The effects of racial status threat on White Americans’ support for Donald Trump: Results of five experimental tests

Abstract

Prior theory and research suggest that threats to the status of White Americans may increase support for Donald Trump. Consistent with this, one previous experiment conducted in early 2016 documented this effect (Major et al., 2018), finding that making salient the declining White majority in the United States increased support for Trump’s presidential candidacy among White participants with high levels of ethnic identification. We report the results of five very similar experiments (total N = 3,076) also conducted in 2016, including one conducted on a national probability sample. The first experiment (conducted in January 2016) found that racial status threat increased Whites’ support for Trump. The other four (conducted from February to October 2016), however, all found null results, and an internal meta-analysis of the five studies found no significant main effect overall. Additionally, none of the studies found an interaction of racial demographic shift and ethnic identification in which racial demographic shift increased Trump support among high-identifying Whites. We conclude by discussing a variety of potential explanations for our findings, including (a) that racial status threats did not increase Whites’ Trump support, (b) that racial status threats increased Trump support early in the 2016 election cycle, but the role of this factor in Trump support declined over time, or (c) that this pattern is an example of a broader tendency of declining experimental treatment effects on candidate support over time in campaigns (Kalla & Broockman, 2018).

Keywords

demographic changes diversity group status threat identity intergroup relations political psychology prejudice race/ethnicity racial status threat 2016 presidential election

What explains Donald Trump’s surprising success as a presidential candidate and his subsequent de facto leadership of the Republican Party? Prominent explanations feature the role of political disenchantment, a “cultural backlash,” and the recent widespread success of right-wing populism in many Western democracies (Lüders et al., 2020). Other analyses highlighting the roles of disinformation, conspiracy theories (DiGrazia, 2017; Sawyer, 2020, 2021), and conspiratorial thinking (van der Linden et al., 2021) are bolstered by the continued belief among many Americans that Donald Trump won the 2020 presidential election, that the election was stolen, and that he would be reinstated as president in the coming months. Many Trump supporters report narratives of grievance and conflict, for example suffering at the hands of elites, conflict between conservatives and liberals, or concerns about nebulous antagonists such as Antifa and the “deep state.”

But perhaps the most prominent explanations of the support for Trump’s 2016 presidential candidacy emphasize the declining standing of the groups that Trump supporters are members of, specifically, the threats to the status of a White (generally Christian) majority that has become accustomed to a position of high status and cultural centrality in America (Blumer, 1958; Bobo, 1999). It is plausible that Trump’s rhetoric around immigration and “American exceptionalism,” his criticisms of antiracist activism, and his call for a return to traditional cultural arrangements would uniquely resonate with many White Americans, potentially addressing their concerns about declining racial standing and cultural centrality. As Bonikowski writes of Trump’s “resentful and exclusionary” rhetoric:

What is at stake in these appeals is the collective status of the group itself—of native-born, White, lower-middle-class people, whose self-understanding as the ethnocultural core of the nation, with all the dignity and respect that entails, is under persistent threat. (2017, p. S203)

Consistent with this line of reasoning, research finds that the narrative of a looming “majority-minority America”—the decline of the White population to a numerical minority (cf. Alba, 2018)—evokes feelings of threat (Outten et al., 2012) and group status concerns (Outten et al., 2018). Researchers have further documented effects of declining racial status on a number of political and racial attitudes of White Americans (Craig et al., 2018). For example, studies making salient the declining White majority to White Americans have reported effects such as a “conservative shift” in policy support (Bai & Federico, 2021; Craig & Richeson, 2014b; Osborn et al., 2020), increased antipathy toward outgroups (Craig & Richeson, 2014a), and reduced support for welfare programs when those programs are perceived to disproportionately benefit racial minorities (Wetts & Willer, 2018).

One prominent study conducted in 2016 (Major et al., 2018) found that viewing a report documenting the declining White majority increased support for Donald Trump among Whites who identified more strongly with their ethnicity. Here we report the results of five very similar tests of the effect of a racial demographic shift manipulation on support for Donald Trump that we conducted between January and October 2016. In the sections that follow, we begin by elaborating on group position theory (Blumer, 1958; Bobo, 1999; Zou & Cheryan, 2017) and related group-based status dynamics. We then turn specifically to the potential role of racial status threat in support for Donald Trump before reporting the results of our five studies. We conclude by discussing a number of possible explanations for our pattern of results.

Group Position Theory and Racial Demographic Shifts

What drives attitudes toward different groups, and how do those attitudes, in turn, influence group members’ behaviors? Group position theory (Blumer, 1958; Bobo, 1999) posits that we should not understand racial prejudice as a purely individual-level phenomenon but instead as a product of the relative positions of different groups within a society. Perceptions of fundamental differences between racial/ethnic groups reinforce boundaries, with groups arranged hierarchically in terms of widely held perceptions of each group’s relative access to power, resources, and recognition.¹ According to this perspective, Whites occupy the dominant racial group position in American society and have the greatest access to resources and power. Consequently, they also have, on average, the most to lose. Negative attitudes and behavior toward outgroups arise from perceptions that group boundaries are being crossed, that groups are making claims to which they are not entitled, or that Whites are not getting their due (Bobo, 1983). Other research points to similar competitive dynamics among subordinate groups (Bobo & Hutchings, 1996; cf. Craig & Richeson, 2012).

Membership in dominant groups offers other benefits beyond access to material, symbolic, and political resources. First, dominant groups are also agentic and thus can restore a threatened sense of personal control (Fritsche et al., 2013). Accordingly, individuals may identify with the most powerful identity to which they can legitimately lay claim (Abascal, 2015).

Another advantage of dominant group membership is the experience of cultural centrality and relative prestige in subnational regions. Built and natural environments reflect histories of group relations and conflict. Group identity affects how people relate to everyday environments (Lewicka, 2011; Lewicka et al., 2019), and views of place often entail perceptions of whether a group belongs or not (Wnuk et al., 2021). Dominant groups disproportionately shape the places they inhabit, leading to phenomena such as segregation, redlining, and conflict over symbols like monuments (Benjamin et al., 2020). Thus, membership in a dominant group offers diverse benefits such as feelings of control, self-esteem, belongingness, and cultural prototypicality.

These considerations suggest that threatening Whites’ numerical majority may evoke defensive reactions for many reasons. Accordingly, several researchers have studied the effects of a racial demographic shift manipulation, showing (typically White American) participants in one experimental condition demographic projections suggesting that Whites will become a numerical minority within a few decades. Consistent with group position theory, these studies have demonstrated effects among White Americans, such as a “conservative shift” in social policy attitudes (Craig & Richeson, 2014b; Wetts & Willer, 2018), increased racial bias or antipathy (Bai & Federico, 2020; Craig & Richeson, 2014a; Outten et al., 2012, 2018), increased support for the Tea Party movement (Willer et al., 2016), an increased sense of victimhood (Craig & Richeson, 2017), and a narrowed definition of “Whiteness” (Abascal, 2020).

