop discrimination on the basis of race, is to stop discriminating on the basis of race,” we contend that a substantial number of Whites are drawn to such policies for their decidedly anti-egalitarian potential (Chow, Lowery, & Hogan, 2013; Knowles, Lowery, Hogan, & Chow, 2009; Sidanius & Pratto, 1999). Indeed, if government cannot know the race of its citizens, its ability to track and reduce racial disparities through race-conscious means would be all but eliminated. Color-blind policy thus removes racial issues from institutional decision-making and public discourse—“setting the agenda” so that race can no longer be effectively discussed and addressed. Thus, for many Whites, support for color-blind policies may reflect the motivation to protect the racial status quo. The present work tests whether Whites engage in agenda setting upon perceiving threats to the racial hierarchy, either by attempting to curtail the collection of racial data or by expressly excluding race from public discussion.

Agenda Setting

Agenda setting refers to an individual or group’s efforts to selectively exclude significant issues from a decision-making process (Bacharach & Baratz, 1962). In this way, power is exercised through “non-decision-making,” in which “A devotes his [sic] energies to creating or reinforcing social and political values and institutional practices that limit the scope of the political process to public consideration of only those issues which are comparatively innocuous to A” (Bacharach & Baratz, 1962, p. 948). We conceptualize agenda setting as a strategy designed to bolster the intergroup hierarchy by barring consideration of issues related to social dominance and subordination. From this perspective, supporting public policies that prohibit the collection of racial data is an agenda-setting maneuver: Absent access to information about race and racial disparities, public discourse about inequality—and institutional efforts to combat it—is not possible.

Recent research in social psychology suggests that individual members of powerful groups engage in strategic agenda setting (see Saguy & Kteily, 2014). For example, during cross-group interactions, members of dominant and subordinate groups tend to prefer different topics of conversation (Saguy, Dovidio, & Pratto, 2008; Saguy, Pratto, Dovidio, & Nadler, 2009). Whereas dominant-group members prefer to discuss commonalities shared with the subordinate group, members of the subordinate group more often wish to address both commonalities and differences in group-based power. Similarly, in the context of intergroup negotiations, dominant-group members prefer to schedule consequential issues (i.e., those at the root of the power difference) for the end of intergroup discussions, whereas subordinate-group members prefer that consequential issues be discussed at the outset of negotiations (Kteily, Saguy, Sidanius, & Taylor, 2013). Dominant-group members’ relative disinterest in “power talk” appears to reflect their desire to protect the intergroup status quo (Saguy et al., 2008; Saguy & Kteily, 2014).

The present analysis extends the notion of agenda setting as a hierarchy-enhancement strategy to the domain of race relations and public policy in the United States. Whereas previous work has examined how powerful individuals attempt to mold discourse in dyadic interactions (Kteily et al., 2013; Saguy et al., 2008; Saguy et al., 2009), we examine attempts by members of the dominant racial group in the United States (Whites) to exclude race from public discourse—either by blocking public discussion of racial topics or by seeking the enactment of “color-blind” policies that would deprive institutions of racial information.

Color-Blind Public Policy

We use the term color-blind policy to denote rules that would restrict or eliminate institutions’ knowledge of individuals’ race. On its face, the notion that people should be treated as individuals rather than as exemplars of racial categories—color-blind ideology—has a hierarchy-attenuating, humanistic “feel” (Knowles et al., 2009). Indeed, this ideology can be taken to embody the Enlightenment ideal that individuals’ outcomes in life should not be influenced by their membership in a particular group (Rousseau, 2009). However, color-blind ideology—depending on how it is construed—can appeal to both egalitarian and anti-egalitarian Whites (Knowles et al., 2009). When construed as a norm of distributive justice, color-blindness permits the enactment of color-conscious policies (e.g., affirmative action) in a bid to mitigate existing inequalities. However, when construed as a dictate of procedural fairness, color-blind ideology requires that race never influence how individuals are treated—thus proscribing the use of color-conscious policies such as affirmative action and enabling existing hierarchies to remain in place (Knowles et al., 2009).

The duality of color-blind ideology raises the question of how to interpret Whites’ support for color-blind policy: Is their support driven by a belief that such policies will help equalize different racial groups’ outcomes (Apfelbaum, Pauker, Sommers, & Ambady, 2010; Plaut, Thomas, & Goren, 2009; Richeson & Nussbaum, 2004)? Or, are Whites drawn to color-blind public policy prescriptions because of these policies’ potential to buttress the extant racial hierarchy? In light of the fact that color-blind policies are fundamentally procedural in nature—in that they weigh against race-conscious processes without regard to decision-makers’ overarching aims—we theorized that support for color-blind policy among some dominant-group members would be uniquely tied to the goal of hierarchy maintenance.

There is historical precedent for the idea that color-blind policies can be used as tools for hierarchy enhancement. In her analysis of racial classification in Latin America, Loveman (2014) examines the consequences of and intentions behind governments’ decisions to omit or include a race question in their national censuses. In Venezuela, which has never conducted a race-conscious census, the presence and living conditions of the Afro-descendent subordinate group has become “statistically imperceptible,” contributing to the country’s “racial democracy myth” (p. 243). From the 1940s to the 1960s, Brazil—a nation in which Euro-descendent citizens enjoy greater power and status than their indigenous and Afro-descendent counterparts—administered race questions in its national census. However, after 1964’s right-wing coup d’état, leaders removed race from the census to subvert “efforts by academics and activists to draw attention to racial inequality in Brazilian society” (p. 244). In contrast, Cuba retained race in its census after the 1959 communist revolution in an explicit effort to track progress toward its stated goal of eliminating racial inequality in Cuban society (p. 246). These examples demonstrate a clear connection between the enactment of color-blind policy and leaders’ motives vis-à-vis their countries’ racial hierarchies.

In the U.S. context, too, the embrace of color-blind policy may betray anti-egalitarian motives. While seemingly defensible from a principled, “post-racial” perspective, the move to ban government and public institutions from collecting racial data would effectively curtail affirmative action, school desegregation plans, attempts to empower minorities through congressional redistricting, and other group-conscious activities meant to reduce racial inequality (Bonilla-Silva, 2003; Carr, 1997; Knowles et al., 2009). In short, race-blind institutions cannot identify and address preexisting racial problems. As such, agenda setting via endorsement of color-blind policy constitutes perhaps the simplest and most efficient means of ensuring that the racial hierarchy is not challenged.¹

Subtle Versus Overt Forms of Hierarchy Maintenance

We argue that some dominant-group members engage in agenda setting to protect the racial hierarchy. If this is correct, then individuals high in anti-egalitarian sentiment should embrace agenda-setting tactics when the hierarchy is threatened. Here, we operationalize anti-egalitarianism in terms of social dominance orientation (SDO; Ho et al., in press; Pratto et al., 1994), an individual difference variable linked to a host of important intergroup attitudes and behaviors (e.g., Overbeck, Jost, Mosso, & Flizik, 2004; Pratto et al., 2000; Pratto, Stallworth, & Sidanius, 1997; Pratto, Tatar, & Conway-Lanz, 1999).

Although long treated as a unidimensional construct, recent work argues that SDO actually consists of two related but distinct dimensions: SDO–Dominance (SDO-D) and SDO–Egalitarianism (SDO-E; Ho et al., in press; Ho et al., 2012; Jost & Thompson, 2000; Kugler, Cooper, & Nosek, 2010). SDO-D is associated with the desire for some groups to actively dominate others, and uniquely predicts support for overt and unsubtle means of group oppression (e.g., the persecution of immigrants, Ho et al., in press; Ho et al., 2012). In contrast, SDO-E is a more passive stance that opposes efforts to reduce intergroup inequality. SDO-E uniquely predicts more subtle means of hierarchy maintenance, such as ideological efforts to inhibit the attenuation of group hierarchies (e.g., “legitimizing myths” that justify dominant-group members’ disproportionate access to social goods; Ho et al., in press; Ho et al., 2012). The policies and attitudes supported by people high in SDO-E are more socially acceptable—at least on their face—than those endorsed by individuals high in SDO-D.