Research has also identified factors that may moderate racial status threat effects. Outten et al. (2018, Study 1) found that the effect of racial demographic shift on anger and fear toward minorities was moderated by the perceived legitimacy of Whites’ privileged position within American society. White participants who felt Whites’ general advantages in society were legitimate responded to threats with greater fear and anger toward minorities in the racial demographic shift condition, as group position theory would predict. Conversely, Whites who did not believe that Whites’ advantages were legitimate responded with less anger and fear. Further, Perkins et al. (2022) found evidence of a conservative shift following a racial demographic shift treatment, though only among participants with a zero-sum view of the economy. Participants with non-zero-sum views of the economy instead exhibited a liberal shift when threatened, though this effect was nonsignificant. Osborn et al. (2020) found a conservative shift among White participants presented with a racial status threat, but only if they were also primed to think of multiculturalism.

Threats to Group Position and Trump Support

A number of scholars and pundits have theorized that support for Donald Trump’s 2016 presidential campaign was rooted, at least in part, in White Americans’ feelings of racial status threat (Mutz, 2018b; Smith & Hanley, 2018; Thompson, 2016; cf. Morgan, 2018). From the perspective of group position theory, the question is, did Donald Trump’s candidacy resonate with and attract Whites who feel their group status is imperiled? It is possible that a variety of Donald Trump’s statements and positions—such as his comments about undocumented immigrants, leadership of the “birther” movement, vilification of Muslims, and criticisms of antiracist activism—mean that his candidacy could have been seen by many as a means for defending the standing of Whites in America. In addition, much of his campaign’s emphasis on a return to embattled cultural traditions—most prominently, the “make America great again” slogan—may have resonated among White Americans mourning for a time when their racial standing was more assured (Hochschild, 2016).

Qualitative evidence is consistent with this account of the motivations of many Trump supporters. For example, Hochschild’s study of White Tea Party supporters—the vast majority of whom would subsequently support Donald Trump’s presidential candidacy—found that many viewed themselves as thwarted in their efforts to achieve the American dream by minorities and immigrants who they felt were granted unfair privileges (2016). Hochschild argues that Trump offered solutions to the problems at the heart of their “deep story” and describes Trump’s promises to restore the status of certain groups as making him the “identity politics candidate for white men” (2016, p. 230). Correlational evidence from nationally representative surveys also supports the intuition that Trump support was uniquely rooted in Whites’ racial resentment (Sides et al., 2018). One analysis of panel data found that animus towards racial/ethnic minorities in 2011 predicted support for Donald Trump years later but did not predict support for other politicians or for the Republican or Democratic parties (Mason et al., 2021).

Present Research

Drawing on prior work on the effects of racial status threats, Major et al. (2018) explored whether a racial threat manipulation would affect support for Donald Trump’s first presidential campaign. In March of 2016, Major et al. (2018) conducted a study of the effects of racial status threat on Whites’ support for Donald Trump’s presidential candidacy. In an online survey experiment (N = 376), they presented White study participants with the same manipulation of information on the declining White majority used by Craig and Richeson (2014a, Study 2) and then measured respondents’ support for Donald Trump, opposition to political correctness, support for anti-immigrant policies, and ethnic identification. Major et al. found that Whites high in ethnic identification who were randomly assigned to read about the rising minority population in the United States reported greater support for Donald Trump’s presidential candidacy, greater support for anti-immigrant policies, and greater opposition to political correctness. Further, these results were mediated by high-identifying Whites’ heightened perceptions of group status threat.²

We independently arrived at the same hypothesis, reasoning that the same racial status threat dynamic previously linked with political conservatism and views on race-related issues would also be linked to support for Trump in the 2016 election cycle. To test this, we conducted five experiments (total N = 3,076) with very similar designs to that of Major et al. (2018), including an experimental treatment that made salient to participants the rising minority population in the United States, which we refer to as a racial demographic shift condition. Like Major et al. (2018), we also included a measure of support for Trump’s candidacy and, in Studies 1–4, a measure of ethnic identification. While we found evidence for a main effect of racial demographic shift on Trump support in our first experiment (January 2016), we did not find this main effect in any of the following experiments, including a study with a nationally representative sample conducted in October 2016. In none of our studies that included a measure of ethnic identification (Studies 1–4) did we find the interaction of racial demographic shift and ethnic identification on Trump support reported in Major et al. Internal meta-analyses do not suggest the racial demographic shift manipulation had an effect on our measure of support for Donald Trump. Further, we found no effect of an economic threat manipulation on support for Donald Trump (Study 5). Next, we detail the methods and results of these studies, after which we conclude by discussing possible explanations that can account for our results and those of Major et al. (2018).

Studies 1–4

To explore whether support for Donald Trump in 2016 was motivated in part by this racial threat dynamic, we conducted four experiments on Mechanical Turk between January and April 2016. Although they were conceived independently of, and conducted at the same time as, Major et al.’s study (2018), our studies could be viewed as conceptual replications. We highlight differences in designs and results in order to propel the literature on this racial status threat dynamic and research using a racial demographic shift paradigm forward.

Much prior research has manipulated racial status threat in fairly similar ways, yet studies differ with respect to how they operationalize the threat and the content of the control condition. Major et al. (2018) contrast the future decline of Whites as the numerical majority (racial demographic shift) to increasing geographic mobility. This approach is similar to the approach taken by Craig and Richeson (2014a, Study 3; 2014b, Studies 2–3) and Bai and Federico (2021). Other studies instead feature a contrast between a racial demographic shift condition and control conditions that also make race/ethnicity salient. Many studies contrast racial demographic shift to the current demographic make-up of the country or region (Abascal, 2020; Craig & Richeson, 2014a, Study 1; Outten et al., 2012, Study 1; Wetts & Willer, 2018, Study 2; Willer et al., 2016, Studies 2 and 4). Others have contrasted projections of the decline of Whites as the majority to projections that Whites will maintain their majority status (Danbold & Huo, 2015, Study 2; Outten et al., 2012, Study 2; Outten et al., 2018, Studies 1–2). Craig and Richeson (2014a, Study 2) contrasted racial demographic shift in the United States to racial demographic shift in the Netherlands to test whether Americans were uniquely threatened by the loss of Whites’ majority status “at home.” Bai and Federico (2020, Study 2) instead contrasted a decline of the global White population to a growing global White population.

In Studies 1–4, we examine the effects of reading a short report about the coming racial demographic shift compared to a report depicting relative demographic stability over a shorter time (as in Abascal, 2020; Wetts & Willer, 2018; Willer et al., 2016). This approach has the advantage of controlling for anything that may be primed by information about the racial/ethnic composition of the United States generally, such as simply making race/ethnicity salient. In contrast, studies with control conditions that do not feature race/ethnicity cannot establish that their results are due to priming changing demographics, as opposed to demographics in general. In Table 1, we present differences between the designs of Studies 1–4, our fifth study, and the studies reported by Major et al. (2018) and Osborn et al. (2020).

Table 1.

Differences between designs of Studies 1–4, Study 5, and two similar studies conducted in 2016.

	Studies 1–4	Study 5	Major et al. (2018)	Osborn et al. (2020, Study 2)
Period of data collection	January 22 to April 24, 2016	October 6 to 16, 2016	March 24 to 27, 2016	November 7–8, 2016
Observations	1,023	2,053	376	256
Data source	Mechanical Turk	GfK (via TESS)	Mechanical Turk	Mechanical Turk
Experimental condition(s)	Racial demographic shift	Racial demographic shift, economic threat	Racial demographic shift	Racial demographic shift
Control condition	Short-term racial demographic stability	Stable geographic mobility	Increasing geographic mobility	Increasing geographic mobility
Factorial design?	No	No	No	Yes, interacted with colorblindness, multiculturalism, or polyculturalism condition
Outcome	Favorability (1 = strongly unfavorable, 5 = strongly favorable) and support (1 = strongly oppose, 5 = strongly support)	Support (1 = strongly oppose, 7 = strongly support)	Positivity (0 = very cold, 100 = very warm) and likelihood of voting for candidate (1 = extremely unlikely, 7 = extremely likely)	Feelings toward Trump (0 = no warmth at all, 100 = extreme warmth)
Effect of racial demographic shift?	In Study 1 (January 22–23)	No	No	In multiculturalism condition
Interaction with ethnic identification?	In Study 4 (April 22–24)	-	Yes	-

Methods

Timing and participants

Participants in Studies 1–4 were recruited from a large, prescreened panel of respondents from Mechanical Turk (MTurk) maintained by the Laboratory for Social Research at Stanford University.³ In total, we recruited 1,305 participants from this panel who identified as moderate or conservative (i.e., who scored 4–7 on a 7-point scale; 1 = extremely liberal, 7 = extremely conservative). Of these, 1,055 participants self-identified as White, of whom 32 were excluded due to missing data. This results in a total analytical sample of 1,023 White participants for Studies 1–4. No participant took part in more than one study. We present descriptive statistics and the timing of data collection for each of these studies in Table 2.

Table 2.

Descriptive statistics for analytical samples.

	Study 1	Study 2	Study 3	Study 4	Study 5
	January 22–23	February 5–9	February 24–25	April 22–24	October 6–16
Observations	268	267	249	239	2,053
Education (% with bachelor’s degree)	43.7%	49.1%	44.6%	43.9%	35.7%
Gender (% male)	50.0%	49.1%	44.6%	43.5%	49.3%
Mean age (SD)	40.84 (13.36)	38.16 (11.82)	40.57 (13.22)	40.51 (13.22)	53.14 (16.86)
Mean political conservatism (SD)	5.09 (0.93)	4.99 (0.94)	5.11 (0.96)	5.06 (0.96)	4.22 (1.59)
Mean support for Donald Trump (SD)	2.89 (1.60)	2.85 (1.62)	3.01 (1.53)	2.75 (1.57)	3.23 (2.29)
Mean perceived group status threat (SD)	4.51 (1.56)	4.47 (1.52)	4.68 (1.55)	4.65 (1.48)	4.29 (1.63)
Mean perceived racial common fate (SD)	3.28 (1.61)	3.30 (1.58)	3.22 (1.59)	3.15 (1.62)	-
Mean ethnic identification (SD)	4.52 (1.60)	4.33 (1.64)	4.57 (1.77)	4.18 (1.75)	-
Mean racial resentment (SD)	4.40 (1.36)	4.53 (1.30)	4.52 (1.32)	4.19 (1.23)	-

Note. Standard deviations of continuous measures in parentheses.

Procedure

In each study, participants were randomly assigned to one of two conditions. In the racial demographic shift condition, participants read information and observed a figure concerning projected demographic changes between 1960 and 2060. The information and figure emphasized the decline of Whites as the majority ethnic group in the United States in recent years and projections for the coming decades. In addition to increasing trend lines for individual minority groups, the figure had a trend line for “all non-Whites” that surpassed the trend line for Whites. These materials were intended to induce racial status threat and evoke a racial status defense response. Participants in this condition then answered questions about the outcomes of these changes, again emphasizing Whites’ decline as the majority ethnic group. In contrast, participants in the control condition read information and saw a similar figure concerning demographic changes between 2000 and 2020, depicting only minor changes and noting that Whites would remain the majority ethnic group in the United States through 2020. Participants in this condition additionally answered comprehension questions highlighting the relative lack of change in the racial/ethnic composition of the United States during this time. Figure 1 illustrates the graphs included in each condition in these four studies. A replication package for this paper is available at OSF (https://osf.io/9wu3p/)

Figure 1.

Images included in the control (left) and racial demographic shift (right) conditions: Studies 1–4.

Measures

Participants provided information concerning their gender, year of birth, household income, educational attainment, and political orientation in a prescreen survey. We recoded educational attainment to match the coding in our Study 5, binning education into the following categories: “less than high school,” “high school,” “some college,” and “bachelor’s degree or higher.” We calculated age using 2016 minus the provided year of birth because participants entered the panel at different points in time. Political conservatism was measured using the question, “In general, do you consider yourself liberal or conservative?” (1 = extremely liberal, 7 = extremely conservative).

We measured four potential mediators. First, we measured perceived group status threat using the item, “If they increase in status, racial minorities are likely to reduce the influence of White Americans in society” (1 = strongly disagree, 7 = strongly agree), which was drawn from prior work (Craig & Richeson, 2014b). We note that Major et al. (2018) used a four-item measure of this construct, and multi-item constructs tend to have better psychometric properties. Second, we measured common fate using one item from the Perceived Common Racial Fate Scale (Lowery et al., 2007), “My opportunities in life are tied to those of my racial group as a whole” (1 = strongly disagree, 7 = strongly agree), in line with prior research (Craig & Richeson, 2014b). Third, we measured ethnic identification—here, the importance to a participant’s identity of their racial/ethnic identity—using four items that are nearly identical to those used by Major et al. (see their supplemental material). Like those used by Major and colleagues, these items are based on the Identity Subscale of the Collective Self-Esteem Scale (Luhtanen & Crocker, 1992). Participants indicated the extent to which they agreed with each item (e.g., “In general, belonging to my race/ethnicity is an important part of my self-image”; 1 = strongly disagree, 7 = strongly agree). The ethnic identification items exhibited high reliability in each study (all Cronbach’s alphas > .86) and were averaged to form a composite. Finally, we measured racial resentment using four items from the work ethic, excessive demands, and deservingness themes of the Symbolic Racism 2000 Scale (Henry & Sears, 2002; e.g., “Over the past few years, Blacks have gotten more economically than they deserve”; 1 = strongly disagree, 7 = strongly agree). These items also exhibited high reliability in each study (all Cronbach’s alphas > .78) and were averaged to form a composite. We standardized these measures for ease of interpretation.