Racial agenda setting bears the hallmarks of a subtle hierarchy-maintenance strategy. Efforts to deprioritize racial discourse and to institute color-blind public policies do not self-evidently betray a desire to preserve unequal intergroup arrangements. In fact, the idea that race should be ignored as a conversation topic or basis for public decision-making may strike many dominant-group members as fair-minded (Apfelbaum et al., 2010; Apfelbaum, Sommers, & Norton, 2008; Bonilla-Silva, 2003; Knowles et al., 2009; Norton, Sommers, Apfelbaum, Pura, & Ariely, 2006). As such, agenda setting should be particularly appealing to Whites committed to more sophisticated means of hierarchy enhancement—namely, those high in SDO-E.

Although we generate strong predictions with respect to SDO-E, we are less certain about the relationship between SDO-D and agenda setting. We suspect that Whites who support unsubtle means of oppression might wish to encourage discussion of race. That is, it may be that Whites high in SDO-D hold to the old-fashioned belief that legitimate differences exist between racial groups and that these differences ought to play an open role in the apportionment of social resources. Moreover, because high-SDO-D individuals prefer “hard power” to “soft power” means of hierarchy maintenance (see Ho et al., in press; Ho et al., 2012), they may not consider color-blind public policy to be an adequate response to the threat posed by those who would subvert the racial hierarchy. When choosing topics for public discourse, then, high-SDO-D Whites might prefer open discussion of racial disparities as a public affirmation of hierarchical intergroup relationships. Alternatively, the disjuncture between SDO-D and agenda setting as a hierarchy-enhancement strategy might be great enough that no relationship exists between the two.

In sum, we propose that Whites committed to relatively passive forms of hierarchy maintenance—that is, those high in SDO-E—will engage in agenda-setting behavior when the hierarchy is threatened. We focus on two exemplars of agenda setting: (a) support for color-blind public policies (Studies 1 and 2) and (b) the strategic exclusion of topics from a public forum (Study 3). For their part, Whites drawn to overt forms of hierarchy maintenance (i.e., those high in SDO-D) may prove indifferent to color-blind policy and the public discussion of race—or perhaps even embrace racial discourse—when the hierarchy is threatened. Three studies tested these hypotheses.

Study 1

Study 1 tested the prediction that high-SDO-E Whites will increase their support for color-blind policies upon perceiving a threat to the existing racial hierarchy. To induce a sense of hierarchy threat, we exposed White participants to information about two organizations: one with a hierarchy-neutral agenda, and one with an explicitly anti-hierarchy mission (reducing intergroup inequality). We reasoned that Whites high in SDO-E would regard the anti-hierarchy organization as a threat to the racial status quo, and respond by increasing their support for color-blind policies. In contrast, we predicted that high-SDO-D Whites would either increase their opposition to color-blind policies in response to hierarchy threat or not be affected by our hierarchy-threat manipulation.

Participants

A total of 200 participants were requested from Amazon Mechanical Turk (“MTurk”; Buhrmeister, Kwang, & Gosling, 2011) in return for a $.51 payment, for an expected yield of 100 non-Hispanic White participants (N = 50 per experimental cell). In all, 198 responses were received. Of these, we excluded 53 non-White or Hispanic participants, 2 who failed to complete large portions of the experimental materials, and 29 who stated that they did not believe the study’s cover story that the two studies were unrelated and that the organizations were real. The final sample of 114 consisted of 67 men and 47 women ranging in age from 18 to 73 years (M = 33.6, SD = 13.1).²

Materials and Method

Participants read that they would take part in two supposedly unrelated studies: one concerning perceptions of organizational mission statements (“Part 1”) and one concerning Internet users’ views and attitudes (“Part 2”).

Part 1

Our experimental manipulation of organization type was embedded in the study involving mission statements. Participants were asked to read a purportedly real organization’s mission statement and subsequently report their impressions of the organization.

Anti-hierarchy organization

In the anti-hierarchy condition, participants read the mission statement of the “Black Equality Alliance” (BEA). The BEA’s goal was described as achieving “equality between Blacks and Whites” through the promotion of “strong affirmative action programs” and “economic reparations for the descendants of slaves,” and noted a “40% increase in private donations” over the past year. This mission statement was intended to create a sense that the existing racial hierarchy was under attack.

Hierarchy-neutral organization

In the hierarchy-neutral condition, participants read the mission statement of an organization called the “Internet Studies Center” (ISC). The statement described the ISC’s goal as examining “who is currently using the Internet, and how they are taking advantage of it” and noted a “40% increase in private donations” to the organization over the past year. This served as our control condition.

After reading the mission statement, participants were asked to describe the purpose of the organization. Then, consistent with the study’s cover story, participants rated their impressions of the statement itself using four items (e.g., “The mission statement was well-written”). After completing this filler task, participants were redirected to a second, supposedly unrelated study.

Part 2

After reading that the second study was about how “Internet users think about social and economic issues,” participants completed, in fixed order, measures of their support for color-blind public policy, anti-egalitarian motives, political orientation, and demographic characteristics.

Support for color-blind public policy

Attitudes toward color-blind policies were measured using two items: “The United States Census Bureau should stop collecting information about race” and “The government should not be in the business of classifying people by race” (1 = strongly disagree, 7 = strongly agree; α = .74).

SDO

Participants’ anti-egalitarian attitudes were measured using Pratto and colleagues’ (1994) 16-item SDO₆ scale. Example items include the following: “It’s probably a good thing that certain groups are at the top and other groups are at the bottom” and “Group equality should be our ideal” (reverse-scored, 1 = very negative, 7 = very positive, α = .94).

Consistent with prior work, we divided the items into protrait (i.e., positively scored) and contrait (i.e., reverse-scored) sets (Ho et al., 2012; Jost & Thompson, 2000; Kugler et al., 2010). The protrait items (e.g., “To get ahead in life, it is sometimes necessary to step on other groups”) tap SDO-D, or the belief that dominant groups should openly and aggressively oppress subordinate groups. The contrait items (e.g., “All groups should be given an equal chance in life”) measure SDO-E, a more subtle form of dominance that opposes efforts to increase groups’ equality. SDO-E items were reversed such that higher values represented stronger preference for intergroup inequality. Both SDO-E and SDO-D displayed good internal reliability (α_SDO-E = .94, α_SDO-D = .94).

Results

Table 1 displays summary statistics for and correlations among the constructs measured in Study 1. Participant gender, age, or level of conservatism did not significantly moderate the results reported here or in any of the following studies and are not discussed further. We first sought to determine whether the manipulation of organization type affected participants’ levels of SDO-E or SDO-D, which were measured after the manipulation. It did not affect either SDO-E (t < 1) or SDO-D (t = 1.12, p = .27).

Table 1.

Means and Standard Deviations of, and Correlations Among, Variables Assessed in Study 1 (N = 114).

Variable	M	SD	1	2	3
1. Color-blind policy endorsement	3.63	1.87	—	−.19*	−.12
2. SDO-E	2.36	1.31		—	.61**
3. SDO-D	2.27	1.35			—

Note. SDO-E and SDO-D = anti-egalitarianism and dominance subcomponents of social dominance orientation, respectively.

p < .05. **p < .01.

We hypothesized that reading an anti-hierarchy (vs. hierarchy-neutral) mission statement would increase endorsement of color-blind public policy among White participants who were high, but not low, in SDO-E. In contrast, we did not have a strong prediction for those high in SDO-D. Thus, we expected a significant SDO-E × Organization Type interaction, and also tested for a SDO-D × Organization Type interaction. Organization type was effects-coded (−1 = hierarchy-neutral condition, 1 = anti-hierarchy condition) and each subscale of SDO was mean-centered and multiplied by organization type to create two-way interaction terms.

Regressing color-blind policy support on SDO-E, SDO-D, organization type, and the two subscales’ interactions with organization type revealed that both interactions were significant (Table 2; see Jost & Thompson, 2000, for a similar analytic strategy). To decompose the SDO-E × Organization Type interaction, we conducted simple slope analyses at 1 SD above and below the mean of SDO-E and SDO-D (Figure 1). As predicted, high-SDO-E participants supported color-blind public policy more after reading about an anti-hierarchy organization than after reading about a hierarchy-neutral organization (B = .54, SE B = .25, β = .33, t = 2.16, p = .03, η² = .04, 95% confidence interval [CI] [.05, 1.04]). Low-SDO-E participants, however, decreased their support for color-blind policy after reading about an anti-hierarchy organization (B = −.56, SE B = .28, β = −.34, t = −2.02, p = .05, η² = .04, 95% CI [−1.11, −.01]).