We measured support for Donald Trump using two items. In the first, we asked participants, “To what extent do you have an unfavorable or favorable opinion of each of the following candidates” (1 = strongly unfavorable, 5 = strongly favorable), followed by a list of candidates. Next, we gave participants the prompt, “To what extent do you support or oppose each of the following candidates” (1 = strongly oppose, 5 = strongly support), followed by the same list of candidates. In each of these studies, we asked about Donald Trump, Ted Cruz, Hillary Clinton, and Bernie Sanders. Additionally, we asked about Jeb Bush in the first three studies, while we asked about John Kasich only in Study 4. The two measures of support for Donald Trump were highly correlated within each study (rs = .97, .95, .94, .95 in Studies 1–4, respectively) and were averaged to form a composite. Our strategy for measuring support for Donald Trump thus differed from the feeling thermometers used by Major et al. (2018) and Osborn et al. (2020) and the measure of expected likelihood of voting for Donald Trump in the general election (1 = extremely unlikely, 7 = extremely likely) also used by Major et al. (2018). In our online supplemental material, we assess the robustness of our findings using different measures of support for Donald Trump: a nominal item asking which candidate the participant most supports out of a list of candidates, agreement with five public statements made by Donald Trump between 2012 and 2016, and our main measure of support minus or divided by average support for the alternative candidates. These alternate analyses produced very similar results to those reported in the main text.

Results

Main effects

To assess the potential effects of racial status threat on support for Donald Trump, we regressed Trump support on racial demographic shift (0 = control, 1 = racial demographic shift) and a set of covariates including gender, age, household income, educational attainment, and political conservatism. This multivariate approach was chosen to obtain less biased estimates of the effect of the treatment.⁴

Table 3 depicts the results. We found a significant effect of racial demographic shift on support for Donald Trump in Study 1 (January 22–23, 2016), in which exposure to the racial demographic shift is associated with a statistically significant increase in support for Donald Trump of approximately 0.31 standard deviations, net of the covariates. However, we found no significant effects of racial demographic shift on support for Donald Trump in Studies 2, 3, or 4. Robustness checks using alternative measures of support yield the same pattern of results: we found comparable, significant effects of racial demographic shift on support for Donald Trump and agreement with statements made by Donald Trump in Study 1, but not in subsequent studies, where coefficients appear to be attenuated or even change sign (Tables S5, S6, S9, and S12).

Table 3.

Effect of racial demographic shift on support for Donald Trump in 2016.

	Study 1	Study 2	Study 3	Study 4	Study 5
	January 22–23	February 5–9	February 24–25	April 22–24	Oct. 6–16
Racial demographic shift	0.31** (0.11)	−0.06 (0.12)	0.11 (0.12)	−0.13 (0.13)	−0.04 (0.04)
Economic threat					0.02 (0.04)
Conservatism	0.33*** (0.06)	0.26*** (0.06)	0.17** (0.06)	0.23*** (0.07)	0.34*** (0.01)
Male (reference: female)	−0.15 (0.12)	−0.04 (0.12)	0.15 (0.12)	0.21 (0.13)	0.00 (0.04)
Age	0.01** (0.00)	0.01* (0.01)	0.01 (0.00)	0.02** (0.00)	0.00 (0.00)
Income	0.03 (0.02)	−0.04* (0.02)	0.00 (0.02)	0.00 (0.02)	−0.01 (0.01)
Education (reference: high school)
Less than high school	0.62 (0.68)	−1.49 (0.97)	0.24 (0.51)	−0.61 (0.58)	0.01 (0.07)
Some college	−0.07 (0.17)	−0.22 (0.19)	0.25 (0.20)	−0.14 (0.19)	−0.15** (0.05)
Bachelor’s degree or higher	−0.12 (0.18)	−0.28 (0.19)	−0.36 (0.20)	−0.38* (0.19)	−0.27*** (0.05)
Intercept	−2.41*** (0.41)	−1.26*** (0.37)	−1.26** (0.39)	−1.60*** (0.41)	−1.28*** (0.09)
Observations	268	267	249	239	2,053
R ²	0.16	0.11	0.13	0.13	0.33
Adjusted R²	0.14	0.09	0.10	0.10	0.33
Residual standard error	0.93	0.96	0.95	0.95	0.82
F statistic	6.36***	4.18***	4.29***	4.20***	112.05***

Note. Measure of Trump support is standardized. Racial demographic shift (0 = control, 1 = racial demographic shift) and economic threat (0 = control, 1 = economic threat) are dummy-coded separately.

p < .05. **p < .01. ***p < .001.

We observed a marginally significant effect of racial demographic shift on racial resentment in Study 1 (β = .23, p = .054), but not in Studies 2, 3, or 4. We also found marginally significant effects of racial demographic shift on perceived group status threat in Study 1 (β = .21, p = .082), Study 2 (β = .21, p = .094), and Study 3 (β = .25, p = .056), but not in Study 4 (β = −.13, p = .345). We do not observe any evidence of an effect of racial demographic shift on our measures of common fate (all ps > .182) or ethnic identification (all ps > .573). Full regression results for these outcomes are presented in Tables S1–S4 in our online supplemental material.

Interaction effects

Following similar reasoning to that of Major et al. (2018), we wished to assess whether the centrality of being White to individuals’ identities was a factor in a potential link between racial status threat and support for Donald Trump. To test whether ethnic identification moderated an effect of racial status threat, we regressed support for Donald Trump on racial demographic shift, ethnic identification, the interaction of the two, and the set of control variables described before. Results are presented in Table 4. We note that in our studies, as in Major et al. (2018), ethnic identification was measured after the experimental manipulation, although it is preferable to measure moderating variables before the manipulation (Montgomery et al., 2018). Although we see no evidence of an interaction effect in the first three studies, we do observe a significant interaction term in Study 4 (April 22–24). The coefficient for ethnic identification (β = .39, p < .001) indicates the association between that variable and support for Donald Trump among participants in the control condition. In the racial demographic shift condition, by contrast, the addition of the interaction term (β = −.35, p = .004) results in a net increase in support for Donald Trump of only 0.04 standard deviations (i.e., 0.39 minus 0.35) for each standard deviation increase in ethnic identification, net of the covariates. As a result, this model suggests that ethnic identification and support for Donald Trump are primarily linked among participants in the control condition in Study 4. We found evidence of similar interactions in Study 4 in four out of five robustness tests using different combinations of alternative measures of Trump support and measures of ethnic identification, which we present in our online supplemental material (Tables S8, S10, S11, S13, and S14). We illustrate the slope of the regression line for each condition in Figure S1 in the online supplemental material. This figure shows the steeper slope—and stronger relationship between Trump support and ethnic identification—in the control condition according to this model.