Table 2.

Results of Regression Predicting Support for Color-Blind Public Policy in Study 1 (N = 114).

Predictor	B	SE B	β	t	p	η²	Lower 95% CI	Upper 95% CI
OT	−.004	.15	−.003	−0.03	.98	.00	−.31	.30
SDO-E	−.07	.16	−.05	−0.41	.68	.002	−.39	.26
SDO-D	−.21	.18	−.17	−1.16	.25	.01	−.56	.15
OT × SDO-E	.42	.16	.34	2.56	.01	.06	.09	.74
OT × SDO-D	−.38	.18	−.32	−2.15	.03	.04	−.74	−.03

Note. OT is coded such that −1 = Internet Studies Center and 1 = Black Equality Alliance. CI = confidence interval; OT = organization type; SDO-E and SDO-D = anti-egalitarianism and dominance subcomponents of social dominance orientation, respectively.

Figure 1.

Color-blind policy endorsement as a function of SDO-E and organization type (Study 1).

The effect of the SDO-D × Organization Type interaction on policy support was also significant, but negative (Figure 2). High-SDO-D participants tended to be more opposed to color-blind public policy after reading about an anti-hierarchy organization than after reading about a hierarchy-neutral organization (B = −.88, SE B = .45, β = −.54, t = −1.94, p = .06, η² = .03, 95% CI [−1.77, .02]). In contrast, low SDO-D participants were more supportive of color-blind public policy after reading about the anti-hierarchy organization than after reading about the hierarchy-neutral organization (B = .87, SE B = .42, β = .53, t = 2.09, p = .04, η² = .04, 95% CI [.05, 1.69]).

Figure 2.

Color-blind policy endorsement as a function of SDO-D and organization type (Study 1).

Discussion

The findings of Study 1 are consistent with the proposition that Whites committed to relatively subtle and ideological means of hierarchy maintenance—that is, those high in SDO-E (Ho et al., in press)—embrace color-blind social policies as an agenda-setting tactic. When led to believe that the hierarchy is under attack (by the fictitious Black Equality Alliance), high-SDO-E participants became significantly more likely to embrace color-blind policies that would curtail the government’s ability to acquire information about, and thus address, racial inequalities. In contrast, high-SDO-D Whites tended to oppose color-blind social policies when the hierarchy was threatened. These findings suggest that a desire for overt intergroup dominance reduces the appeal of relatively subtle hierarchy-maintaining strategies.

Unexpectedly, Study 1 also found significant effects among participants low in SDO-E and SDO-D. Specifically, although low SDO-E Whites’ support for color-blind policy was relatively high after reading about a neutral organization, their support decreased after reading about an anti-hierarchy organization. This finding suggests that thinking about racial differences in the anti-hierarchy condition reduced these Whites’ support for color-blind policy (just as it increased support among anti-egalitarian Whites). Speculatively, such a policy may strike egalitarian Whites as benign (or even desirable) unless or until they are made aware of persistent racial inequalities. These participants may believe that, in a world without inequality, there would be no reason for institutions to track people’s race. However, when racial inequality is made salient, they may believe that there is risk in not monitoring the existence of such inequalities.

In Study 1, the anti-hierarchy mission statement differed from the hierarchy-neutral statement, not only in its implications for the intergroup hierarchy, but also in its mention of race and the historical existence of slavery in the United States. Thus, it is not clear whether high-SDO-E participants increased their support for color-blind policy out of concern for the fate of the intergroup hierarchy or, rather, as a more general reaction to overtly racial rhetoric. Thus, in Study 2, we utilized a different manipulation of hierarchy threat that references race in both conditions. We also described the study as being explicitly about racial attitudes, potentially inducing less suspicion among participants while providing a conceptual replication of our previous effect. In addition, the manipulation allowed us to provide more precision regarding the motivation underlying high-SDO-E and SDO-D Whites’ support for color-blind policy.

Study 2

The results among high-SDO Whites in Study 1 are consistent with either of two motivational accounts. First, it may be that anti-egalitarian Whites’ support for color-blind policy stems from a general concern for the health of the intergroup hierarchy. Alternatively, endorsement of color-blind policy could reflect a more specific desire vis-à-vis the racial hierarchy—namely, to safeguard the White ingroup’s dominant position. To clarify the motive underlying support for color-blind policy, Study 2 leveraged the fact that intergroup inequity can be framed either as dominant-group advantage or subordinate-group disadvantage (Branscombe, 1998; Lowery, Chow, Knowles, & Unzueta, 2012; Lowery, Knowles, & Unzueta, 2007; Powell, Branscombe, & Schmitt, 2005). Although these frames are formally interchangeable, they have different psychological implications for members of the dominant group: Whereas the advantage frame makes salient the dominant group’s privileged position, the disadvantage frame focuses attention on deprivations faced by the subordinate group (Lowery et al., 2012; Lowery, Unzueta, Knowles, & Goff, 2006). If the agenda-setting behavior observed thus far reflects ingroup-focused concerns, then Whites high in SDO-E should react to hierarchy threat by increasing their support for color-blind policy especially (or only) when inequity is framed as White advantage. Likewise, Whites high in SDO-D should increase their opposition to color-blind policies in response to White advantage, as they did in Study 1. If, however, agenda setting reflects a more general desire to maintain the intergroup hierarchy, then how inequity is framed should not affect high-SDO-E and high-SDO-D Whites’ levels of support.

Study 2 also utilized a new hierarchy-threat manipulation. Whereas Study 1 created threat by exposing Whites to an organization aimed at reducing racial inequalities, the present study induced threat by portraying the hierarchy as vulnerable to change (i.e., unstable). A sense that the intergroup status quo is tenuous can trigger attempts by dominant-group members to buttress the hierarchy (Chow et al., 2013; Knowles et al., 2009; Saguy & Kteily, 2014; Tajfel & Turner, 1979). Thus, we predicted that perceptions of racial hierarchy instability would increase support for color-blind policy among individuals high in SDO-E. We did not have strong predictions regarding individuals high in SDO-D.

Study 2 differed from Study 1 in two additional respects. First, we measured color-blind policy support by gauging participants’ reactions to a fictitious bill purportedly before the U.S. House of Representatives that would ban the government from collecting information about citizens’ race or ethnicity. The bill was crafted to closely match the language used in California’s Proposition 54, which specifically prohibits the government from collecting data about, or presenting data separated by, race (with a few exceptions, such as allowing police officers to use race when describing a suspect or when assigning an undercover agent to a case).

Second, we randomly assigned participants to complete either the old version of the SDO scale, used in Study 1, or a new measure of SDO (Ho et al., in press) specifically designed to distinguish between SDO-E and SDO-D. In Study 1, we divided the older SDO scale (i.e., SDO₆; Pratto et al., 1994) into its constituent parts by averaging the positively and negatively worded items separately (Jost & Thompson, 2000). However, this is potentially problematic from a psychometric perspective, as it perfectly confounds SDO dimension with the direction of item wording. In contrast, the new scale contains a balanced set of protrait and contrait items tapping SDO-D and SDO-E (Ho et al., in press). We manipulated which SDO scale participants were administered as a between-subjects factor, but did not expect SDO version to influence the results.

Participants

In all, 550 participants were requested from Amazon Mechanical Turk in return for a $1 payment, for an expected yield of 400 non-Hispanic White participants (N = 100 per experimental cell). A total of 529 responses were received. Of these, we excluded 123 non-Whites or Hispanics, 16 foreign nationals, 15 non-native English speakers, and 4 participants who failed at least one of three attention-check questions (see Online Supplement). The final sample of 371 consisted of 209 men and 162 women ranging in age from 18 to 70 years (M = 33.9, SD = 11.4).

Materials and Method

Participants were told that the study was about their social attitudes and that they should consider social-scientific findings regarding economic inequalities between Whites and minorities. They were then randomly assigned to read one of two passages, in which we embedded the manipulation of inequity frame (see Lowery et al., 2012, for a similar manipulation). Participants read,

Prior research has led most social scientists to agree that, even today, Whites [minorities] in America continue to enjoy undeserved advantages [suffer from undeserved disadvantages] that minorities [Whites] do not, particularly in the realm of employment. Below are some ways in which Whites are advantaged [minorities are disadvantaged], compiled from economic research.