Table 4.

Interaction effect of racial demographic shift and ethnic identification on support for Donald Trump in 2016.

	Study 1	Study 2	Study 3	Study 4
	January 22–23	February 5–9	February 24–25	April 22–24
Racial demographic shift	0.29** (0.11)	−0.05 (0.12)	0.11 (0.12)	−0.11 (0.12)
Ethnic ID	0.28** (0.09)	0.21* (0.08)	0.20* (0.08)	0.39*** (0.08)
Racial Demographic Shift x Ethnic ID interaction	−0.14 (0.12)	−0.08 (0.12)	−0.13 (0.12)	−0.35** (0.12)
Conservatism	0.30*** (0.06)	0.25*** (0.06)	0.16* (0.06)	0.20** (0.06)
Male (reference: female)	−0.15 (0.11)	−0.06 (0.12)	0.16 (0.12)	0.26* (0.12)
Age	0.01** (0.00)	0.01* (0.01)	0.01 (0.00)	0.02*** (0.00)
Income	0.03 (0.02)	−0.04* (0.02)	0.00 (0.02)	0.00 (0.02)
Education (reference: high school)
Less than high school	0.72 (0.66)	−1.34 (0.96)	0.30 (0.51)	−0.60 (0.55)
Some college	−0.07 (0.17)	−0.20 (0.19)	0.25 (0.19)	−0.16 (0.18)
Bachelor’s degree or higher	−0.12 (0.17)	−0.26 (0.19)	−0.34 (0.20)	−0.37 (0.19)
Intercept	−2.25*** (0.40)	−1.13** (0.37)	−1.22** (0.39)	−1.48*** (0.39)
Observations	268	267	249	239
R ²	0.21	0.15	0.15	0.20
Adjusted R²	0.18	0.11	0.11	0.17
Residual standard error	0.91	0.94	0.94	0.91
F statistic	6.70***	4.35***	4.13***	5.83***

Note. Support for Donald Trump and ethnic identification are standardized. Racial demographic shift is dummy-coded (1 = racial demographic shift, 0 = control condition).

p < .05. **p < .01. ***p < .001.

Internal meta-analysis of Studies 1–4

Before turning to our next study, which deviates from the design and sample procedure of Studies 1–4 in a few ways, we report the results of an internal meta-analysis of these first four studies. Although Studies 1–4 were conducted at different points in time, each sample was drawn from the same prescreened panel, and the procedures and materials in each study were identical. At this stage, we assumed that sampling was the primary source of uncertainty, and a fixed-effects model was appropriate for this meta-analysis (e.g., Hedges & Vevea, 1998). We conducted this analysis using the meta library for R (Balduzzi et al., 2019; Harrer et al., 2021). This analysis yielded a nonsignificant summary effect size estimate, Hedges’s g = 0.03, p = .666. The test for heterogeneity was nonsignificant, Q = 5.64, p = .131.

Study 5

We also fielded a better-powered version of our survey experiment on a national probability sample just before the 2016 presidential election. In this study, we once again used a racial demographic shift manipulation to test whether racial status threat drove support for Donald Trump’s campaign. However, this study featured the addition of an economic threat condition. Economic anxiety was and continues to be an important factor theorized by many to drive support for Donald Trump’s candidacy. One version of this claim suggests that some Americans trusted Donald Trump to look out for their economic interests more than other candidates or the traditional party system in the United States (e.g., Frank, 2016). Other varieties of this argument focus on a relationship between economic threat and ingroup–outgroup dynamics such as scapegoating immigrants (Hirsch et al., 2021; Stephan et al., 2005) or a more general relationship between economic threat and authoritarian attitudes (Feldman & Stenner, 1997; Rickert, 1998). There has been disagreement about how to operationalize distinct threats to group (e.g., racial/ethnic) status versus economic anxiety (Morgan, 2018; Mutz, 2018a). Our manipulation of economic threat is intended to prime sociotropic economic threat (Kinder & Kiewiet, 1981), inducing a sense that the health of the American economy in general is in peril (see also Hainmueller & Hopkins, 2014).

With respect to the control and racial demographic shift conditions, Study 5 more closely resembles the design of Major et al.’s (2018) experiment. Rather than contrasting the changing racial/ethnic make-up of the population over the coming decades to more stable, short-term demographic trends (as in Studies 1–4), Study 5 contrasts racial demographic shift to a control condition concerning geographic mobility (see Table 1). Notably, the control condition in Study 5 concerns static geographic mobility, in contrast to the changing geographic mobility condition in Major et al. (2018).

Methods

Timing and participants

Study 5 was fielded via the Time-Sharing Experiments for the Social Sciences (hereafter TESS) project to a national probability sample between October 6 and 16, 2016, well after Donald Trump was declared the Republican nominee and just a few weeks before the presidential election on November 8. We were not aware of Major et al.’s (2018) research when we submitted our proposal for this study to TESS in April or as we revised the proposal over the summer. Midway through data collection, on October 7, The Washington Post released the infamous “Access Hollywood tape,” in which Donald Trump appears to describe a pattern of committing sexual assault. Although this event seemed to reduce support for his candidacy (e.g., Rhodes et al., 2020), we found no evidence in unreported analyses of this data set that this event moderated an effect of either racial demographic shift or economic threat on support for his candidacy.

Procedure

We recruited a national probability sample of 2,058 White participants through GfK, with support from TESS. Five participants were excluded due to missing data. Participants were randomly assigned to one of three conditions. In the control condition (n = 691), participants read an article about stable rates of geographic mobility in the United States. In the racial demographic shift condition (n = 694), participants received the same information as participants in the treatment conditions of Studies 1–4. Finally, in the economic threat condition (n = 668), participants read about instability in the U.S. economy, growing personal debt, increasing strain on support systems like social security, and looming increases in precarity in general. Data, our proposal, and all materials for Study 5 are available at OSF (https://osf.io/d3agv/).

Measures

Participants provided information including political ideology, age, gender, household income, and educational attainment as part of their participation in the GfK panel. Political conservatism was measured with an item like that used in Studies 1–4, “In general, do you think of yourself as . . .” (1 = extremely liberal, 7 = extremely conservative). Support for Donald Trump and Hillary Clinton were measured using the item, “To what extent do you support or oppose each of the following candidates?” (1 = strongly oppose, 7 = strongly support), with the two candidates presented in a randomized order.

Group status threat was again measured using the item, “If they increase in status, racial minorities are likely to reduce the influence of White Americans in society” (1 = strongly disagree, 7 = strongly agree).