1. Relative to equally qualified minorities [Whites], being White [a minority group-member] increases [decreases] the chance of being hired for a prestigious position.

2. Whites [minorities] receive higher [lower] salaries than equally qualified minorities [Whites]. (White advantage/Minority disadvantage conditions)

After reading the description of inequality, participants were randomly assigned to read one of two forecasts concerning the future of national economic disparities (see Chow et al., 2013, for a similar approach). These passages constituted our manipulation of hierarchy instability. Specifically, participants read,

The above conclusions are based on the most recent Census data. In addition to these findings, researchers have concluded that it is likely [unlikely] that the distribution of income and political power within the United States will change substantially within the next 10 years. (Unstable/stable hierarchy conditions)

Participants were then administered the following measures in fixed order.

Manipulation check

Participants indicated their responses to two items to ensure that the hierarchy instability manipulation had the intended effect: “The distribution of income and political power between racial groups within the United States will change substantially in the next 10 years, at the expense of Whites” and “10 years from now, the political and economic power of Whites in the United States will have decreased as the political and economic power of other groups will have grown” (1 = strongly disagree, 7 = strongly agree, α = .84).³

Color-blind policy support

Participants then rated their support for a supposedly real congressional bill. Specifically, participants were told that “[r]ecently, a member of the U.S. House of Representatives proposed a bill that would significantly alter the ways in which the American government can track the makeup of the U.S. population.” Participants then read the following description of the (fictitious) bill:

The Ending Classification By Race (ECR) Act

Effective January 1, 2014, all offices of the federal government are prohibited from classifying American citizens according to their race. “Classifying” is defined as separating, sorting, or organizing persons or personal data. Exemptions include: law enforcement descriptions, and prisoner and undercover assignments. The United States Census Bureau would hereafter be barred from asking citizens’ race in any of its data collection activities, including but not limited to the Decennial Census and American Community Surveys.

This description was accompanied by a picture of the congressman who sponsored the bill, a White male standing in front of an American flag and a shelf full of legal texts. After reading the bill summary, participants rated their agreement with the following question: “I support the Ending Classification by Race (ECR) Act” (1 = strongly disagree, 7 = strongly agree).

SDO

Participants were randomly assigned to either complete the original SDO₆ scale (α_SDO-D = .94, α_SDO-E = .92), or Ho and colleagues’ (in press) new SDO₇ measure (α_SDO-D = .89, α_SDO-E = .92). Sample items from the Ho et al. scale include, “We shouldn’t try to guarantee that every group has the same quality of life” (protrait SDO-E), “We should work to give all groups an equal chance to succeed” (contrait SDO-E), “Some groups of people are simply inferior to other groups” (protrait SDO-D), and “Groups at the bottom should not have to stay in their place” (contrait SDO-D). Scores were subsequently transformed such that higher scores represented greater anti-egalitarian xz attitudes.

Results

Manipulation check

We first sought to verify that the hierarchy instability manipulation had the intended effect of increasing hierarchy threat. To test this, we regressed the manipulation check measure on hierarchy instability and inequity frame and their interaction. We observed only a significant main effect of the hierarchy instability manipulation (B = 1.35, SE B = .19, β = .45, t = 7.19, p < .001, η² = .23, 95% CI [.98, 1.72]).

Equivalence of SDO versions

Recall that participants were randomly assigned to complete either the original SDO₆ scale (Pratto et al., 1994) or the new scale developed by Ho and colleagues (SDO₇; Ho et al., in press). The SDO version did not qualify any of our effects, ts < 1, providing further evidence that the new SDO scale functions similarly to the original SDO scale. Thus, for simplicity, the analyses reported below were conducted without SDO version as a moderator.

Main analysis

Table 3 displays summary statistics for and correlations among the constructs measured in Study 2. Both hierarchy instability (−1 = unstable, 1 = stable) and inequity frame conditions (−1 = minority disadvantage, 1 = White advantage) were effects-coded. SDO-E and SDO-D were mean-centered and multiplied by hierarchy instability and inequity frame variables to create interaction terms.

Table 3.

Means and Standard Deviations of, and Correlations Among, Variables Assessed in Study 2 (N = 371).

Variable	M	SD	1	2	3
1. Color-blind bill endorsement	4.11	1.96	—	−.12*	−.10*
2. SDO-E	2.46	1.23		—	.71**
3. SDO-D	2.29	1.20			—

Note. SDO-E and SDO-D = anti-egalitarianism and dominance subcomponents of social dominance orientation, respectively.

p < .05. **p < .01.

We first sought to determine whether the inequity frame or hierarchy stability manipulations affected participants’ levels of SDO-E or SDO-D, which were measured after the manipulation. Regressing SDO-E on inequity frame, hierarchy stability, and their interaction revealed no significant effects, ts < 1.78, nor did regressing SDO-D on the same variables, ts < 1.14.

If Whites’ agenda-setting behavior reflects concerns over the ingroup’s dominant social position—and not merely the overall health of the intergroup hierarchy—high-SDO-E participants should increase their support for color-blind legislation only when their attention is focused on the White ingroup’s advantages. This amounts to a predicted three-way SDO-E × Inequity Frame × Hierarchy Instability interaction on color-blind policy support. If, however, support for color-blind policies is driven by a general desire to protect the hierarchy, hierarchy instability should increase policy support among high-SDO-E participants in both the White advantage and Minority disadvantage conditions. This pattern would manifest in a two-way SDO-E × Hierarchy Instability interaction on policy support, but without a three-way interaction involving inequity frame. As in Study 1, we did not have a strong prediction regarding the effects of SDO-D, but tested whether it interacted with either hierarchy instability or inequity frame.

To test our predictions, we regressed support for the color-blind bill on SDO-E, SDO-D, hierarchy instability, inequity frame, and their two- and three-way interactions. As shown in Table 4, the three-way SDO-E × Inequity Frame × Hierarchy Instability interaction was significant, whereas the three-way SDO-D × Inequity Frame × Hierarchy Instability interaction was marginally significant. Importantly, the two-way SDO-E × Hierarchy Instability and SDO-D × Hierarchy Instability interactions were not significant, suggesting that support for color-blind policies is not driven by a general motivation to protect the hierarchy for high-SDO-E and SDO-D participants, but rather by a concern for ingroup dominance in particular.

Table 4.

Results of Regression Predicting Support for the Color-Blind Congressional Bill in Study 2 (N = 371).

Predictor	B	SE B	β	t	p	η²	Lower 95% CI	Upper 95% CI
IF	−.06	.10	−.03	−0.56	.58	.00	−.26	.15
HS	.002	.10	.001	0.02	.66	.00	−.20	.21
SDO-E	−.30	.12	−.19	−2.46	.01	.02	−.54	−.06
SDO-D	.06	.13	.03	0.45	.66	.00	−.20	.32
IF × HS	−.009	.10	−.005	−0.09	.93	.00	−.21	.19
IF × SDO-E	−.32	.12	−.20	−2.66	.008	.02	−.56	−.08
IF × SDO-D	.26	.13	.15	1.97	.05	.01	.001	.53
HS × SDO-E	−.11	.12	−.07	−0.90	.37	.00	−.35	.13
HS × SDO-D	.19	.13	.11	1.44	.15	.01	−.07	.46
IF × HS × SDO-E	−.32	.12	−.20	−2.66	.008	.02	−.56	−.08
IF × HS × SDO-D	.24	.13	.13	1.78	.08	.01	−.03	.50

Note. IF is coded such that −1 = minority disadvantage and 1 = White advantage. For hierarchy instability, −1 = unstable and 1 = stable. CI = confidence interval; IF = inequity frame; HS = hierarchy stability; SDO-E and SDO-D = anti-egalitarianism and dominance subcomponents of social dominance orientation, respectively.

Simple effects testing revealed that the two-way Inequity Frame × Hierarchy Instability interaction was significant for high-SDO-E participants (B = −.41, SE B = 0.19, β = −.21, t = −2.17, p = .03, η² = .01, 95% CI [−.77, −.04]), see Figure 3. When ingroup advantage was salient, high-SDO-E participants tended to support color-blind policy more strongly when they also perceived the ingroup’s dominant position to be unstable (B = −.54, SE B = 0.30, β = −.27, t = −1.78 p = .08, η² = .01, 95% CI [−1.13, .06]). In contrast, hierarchy instability did not affect policy support when outgroup disadvantage was salient (B = .27, SE B = 0.22, β = .14, t = 1.25, p = .21, η² = .00, 95% CI [−.16, .71]). Thus, anti-egalitarian Whites responded to hierarchy instability by increasing their support for a color-blind policy when thinking about their ingroup’s advantages, but not when thinking about an outgroup’s disadvantages.

Figure 3.

Support for color-blind bill as a function of hierarchy instability and inequity frame for participants high in SDO-E (Study 2).

We also observed a significant, but unpredicted, Inequity Frame × Hierarchy Instability interaction for participants low in SDO-E (B = .39, SE B = 0.18, β = .20, t = 2.20, p = .03, η² = .01, 95% CI [.04, .73]), see Figure 4. When ingroup advantage was salient, low SDO-E participants were more likely to support a color-blind bill when they thought the racial hierarchy was stable than when it was unstable (B = .52, SE B = 0.26, β = .27, t = 2.04, p = .04, η² = .01, 95% CI [.02, 1.03]). Hierarchy instability did not affect policy support when outgroup disadvantage was salient (B = −.25, SE B = 0.24, β = −.13, t = −1.05, p = .30, η² = .00, 95% CI [−.73, .22]).

Figure 4.

Support for color-blind bill as a function of hierarchy instability and inequity frame for participants low in SDO-E (Study 2).

Decomposition of the SDO-D × Inequity Frame × Hierarchy Instability interaction revealed no significant Inequity Frame × Hierarchy Instability interaction among participants high in SDO-D (B = .25, SE B = .18, β = .13, t = 1.42, p = .16, η² = .01, 95% CI [−.10, .59]), nor was there an Inequity Frame × Hierarchy Instability interaction among participants low in SDO-D (B = −.27, SE B = .18, β = −.14, t = −1.48, p = .14, η² = .01, 95% CI [−.62, .09]).

Supplemental analysis

Our results suggest that high-SDO-E Whites supported color-blind policy in an effort to protect the White group’s social position. Participants may have done so, however, not because the White group is the ingroup, but because the White group is dominant. In other words, high SDO-E individuals may prefer that any dominant group stay on top, regardless of whether or not they belong to that group. If this is the case, then inequality framed as (unstable) dominant-group advantage may be more threatening than inequality framed as (unstable) subordinate-group disadvantage regardless of group membership. We tested for this possibility by conducting the same analyses as above on the data from the 123 non-White participants in our original sample. If subordinate-group members also increased their support for the color-blind bill when the dominant group’s position was unstable, it would weaken our argument that the effects we observe among Whites are driven by an interest in maintaining ingroup dominance per se.

This analysis did not produce a significant SDO-E × Inequity Frame × Hierarchy Instability interaction (B = .28, SE B = .21, β = .19, t = 1.34, p = .18, η² = .02, 95% CI [−.13, .69]), nor a significant SDO-D × Inequity Frame × Hierarchy Instability interaction, t < 1. Thus, non-White participants high in SDO-E and SDO-D did not respond to the hierarchy threat associated with an unstable hierarchy in which Whites are dominant by changing their support for a color-blind bill. These results suggest that the effects we observe among White participants are driven by their desire for the ingroup to maintain its dominance in particular, rather than by a desire for any given group to maintain a dominant position.

Discussion

Study 2 tested whether high-SDO-E and high-SDO-D Whites’ support for color-blind policy reflects a desire to maintain the White ingroup’s dominant position or a desire to maintain hierarchies in general. The results provide evidence that anti-egalitarian Whites will strategically support color-blind policies to buttress the ingroup’s advantaged position. When anti-egalitarian (high SDO-E) Whites were focused on the ingroup’s undeserved privileges (White advantage), they were more likely to support color-blind policies when the hierarchy was unstable than when it was stable. When inequity was framed as the subordinate group’s disadvantage, however, anti-egalitarian Whites did not increase support for color-blind policy in response to an unstable hierarchy.

In addition, and interestingly, low SDO-E Whites showed the opposite pattern: When focused on White advantage, they supported color-blind policies more when the hierarchy was stable than when it was unstable. In our view, it is possible that egalitarian Whites, led to believe that their ingroup’s advantages were truly intractable, saw no utility (and perhaps only the prospect of social discord) in allowing racial data to be collected. Alternatively, in a world where the dominant group’s position is firmly entrenched and current efforts to reduce its position are unsuccessful, a color-blind bill might serve as a last ditch attempt to rectify existing disparities.

Finally, counter to Study 1, we did not observe a significant interactive effect of SDO-D on color-blind policy support in Study 2. Although we address this discrepancy more in the general discussion, we submit that there might be additional factors that affect when high-SDO-D individuals will increase their opposition to subtle hierarchy-maintaining policies, such as color-blind policies. For example, in Study 1, the success of the anti-hierarchy organization suggested active challenge on the part of the subordinate group, potentially leading high-SDO-D individuals to prefer more overt, oppressive strategies to the more subtle option offered by color-blind policies. However, in Study 2, it was less clear if the instability of the hierarchy was due to active challenge on the part of the subordinate group or other contextual factors, and thus did not affect high-SDO-D individuals’ attitudes towards the color-blind bill.

To this point, we have provided evidence that Whites high in SDO-E increase their support for a subtly hierarchy-enhancing policy—color-blind policy—when they perceive the ingroup’s dominant position to be under threat. It has been our overarching contention that this behavior is driven by anti-egalitarian dominant-group members’ desires to “set the agenda” by preventing racial disparities from being known and thus addressed. Study 3 is an attempt to provide more concrete evidence that, beyond simply supporting color-blind policies as an agenda-setting technique, high-SDO-E (and, potentially, SDO-D) Whites’ selection of publicly addressable topics will be influenced by their perceptions of hierarchy threat.

Study 3

In Study 3, we leveraged increasing media attention to the 2016 U.S. Presidential elections and asked participants for their preferred list of topics to be addressed in a presidential debate. In the United States, once the two dominant political parties have each settled upon a candidate, the two presidential candidates typically square off in a series of debates, which are televised live. The formats of the debates can vary, with one common format being the “town hall,” in which everyday Americans are brought in and the candidates entertain questions from the crowd. Following this format, we asked our participants to choose the topics, of which race relations was one, they would like presidential candidates to address in a town hall debate. By choosing (or not) to include race relations in the presidential debate, participants could actively set the agenda.

Importantly, before asking them for their topic preferences, we manipulated whether participants believed social group differences in the United States were legitimate or illegitimate by exposing them to a newspaper article suggesting that the American economic system was either meritocratic or not. Most Americans believe that individuals’ life outcomes ought to be allocated on the basis of their individual abilities and efforts, rather than on the basis of factors outside of their control, such as group membership (Hochschild, 1995; Sniderman & Piazza, 1993). Indeed, the suggestion that America is not a meritocracy can lead to self-protective behaviors designed to buttress the myth of meritocracy, such as the denial of racial disparities (Knowles & Lowery, 2012). Thus, asserting that Americans’ life outcomes are either the result of individual ability (i.e., merit) or group membership (e.g., race, gender) should affect the perceived legitimacy of the social hierarchy, and thereby indicate whether the hierarchy is open to challenge.

If, as we contend, high-SDO-E Whites are increasing their support for color-blind policies in response to hierarchy threat in an attempt to set a race-free agenda, we should observe that when the hierarchy is seen as illegitimate, and thus susceptible to challenge, high-SDO-E Whites should be less likely to choose race as a discussion topic. However, when the hierarchy is seen as legitimate, high-SDO-E Whites will not see the hierarchy as under threat, and therefore not shy away from discussing race relations. We did not have strong predictions for high SDO-D participants, given the lack of coherence in findings across Studies 1 and 2.