Results

Main effects

To test whether this larger, more representative study provided evidence of a relationship between racial status threat and support for Donald Trump in 2016, we regressed Trump support on dummy variables for the racial demographic shift and economic threat conditions, leaving the control condition as the reference category. In our models, we also included the control variables used in Studies 1–4: gender, age, household income, educational attainment, and political conservatism. We used a four-category measure of education that matched our previous studies (“less than high school,” “high school,” “some college,” or “bachelor’s degree or higher”) and recoded household income so that the bins more closely matched the coding in Studies 1–4. Our substantive results and conclusions do not differ when we code household income or educational attainment differently.

We present the results of our regression model in the rightmost column of Table 3. We found no evidence of a main effect of either racial demographic shift or economic threat on Trump support.⁵ However, we did observe a significant main effect of each treatment on perceived group status threat. Relative to the control condition, agreement with the group status threat measure was 0.41 standard deviations higher in the racial demographic shift condition, net of the other covariates (β = .41, p < .001). In contrast, agreement with this measure was significantly lower in the economic threat condition, β = −.11, p = .036. Consequently, we tested whether group threat mediated an indirect effect of either of these treatments on support for Donald Trump.

Mediation analysis

First, we tested for simple mediation of an effect of racial demographic shift (dummy-coded) on support for Donald Trump by perceived group status threat. We included the same control variables and a dummy variable for the economic threat condition, and we used the PyProcessMacro (André, 2021) implementation of the PROCESS macro for the Python programming language. The 95% confidence interval for the estimate of the indirect effect through perceived group status threat does not contain zero [0.01, 0.04], β = .03.

This indirect effect is consistent with our group position theory account: racial status threat induced by the racial demographic shift treatment is associated with increased perceived group status threat, which is, in turn, associated with increased support for Donald Trump. However, we note that in addition to the lack of a main (or total) effect, the nonsignificant estimate of the direct effect has the opposite sign, β = −.08, 95% CI [−0.16, 0.12]. In other words, we found evidence of a mechanism that is consistent with our account, but this appears, in our analysis, to be a suppressor effect that warrants more cautious interpretation given the overall pattern of our findings.

Next, we tested for an indirect effect of economic threat (dummy-coded) on support for Donald Trump mediated by perceived group status threat. We used the same procedure and control variables as well as a dummy variable for the racial demographic shift condition. The 95% confidence interval for the indirect effect through perceived group status threat does not contain zero [−0.02, −0.001], β = −.01. This suggests that sociotropic economic threat was associated with decreased perceived group status threat relative to the control condition, which was, in turn, associated with decreased support for Donald Trump. We similarly found that the nonsignificant direct effect has the opposite sign, β = .02, 95% CI [−0.07, 0.11]. Given the magnitude of the coefficient (β = −.01), the lack of a main (or total) effect, and the contrasting sign of the estimate of the direct effect in the mediation analysis, we hesitate to assign too much importance to this finding.

Internal meta-analysis of Studies 1–5

Finally, we conducted an internal meta-analysis of all five studies. We note again that Study 5 differs from the previous studies in several ways: the data were collected from a different panel through a different procedure, the control condition is different (and did not prime race/ethnicity), and we measured Trump support alone, rather than measuring and averaging (highly correlated) support and favorability measures, as we did for Studies 1–4. Given these overt sources of heterogeneity, we present results from both fixed and random effects models. In these models, we compare only the racial demographic shift condition to the control condition. According to both models, the summary effect is negative and nonsignificant. The fixed effects model (Hedges’s g = −0.04, p = .331) produces a larger estimate than the random effects model (Hedges’s g = −0.01, p = .851). The test of heterogeneity is also nonsignificant, Q = 7.60, p = .107. Figure 2 presents the results in greater detail, showing the effect sizes and confidence intervals for each study, and summary effect sizes for both the fixed effects model and the random effects model.

Figure 2.

Results of the internal meta-analysis.

General Discussion

Many potential explanations for support for Donald Trump as a political and cultural figure have been proposed. These range from broad arguments about factors like economic interests (Frank, 2016), authoritarianism (Smith & Hanley, 2018), and Christian nationalism (Whitehead et al., 2018) to narrower arguments about issues like threatened masculinity (Carian & Sobotka, 2018) and the use of language that may evoke a sense of crisis (Homolar & Scholz, 2019; Schrock et al., 2017). We have focused on the role of concerns about group status among Whites. Major et al. (2018) found that making salient the declining White majority in the United States increased support for Donald Trump’s presidential candidacy among White participants with high levels of ethnic identification. In the same period in 2016, we conducted five very similar experiments, also manipulating racial status threat via a report on the declining proportion of White Americans (racial demographic shift), also measuring support for Trump, and also measuring ethnic identification in the first four studies. We found a significant main effect of racial demographic shift on Whites’ Trump support only in our first experiment, conducted in January 2016. We did find evidence for an interaction of racial status threat and ethnic identification on Whites’ Trump support in Study 4, but the character of this interaction is difficult to interpret and very different from the interaction reported by Major and colleagues.

Here, we consider several possible interpretations that may account for our findings and those of Major and colleagues. The pattern of results found in these two papers could be driven by none, one, or multiple of the dynamics discussed next, or by other dynamics entirely.

1. Racial status threat was not causally linked with Whites’ support for Donald Trump. This is perhaps the most straightforward interpretation, given that in the studies as a whole, we did not find a reliable main effect of racial status threat, or a reliable interaction with ethnic identity, on Trump support. This interpretation may be surprising given prior research linking racial status threat with conservatism, racial resentment, and Tea Party support, as well as Trump’s positions and rhetoric on topics like immigration, the support for which is often attributed to racial status concerns. However, it is possible that the link between racial status threats and Trump support was not causal.

2. Views on Trump’s candidacy became less treatable over the course of his campaign. Here we found a main effect of racial status threat in the earliest study. Further, Major et al. (2018) found an effect among high-identifying Whites while the campaign was still in the primaries. A recent meta-analysis of field experiments of persuasion interventions embedded in campaigns found that effects of these interventions were only detectable early in the campaigns, not later when views on candidates had become crystallized and the information environment had become denser (Kalla & Broockman, 2018). In these later stages, it is harder to change impressions of candidates. In the context of the 2016 presidential campaign, which was extensively covered in the media and received considerable attention on various social media platforms, it is plausible that the same, relatively minimal experimental treatments could have shifted individuals’ views on candidates very early in the cycle, but not later.

3. Psychological bases of support for Trump shifted over time. It might be the case that feelings of racial status threat drove Whites’ support for Trump early in the campaign, but less so or not at all later. This shift over time could be driven by different possible dynamics. First, it could be that Trump moved away from the racially incendiary rhetoric that characterized the beginning of his primary campaign, making him a less attractive candidate for racially threatened Whites. Alternatively, it could be that, as Trump pulled ahead in the primaries, his base broadened. As his base broadened, so too did the psychological bases of support for Trump, with many supporting him because he was the most viable Republican in the race.