Participants

A total of 300 participants were requested from Amazon Mechanical Turk in return for a $1 payment, for an expected yield of 200 non-Hispanic White participants (N = 100 per experimental cell). In all, 299 responses were received. Of these, we excluded 62 non-Whites or Hispanics, 5 foreign nationals, and 7 non-native English speakers. The final sample of 225 consisted of 117 men and 108 women ranging in age from 18 to 68 years (M = 33.46, SD = 11.37).

Materials and Method

Participants read that they would take part in two supposedly unrelated studies: one concerning American’s social attitudes (“Part 1”) and one concerning the 2016 Presidential election (“Part 2”).

Part 1

Our experimental manipulation of hierarchy legitimacy was embedded in the study involving social attitudes. Participants were told that prior to providing their opinions about social policies in the United States, we wanted them to consider the results of recent social science research. They were then asked to read a purportedly real newspaper article about whether personal attributes or group membership are more influential in determining life outcomes.

Legitimate hierarchy

In the legitimate condition, participants were told that “according to the [American Sociological Association], while both personal and demographic factors matter, by far the largest influence on people’s life outcomes comes from personal attributes, such as IQ, and personality traits, such as conscientiousness and ambition.” The article subsequently quoted one of the researchers as saying, “It’s really interesting that personal merit—how smart you are, how hard you work—is the primary determinant of wealth in the present-day economy.” The article concluded with the following statement: “[ . . . ] the ASA’s work suggests that group-based discrimination plays only a limited role in people’s important life outcomes.” In this way, participants were led to believe that the existing hierarchy was legitimate.

Illegitimate hierarchy

In the illegitimate condition, participants read that “[ . . . ] by far the largest influence on people’s life outcomes comes from their race, gender, and family income.” The article subsequently quoted one of the researchers as saying, “It’s really interesting that group membership—if you are male or born into a wealthy family—is the primary determinant of wealth in the present-day economy.” The article concluded with the following statement: “ . . . the ASA’s work suggests that group-based discrimination plays a major role in people’s important life outcomes.” In this way, participants were led to believe that the existing hierarchy was illegitimate.

After reading the newspaper article, participants were asked to complete Ho and colleagues’ (2012) SDO scale (α_SDO-E = .94, α_SDO-D = .94). They were then redirected to a second, supposedly unrelated study.

Part 2

After reading that the second study was about “what topics are most important in determining support for Presidential candidates,” participants were given a description of the “town hall” format used in many presidential debates. They were then told to imagine that they were responsible for choosing what topics the candidates would address during the debate. All participants were told that the topics of health care reform, education, national security, and foreign policy were guaranteed to be covered in the debate. However, due to time constraints, the candidates would be able to cover only three additional topics. The participants were to indicate which additional topics they wanted the candidates to address.

Topic choice

Participants were told to choose three topics from a given set. This set included, in randomized order, the following: immigration reform, energy policy, climate change, gun control, tax reform, and race relations. They could also write in an option. Whether or not race relations was included in participants’ chosen set of topics constituted our dependent variable.

After providing their topic choices, participants completed a set of demographic questions, were debriefed regarding the newspaper article manipulation, and thanked for their participation.

Results

We first sought to determine whether our manipulation of hierarchy legitimacy affected participants’ levels of SDO-E and SDO-D. An independent-samples t-test revealed that it did not, ts < 1.2.

Because our dependent variable was a dichotomous binary measure (1 = race included, 0 = not included), we used logistic regression to conduct our analyses. We first mean-centered the SDO-E and SDO-D measures, and then computed interaction terms by multiplying the hierarchy legitimacy variable (−1 = legitimate, 1 = illegitimate) with the mean-centered SDO-E and SDO-D measures. We then regressed topic choice on hierarchy legitimacy, SDO-E, SDO-D, and the Hierarchy Legitimacy × SDO-E and Hierarchy Legitimacy × SDO-D two-way interactions. This analysis revealed a marginally significant main effect of SDO-E, indicating that the higher the individuals’ levels of SDO-E, the more likely they were to include race relations in the discussion (see Table 5). In contrast, the higher the individuals’ levels of SDO-D, the less likely they were to include race relations. Importantly, we observed significant Hierarchy Legitimacy × SDO-E and Hierarchy Legitimacy × SDO-D interactions.

Table 5.

Results of Regression Predicting Likelihood of Including Race in a Presidential Debate in Study 3 (N = 225).

Predictor	B	SE B	Wald	Exp(b)	p
HL	−0.11	.30	0.14	0.89	.71
SDO-E	0.64	.35	3.37	1.89	.07
SDO-D	−0.88	.36	5.87	0.42	.02
HL × SDO-E	−1.10	.46	5.74	0.33	.02
HL × SDO-D	1.32	.49	7.38	3.73	.01

Note. HL = hierarchy legitimacy. SDO-E and SDO-D = anti-egalitarianism and dominance subcomponents of social dominance orientation, respectively.

Decomposition of the Hierarchy Legitimacy × SDO-E interaction indicated that Whites high in SDO-E were less likely to include race as a discussion topic when the hierarchy was illegitimate than when it was legitimate (B = −1.55, SE B = .67, Wald = 5.33, odds ratio [OR] = .21, p = .02). In contrast, low SDO-E Whites were more likely to include race when the hierarchy was illegitimate than when it was legitimate (B = 1.33, SE B = .67, Wald = 3.93, OR = 3.77, p = .05), see Figure 5. In addition, Whites high in SDO-D were more likely to include race when the hierarchy was illegitimate than when it was legitimate (B = 1.70, SE B = .74, Wald = 5.29, OR = 5.50, p = .02), but Whites low in SDO-D showed the opposite pattern (B = −1.93, SE B = .72, Wald = 7.13, OR = .15, p = .01), see Figure 6.

Figure 5.

Likelihood of including race as a discussion topic as a function of legitimacy condition and SDO-E (Study 3).

Figure 6.

Likelihood of including race as a discussion topic as a function of legitimacy condition and SDO-D (Study 3).

Discussion

The results of Study 3 provided additional evidence that anti-egalitarian Whites will attempt to set the agenda by avoiding the discussion of race in response to hierarchy threat. Specifically, we found that high-SDO-E Whites who believed the hierarchy was illegitimate were less likely to include race relations as a topic to be addressed in a nationally televised presidential debate. Oddly, high-SDO-D Whites were more likely to include it as a topic when the hierarchy was illegitimate. In addition, low SDO-E Whites responded to hierarchy threat by being more likely to include race, while those low in SDO-D were less likely to include it.

General Discussion

This article provides evidence that some members of the dominant racial group engage in behaviors designed to “set the agenda”—by supporting color-blind policies or through the exclusion of race in public discourse—to buttress the ingroup’s dominance. For these Whites, support for color-blind rules serves to “set the racial agenda” by removing the issue of racial inequality from public discourse and institutional decision-making. This maneuver promises to prevent tracking and discussion of racial inequalities, and therefore implementation of actions designed to remedy such inequalities. In Study 1, anti-egalitarian Whites increased their support for color-blind public policies when confronted with an organization that challenges the status quo. In contrast, pro-hierarchy Whites decreased their support for the policies in response to exposure to an anti-hierarchy organization. In Study 2, anti-egalitarians increased their support for color-blind public policy only when the ingroup’s advantages, but not the subordinate group’s disadvantages, were made to seem unstable. Thus, it appears that concern for ingroup dominance—and not merely the general health of the racial hierarchy—drives support for color-blind policy among anti-egalitarian Whites. Finally, in Study 3, anti-egalitarian Whites were less likely to include race relations as a debate topic in a nationally publicized forum when they perceived the hierarchy to be illegitimate. In contrast, pro-hierarchy Whites were more likely to want race relations to be discussed by presidential candidates when the hierarchy was illegitimate.

In our view, the discouragement of discussing race or collecting racial data—either by support for color-blind public policies or by excluding race as a publicly debated issue—represents an effort by dominant-group members to control the agenda with respect to intergroup inequalities. Indeed, the fact that anti-egalitarian Whites increased their support for such policies only when the ingroup’s dominance was under threat belies the oft-made claim that support for such policies will result in hierarchy-attenuating outcomes. Rather, our findings converge with recent work suggesting that pro-hierarchy dominant-group members will sometimes behave in ways that, while superficially consistent with egalitarian sentiments, actually serve to reinforce group-based hierarchy (Chow et al., 2013; Knowles et al., 2009; Unzueta, Knowles, & Ho, 2012).

Differences Between SDO-D and SDO-E

In the present studies, we relied on recent work that distinguishes between distinct dimensions of SDO (Ho et al., 2012). SDO-D reflects the belief that dominant groups should openly and aggressively oppress subordinate groups, whereas SDO-E reflects a more subtle form of dominance that opposes efforts to increase groups’ equality. Because agenda setting is not a strategy that relies on overt oppression of subordinate groups, we predicted (and found) evidence for this tactic only among Whites high in SDO-E. In contrast, Whites high in SDO-D were more likely to want to discuss race when the hierarchy was threatened (Study 3) or to oppose color-blind policies when a subordinate group was seen as successfully agitating against the hierarchy (Study 1). The data thus suggest that, whereas being a subtle anti-egalitarian (high-SDO-E) engenders attempts to remove race from the public agenda, being an unsubtle, “oppression-focused,” pro-hierarchy (high SDO-D) dominant-group member can sometimes trigger efforts to encourage the discussion of race.

The finding that the SDO-E and SDO-D dimensions appear to be leading individuals to engage in conflicting behaviors is inconsistent with prior work, which has largely treated SDO-E and SDO-D as theoretically related, but distinguishable, constructs that operate in tandem. Instead, our data suggest that desires to prevent others from gaining parity are not equivalent to desires to increase the dominant position of one’s group. In this way, our findings suggest that one difference between high SDO-E and SDO-D dominant-group members is their preference for hierarchy maintenance using approach and avoidance strategies (see Higgins, 1998). That is, while high SDO-E Whites might think that the hierarchy is more effectively maintained by avoiding the discussion or acknowledgment of race, high SDO-D Whites might think that hierarchy maintenance is more effective when racial differences are openly viewed. Indeed, sometimes the airing of legitimate disparities can result in hierarchy-enhancing outcomes; some research suggests that Whites who are exposed to information that the prison population is “more Black” are more likely to support punitive law enforcement policies (Hetey & Eberhardt, 2014). Thus, evidence of extreme racial differences can sway policy preferences in ways that perpetuate the inequality. Future research might explore more closely when intergroup differences will be emphasized or avoided, and whether differences in SDO-E and SDO-D might also stem from preferences regarding how hierarchies can be achieved and maintained.

Although our primary focus was on anti-egalitarian Whites, our findings suggest that pro-egalitarians and anti-hierarchical individuals also respond to threats to the hierarchy. For example, in Study 1, low-SDO-E participants decreased their support for color-blind public policy when they read about an anti-hierarchy organization, but in Study 2, low SDO-E participants displayed the highest levels of support for color-blind policies when told that the ingroup’s advantages were stable. Similarly, in Study 1, low SDO-D participants increased their support for color-blind policies when they read about a successful anti-hierarchy organization (but did not differentially respond to (un)stable ingroup advantage in Study 2). Low SDO-E and SDO-D Whites also changed their willingness to include race relations as a debate topic in response to hierarchy threat. Speculatively, drawing on prior work showing that individuals’ construals of color-blind ideology can fluctuate between its hierarchy-enhancing and hierarchy-attenuating forms (Knowles et al., 2009), it may be that Whites’ expectations about the consequences of color-blind policies change depending on contextual cues as well. For example, low SDO-E Whites who were told that the ingroup’s advantage was unlikely to change might have believed that, since attending to race has not wrought the desired changes, it might be helpful to try the strategy of not attending to race. Likewise, low SDO-D Whites who believe that anti-hierarchy organizations are succeeding might believe that color-blind policies will build upon the success of the anti-hierarchy organizations, bringing society yet closer to a less hierarchical ideal. Indeed, it may be that the perceived starting point—that racial inequity exists but is intractable versus exists but is being addressed versus has been successfully addressed—may significantly influence dominant-group members’ expectations regarding the effects of color-blind policies.

The Interdependent Nature of Hierarchy Maintenance and the Specter of Illegitimacy

Hierarchies are much easier to maintain when the relationship between the dominant and subordinate groups is consensual (Gould, 2002), leading pro-hierarchy dominant-group members to be sensitive to the subordinate-group experience (Chow et al., 2013; Knowles et al., 2009). In keeping with these prior findings, our results, too, suggest that the specter of hierarchy threat never strays far from the minds of some White Americans, and can lead them to be highly attuned to the attitudes of subordinate groups toward the hierarchy. For example, in Study 1, simply being told that an anti-hierarchy organization exists was sufficient to induce a sense of hierarchy threat among Whites high in SDO-E such that they increased their support for color-blind policies. Moreover, in Study 2, we observed the same increase in color-blind public policy support among high-SDO-E participants only in the unstable White advantage condition. Taken together, our results suggest that simply knowing that subordinate-group members have their own hierarchical agenda—as evidenced by their attempts to organize and engage in political action through an anti-hierarchy organization—can lead Whites to contend with the possibility that their ingroup’s dominance is unwarranted and subject to change.

Finally, we do not believe that our effects are purely constrained to the Black/White U.S. relationship. From our perspective, agenda setting is a tactic potentially to be used by any dominant group when the ingroup’s dominance is threatened. For example, in Study 2, high-SDO-E Whites’ increase in support for color-blind policies was only in response to White advantage, but not minority disadvantage, suggesting that the threat most concerning for dominant-group members is loss of the ingroup’s dominant position in particular. In addition, our manipulation of hierarchy legitimacy in Study 3 did not reference any particular racial group, nor did we specify which race relations would be discussed in the Presidential debate. Moreover, historical and cross-cultural analyses suggest that agenda setting is a tactic that has been used in the past and in other countries, involving other social groups. Thus, we believe that agenda setting is a general-use maneuver available to any anti-egalitarian dominant group.

Conclusion

In the present work, we have focused on agenda setting as a form of negation—of removing topics that run counter to the dominant group’s interests. We would like to note that preventing the collection of racial data—as would be done under color-blind public policy—such that racial differences cannot be detected represents a more insidious form of agenda setting than specifically precluding topics for discussion. This form of agenda setting not only prevents the addressing of intergroup differences, but prevents the very formation of knowledge that would enable subordinate groups to know that there are intergroup differences that could be addressed. In this way, support for color-blind policy resembles the third dimension of power (Lukes, 2005), in which latent conflicts between the dominant and subordinate groups are not observed.

Footnotes

Acknowledgements

We are grateful to Jeff Galak, Brian Lowery, and Miguel Unzueta for helpful comments on earlier versions of the article.

Authors’ Note

Rosalind M. Chow and Eric D. Knowles contributed equally to this paper.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

The online supplemental material is available at

Notes

References

Anderson

M. J.

(1988). The American census: A social history. New Haven, CT: Yale University Press.

Apfelbaum

E. P.

Pauker

Sommers

S. R.

Ambady

(2010). In blind pursuit of racial equality? Psychological Science, 21, 1587-1592. doi:10.1177/0956797610384741

Apfelbaum

E. P.

Sommers

S. R.

Norton

M. I.

(2008). Seeing race and seeming racist? Evaluating strategic colorblindness in social interaction. Journal of Personality and Social Psychology, 95, 918-932. doi:10.1037/a0011990

Bacharach

Baratz

M. S.

(1962). Two faces of power. American Political Science Review, 56, 974-952.

Bonilla-Silva

(2003). Racism without racists: Color-blind racism and the persistence of racial inequality in the United States. Lanham, MD: Rowman & Littlefield.

Branscombe

N. R.

(1998). Thinking about one’s gender group’s privileges or disadvantages: Consequences for well-being in women and men. British Journal of Social Psychology, 37, 167-184.

Buhrmeister

Kwang

Gosling

S. D.

(2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6, 3-5. doi:10.1177/1745691610393980

Carr

L. G.

(1997). “Color-blind” racism. Thousand Oaks, CA: SAGE.

Chow

R. M.

Lowery

B. S.

Hogan

C. M.

(2013). Appeasement: Whites’ strategic support for affirmative action. Personality and Social Psychology Bulletin, 39, 332-345. doi:10.1177/0146167212475224

10.

Gould

R. V.

(2002). The origins of status hierarchies: A formal theory and empirical test. American Journal of Sociology, 107(5), 1143-1178.