Although our studies were not designed to test the existence of trends in the psychological bases of support for Trump, they do offer the ability to take a cautious, exploratory look at certain trends. Table 5 presents the Pearson correlation coefficients between Trump support and several key variables in the control condition in each of our five studies.⁶ Notably, we found that the self-reported group status threat item correlated more highly with Trump support in Study 1 than in Studies 2–5, consistent with our experimental findings. This suggests that racial status threat was a factor early in Trump’s campaign but became less significant over time. Beyond this, we see little clear evidence of trends in the relationships between Trump support and these variables over time. However, we do see declines from Study 1 (when we found the expected effect) to Study 3 in the correlations between Trump support and political conservatism and between Trump support and racial resentment.⁷ Correlations between Trump support and ethnic identification, Trump support and racial resentment, and Trump support and identification with the Republican Party were highest in Study 4, when we found evidence of an interaction effect.

Table 5.

Correlations between support for Donald Trump and other key measures in control condition.

	Study 1	Study 2	Study 3	Study 4	Study 5
	January 22–23	February 5–9	February 24–25	April 22–24	October 6–16
Premanipulation
Political conservatism	0.38***	0.21*	0.18	0.26**	0.54***
Party identification^†	-	-	-	-	−0.65***
Age	0.23**	0.19*	0.07	0.21*	0.09*
Income	0.00	−0.01	−0.05	−0.01	−0.09*
Postmanipulation
Perceived group status threat	0.24**	0.10	0.15	0.15	0.11**
Perceived racial common fate	0.13	0.17	0.29***	0.24**	-
Ethnic identification	0.28***	0.20*	0.24**	0.39***	-
Racial resentment	0.47***	0.37***	0.36***	0.53***	-
Identification with Republican Party	0.42***	0.37***	0.50***	0.57***	-
Identification with Democratic Party	−0.05	−0.07	−0.18*	0.01	-

Note. ^†Party identification was measured only in Study 5 (1 = strong Republican, 7 = strong Democrat).

p < .05. **p < .01. ***p < .001.

4. Confounded racial status threat manipulations. Our manipulation in Studies 1–4 was designed to manipulate racial status threat while holding constant the salience of race/ethnicity and racial/ethnic differences. In these studies, we used a control condition that also reports on racial/ethnic demographic trends over time in the United States but does so in a way that highlights the declining White majority far less than our treatment condition, which projects trends out to 2060, by which time Whites are expected to become a numeric minority (see Wetts & Willer, 2018). However, Major et al. (2018) used a control condition (trends in geographic mobility rates) that does not refer to race or racial differences, meaning that their manipulation varies both the salience of race and racial differences in the United States, as well as racial status threat.

The differences between our and Major et al.’s control conditions create alternative explanations for our findings. On the one hand, it could be that Major et al.’s (2018) findings were driven not by racial status threat but instead by making salient race/ethnicity or racial/ethnic differences, with this salience increasing high-identifying Whites’ support for Trump for reasons besides racial status threat. Another possibility is that by making race salient, our control conditions in Studies 1–4 primed a sense of racial status threat in a way similar to the racial demographic shift treatment. Consistent with this explanation, when we switched control conditions in Study 5 to a control condition more like Major et al.’s, we found a much larger effect on our perceived group status threat measure. On the other hand, we still did not find an effect of racial status threat on Trump support in Study 5. Additionally, the manipulation we used in Studies 1–4 showed racial status threat effects in other research (Abascal, 2020; Wetts & Willer, 2018; Willer et al., 2016) and in our Study 1.

5. Other differences between designs. While our studies were strikingly similar to that of Major and colleagues’, there were nonetheless differences. For example, Major and colleagues’ dependent measures included more intervals than ours. They used a feeling-thermometer-style item with 100 intervals and a 7-point Likert scale assessing likelihood of voting for Trump. Scales with many intervals can be better for detecting small treatment effects. That said, our composite measure of Trump support (averaging favorability and support) in Studies 1–4 had eight intervals. Further, we did find a robust effect in Study 1, and Study 5 should have been sufficiently powered to detect small effects even with its coarser dependent measure (see, for example, the highly significant effect on perceived group status threat we found in Study 5).

We believe that all these potential explanations are plausible, though we have a limited ability to adjudicate which of these, if any, accounts for the divergent results. Of the previous interpretations, we would highlight 1, 2, and 3 as particularly plausible. Interpretation 1 is attractive because it is the simplest interpretation of an unreliable effect. Interpretation 2 is attractive because it aligns with common sense and the results of a major meta-analysis on persuasion efforts in campaigns (Kalla & Broockman, 2018). Interpretation 3 is attractive because it is extremely likely that partisanship becomes a stronger basis of candidate support as a candidate becomes the presumptive, and eventually the actual, nominee of a party, and this necessarily crowds out other bases of support.

Beyond these interpretations, myriad other standard interpretations for unreliable results are possible, such as small differences in study populations and specific events happening in the world at the time of the studies. While we cannot be certain of what accounts for the differences between our results and those of Major and colleagues, we believe the scientific record is most helpful if it presents all these findings so that future researchers may be fully informed. We hope that laying out our findings and reviewing differences between studies in this area will contribute to the progress of this literature.

Future work may consider new variations on racial demographic shift manipulations (e.g., Bai & Federico, 2021) and also uncover when different control conditions are appropriate. In addition, there may be opportunities to combine the various proposed group-based threats in new ways. For example, Craig and Richeson (2014b, Study 3) demonstrated that assuaging material concerns may mitigate general group status threat dynamics, and future work can probe this further to resolve fundamental questions regarding the relationships between status and economic concerns at the group level (cf. Quillian, 1995). Researchers may also more systematically assess the kinds of policy positions and candidates that benefit from racial status threat dynamics. We hope that future work builds on our two papers and develops a deeper understanding of what reactions racial status threats cause among White Americans, what attitudes and behaviors those reactions are likely to trigger, when these effects obtain, when they do not, and why.

Supplemental Material

sj-docx-1-gpi-10.1177_13684302211048893 – Supplemental material for The effects of racial status threat on White Americans’ support for Donald Trump: Results of five experimental tests

Supplemental material, sj-docx-1-gpi-10.1177_13684302211048893 for The effects of racial status threat on White Americans’ support for Donald Trump: Results of five experimental tests by Sheridan Stewart and Robb Willer in Group Processes & Intergroup Relations

Footnotes

Acknowledgements

We thank James Druckman and Chrystal Redekopp for their contributions to this project.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: We thank the National Science Foundation and the Time-Sharing Experiments for the Social Sciences project for supporting this research.