11.

Hetey

R. C.

Eberhardt

J. L.

(2014). Racial disparities in incarceration increase acceptance of punitive policies. Psychological Science, 25(10), 1949-1954. http://doi.org/10.1177/0956797614540307

12.

Higgins

E. T.

(1998). Promotion and prevention: Regulatory focus as a motivational principle. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (pp. 1-46). San Diego, CA: Academic Press.

13.

A. K.

Sidanius

Kteily

Sheehy-Skeffington

Pratto

Henkel

K. E.

. . . Stewart

A. L.

(in press). The nature of social dominance orientation: Theorizing and measuring preferences for intergroup inequality using the new SDO7 scale. Journal of Personality and Social Psychology.

14.

A. K.

Sidanius

Pratto

Levin

Thomsen

Kteily

Sheehy-Skeffington

(2012). Social dominance orientation: Revisiting the structure and function of a variable predicting social and political attitudes. Personality and Social Psychology Bulletin, 38, 583-606. doi:10.1177/0146167211432765

15.

Hochschild

J. L.

(1995). Facing up to the American Dream: Race, class, and the soul of the nation. Princeton, NJ: Princeton University Press.

16.

HoSang

D. M.

(2010). Racial propositions: Ballot initiatives and the making of postwar California. Berkeley: University of California Press.

17.

Jost

J. T.

Thompson

E. P.

(2000). Group-based dominance and opposition to equality as independent predictors of self-esteem, ethnocentrism, and social policy attitudes among African Americans and European Americans. Journal of Experimental Social Psychology, 36, 209-232. doi:10.1006/jesp.1999.1403

18.

Knowles

E. D.

Lowery

B. S.

(2012). Meritocracy, self-concerns, and Whites’ denial of racial inequity. Self and Identity, 11(2), 202–222. http://doi.org/10.1080/15298868.2010.542015

19.

Knowles

E. D.

Lowery

B. S.

Hogan

C. M.

Chow

R. M.

(2009). On the malleability of ideology: Motivated construals of color blindness. Journal of Personality and Social Psychology, 96, 857-869. doi:10.1037/a0013595

20.

Kteily

Saguy

Sidanius

Taylor

D. M.

(2013). Negotiating power: Agenda ordering and the willingness to negotiate in asymmetric intergroup conflicts. Journal of Personality and Social Psychology, 105, 978-995. doi:10.1037/a0034095

21.

Kugler

M. B.

Cooper

Nosek

B. A.

(2010). Group-based dominance and opposition to equality correspond to different psychological motives. Social Justice Research, 23, 117-155. doi:10.1007/s11211-010-0112-5

22.

Loveman

(2014). National colors: Racial classification and the state in Latin America. New York, NY: Oxford University Press.

23.

Lowery

B. S.

Chow

R. M.

Knowles

E. D.

Unzueta

M. M.

(2012). Paying for positive group esteem: How inequity frames affect whites’ responses to redistributive policies. Journal of Personality and Social Psychology, 102, 323-336. doi:10.1037/a0024598

24.

Lowery

B. S.

Knowles

E. D.

Unzueta

M. M.

(2007). Framing inequity safely: Whites’ motivated perceptions of racial privilege. Personality and Social Psychology Bulletin, 33, 1237-1250. doi:10.1177/0146167207303016

25.

Lowery

B. S.

Unzueta

M. M.

Knowles

E. D.

Goff

P. A.

(2006). Concern for the ingroup and opposition to affirmative action. Journal of Personality and Social Psychology, 90, 961-974. doi:10.1037/0022-3514.90.6.961

26.

Lukes

(2005). Power: A radical view (2nd ed.). New York: Palgrave Macmillan.

27.

Norton

M. I.

Sommers

S. R.

Apfelbaum

E. P.

Pura

Ariely

(2006). Color blindness and interracial interaction: Playing the political correctness game. Psychological Science, 17, 949-953.

28.

Overbeck

J. R.

Jost

J. T.

Mosso

C. O.

Flizik

(2004). Resistant versus acquiescent responses to ingroup inferiority as a function of social dominance orientation in the USA and Italy. Group Processes and Intergroup Relations, 7, 35-54.

29.

Parents Involved in Community Schools v. Seattle School District No. 1. (2007). U.S.

30.

Plaut

V. C.

Thomas

K. M.

Goren

M. J.

(2009). Is multiculturalism or color blindness better for minorities? Psychological Science, 20, 444-446.

31.

Powell

A. A.

Branscombe

N. R.

Schmitt

M. T.

(2005). Inequality as ingroup privilege or outgroup disadvantage: The impact of group focus on collective guilt and interracial attitudes. Personality and Social Psychology Bulletin, 31, 508-521.

32.

Pratto

Liu

J. H.

Levin

Sidanius

Shih

Bachrach

Hegarty

(2000). Social dominance orientation and the legitimization of inequality across cultures. Journal of Cross-Cultural Psychology, 31, 369-409.

33.

Pratto

Sidanius

Stallworth

L. M.

Malle

B. F.

(1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741-763. doi:10.1037/0022-3514.67.4.741

34.

Pratto

Stallworth

L. M.

Sidanius

(1997). The gender gap: Differences in political attitudes and social dominance orientation. British Journal of Social Psychology, 36, 49-68.

35.

Pratto

Tatar

D. G.

Conway-Lanz

(1999). Who gets what and why: Determinants of social allocations. Political Psychology, 20, 127-150. Retrieved from http://www.blackwellpublishing.com/journal.asp?ref=0162-895X&site=1

36.

Proposition 54. (2003). Smart Voter. Retrieved from http://www.smartvoter.org/2003/10/07/ca/state/prop/54/

37.

Richeson

J. A.

Nussbaum

R. J.

(2004). The impact of multiculturalism versus color-blindness on racial bias. Journal of Experimental Social Psychology, 40, 417-423.

38.

Robinson

P. M.

Connerly

Douglass

R. E.

(2002). Strength in numbers: Race and the census. Uncommon Knowledge with Peter Robinson. Retrieved from http://www.hoover.org/multimedia/uncommon-knowledge/26953

39.

Rousseau

J.-J.

(2009). Discourse on the origin of inequality. New York, NY: Oxford University Press. (Original work published 1755)

40.

Saguy

Dovidio

J. F.

Pratto

(2008). Beyond contact: Intergroup contact in the context of power relations. Personality and Social Psychology Bulletin, 34, 432-445. doi:10.1177/0146167207311200

41.

Saguy

Kteily

(2014). Power, negotiations, and the anticipation of intergroup encounters. European Review of Social Psychology, 25(1), 107-141. doi:10.1080/10463283.2014.957579

42.

Saguy

Pratto

Dovidio

J. F.

Nadler

(2009). Talking about power: Group power and the desired content of intergroup interactions. In Demoulin

Leyens

J.-P.

Dovidio

J. F.

(Eds.), Intergroup misunderstandings: Impact of divergent social realities (pp. 213-232). New York: Psychology Press.

43.

Seitz-Wald

(2010). Beck: The census is the government’s attempt to “increase slavery.” ThinkProgress. Retrieved from http://thinkprogress.org/politics/2010/03/09/85755/beck-census-slavery/

44.

Sidanius

Pratto

(1999). Social dominance: An intergroup theory of social hierarchy and oppression. New York, NY: Cambridge University Press.

45.

Sniderman

P. M.

Piazza

(1993). The scar of race. Cambridge, MA: Harvard University Press.

46.

Sperry

(2015, July 18). Obama collecting personal data for a secret race database. New York Post. Retrieved from http://nypost.com/2015/07/18/obama-has-been-collecting-personal-data-for-a-secret-race-database/

47.

Tajfel

Turner

J. C.

(1979). An integrative theory of intergroup conflict. In Austin

W. G.

Worschel

(Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks/Cole.

48.

Unzueta

M. M.

Knowles

E. D.

(2012). Diversity is what you want it to be: How social-dominance motives affect construals of diversity. Psychological Science, 23, 303-309. doi:10.1177/0956797611426727

49.

Whitehead

J. W.

(2010). The race question on the 2010 census raises serious questions. The Huffington Post. Retrieved from http://www.huffingtonpost.com/john-w-whitehead/the-race-question-on-the_b_502903.html

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.48 MB