ORCID iD

Sheridan Stewart

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Abascal

(2015). Us and them: Black–White relations in the wake of Hispanic population growth. American Sociological Review, 80, 789–813. https://doi.org/10.1177/0003122415587313

Abascal

(2020). Contraction as a response to group threat: Demographic decline and Whites’ classification of people who are ambiguously White. American Sociological Review, 85, 298–322. https://doi.org/10.1177/0003122420905127

Alba

(2018). What majority-minority society? A critical analysis of the Census Bureau’s projections of America’s demographic future. Socius: Sociological Research for a Dynamic World, 4, 1–10. https://doi.org/10.1177/2378023118796932

André

(2021). PyProcessMacro: A Python implementation of Andrew F. Hayes’ “Process” macro [Computer software]. https://github.com/QuentinAndre/pyprocessmacro

Bai

Federico

C. M.

(2020). Collective existential threat mediates White population decline’s effect on defensive reactions. Group Processes & Intergroup Relations, 23, 361–377. https://doi.org/10.1177/1368430219839763

Bai

Federico

C. M.

(2021). White and minority demographic shifts, intergroup threat, and right-wing extremism. Journal of Experimental Social Psychology, 94, Article 104114. https://doi.org/10.1016/j.jesp.2021.104114

Balduzzi

Rücker

Schwarzer

(2019). How to perform a meta-analysis with R: A practical tutorial. Evidence-Based Mental Health, 22, 153–160. https://doi.org/10.1136/ebmental-2019-300117

Benjamin

Block

Jr. Clemons

Lair

Wamble

(2020). Set in stone? Predicting Confederate monument removal. PS: Political Science & Politics, 53, 237–242. https://doi.org/10.1017/S1049096519002026

Blumer

(1958). Race prejudice as a sense of group position. The Pacific Sociological Review, 1, 3–7. https://doi.org/10.2307/1388607

10.

Bobo

(1983). Whites’ opposition to busing: Symbolic racism or realistic group conflict? Journal of Personality and Social Psychology, 45, 1196–1210. https://doi.org/10.1037/0022-3514.45.6.1196

11.

Bobo

L. D.

(1999). Prejudice as group position: Microfoundations of a sociological approach to racism and race relations. Journal of Social Issues, 55, 445–472. https://doi.org/10.1111/0022-4537.00127

12.

Bobo

Hutchings

V. L.

(1996). Perceptions of racial group competition: Extending Blumer’s theory of group position to a multiracial social context. American Sociological Review, 61, 951–972. https://doi.org/10.2307/2096302

13.

Bonikowski

(2017). Ethno-nationalist populism and the mobilization of collective resentment. British Journal of Sociology, 68, S181–S213. https://doi.org/10.1111/1468-4446.12325

14.

Carian

Sobotka

(2018). Playing the Trump card: Masculinity threat and the U.S. 2016 presidential election. Socius: Sociological Research for a Dynamic World, 4, 1–6. https://doi.org/10.1177/2378023117740699

15.

Craig

M. A.

Richeson

J. A.

(2012). Coalition or derogation? How perceived discrimination influences intraminority intergroup relations. Journal of Personality and Social Psychology, 102, 759–777. https://doi.org/10.1177/0002716218766269

16.

Craig

M. A.

Richeson

J. A.

(2014a). More diverse yet less tolerant? How the increasingly diverse racial landscape affects White Americans’ racial attitudes. Personality and Social Psychology Bulletin, 40, 750–761. https://doi.org/10.1177/0146167214524993

17.

Craig

M. A.

Richeson

J. A.

(2014b). On the precipice of a “majority-minority” America: Perceived status threat from the racial demographic shift affects White Americans’ political ideology. Psychological Science, 25, 1189–1197. https://doi.org/10.1177/0956797615578857

18.

Craig

M. A.

Richeson

J. A.

(2017). Information about the U.S. racial demographic shift triggers concerns about anti-White discrimination among the prospective White “minority.” PLoS ONE, 12, Article e0185389. https://doi.org/10.1371/journal.pone.0185389

19.

Craig

M. A.

Rucker

J. M.

Richeson

J. A.

(2018). Racial and political dynamics of an approaching “majority-minority” United States. The Annals of the American Academy of Political and Social Science, 677, 204–214. https://doi.org/10.1177/0002716218766269

20.

Danbold

Huo

Y. J.

(2015). No longer “all-American”? Whites’ defensive reactions to their numerical decline. Social Psychological and Personality Science, 6, 210–218. https://doi.org/10.1177/1948550614546355

21.

DiGrazia

(2017). The social determinants of conspiratorial ideation. Socius: Sociological Research for a Dynamic World, 3, 1–9. https://doi.org/10.1177/2378023116689791

22.

Feldman

Stenner

(1997). Perceived threat and authoritarianism. Political Psychology, 18, 741–770. https://doi.org/10.1111/0162-895x.00077

23.

Frank

(2016, March 8). Millions of ordinary Americans support Donald Trump: Here’s why. The Guardian. https://www.theguardian.com/commentisfree/2016/mar/07/donald-trump-why-americans-support

24.

Fritsche

Jonas

Ablasser

Beyer

Kuban

Manger

Schultz

(2013). The power of we: Evidence for group-based control. Journal of Experimental Social Psychology, 49, 19–32. https://doi.org/10.1016/j.jesp.2012.07.014

25.

Hainmueller

Hopkins

D. J.

(2014). Public attitudes toward immigration. Annual Review of Political Science, 17, 225–249. https://doi.org/10.1146/annurev-polisci-102512-194818

26.

Harrer

Cuijpers

Furukawa

T. A.

Ebert

D. D.

(2021). Doing meta-analysis with R: A hands-on guide. Chapman & Hall/CRC Press.

27.

Hedges

L. V.

Vevea

J. L.

(1998). Fixed- and random-effects models in meta-analysis. Psychological Methods, 3, 486–504. https://doi.org/10.1037/1082-989X.3.4.486

28.

Henry

P. J.

Sears

D. O.

(2002). The Symbolic Racism 2000 Scale. Political Psychology, 23, 253–283. https://doi.org/10.1111/0162-895X.00281

29.

Hirsch

Veit

Fritsche

(2021). Blaming immigrants to enhance control: Exploring the control-bolstering functions of causal attribution, in-group identification, and hierarchy enhancement. Journal of Theoretical Social Psychology, 5, 114–131. https://doi.org/10.1002/jts5.73

30.

Hochschild

A. R.

(2016). Strangers in their own land: Anger and mourning on the American right. The New Press.

31.

Homolar

Scholz

(2019). The power of Trump-speak: Populist crisis narratives and ontological security. Cambridge Review of International Affairs, 32, 344–364. https://doi.org/10.1080/09557571.2019.1575796

32.

Kalla

J. L.

Broockman

(2018). The minimal persuasive effects of campaign contact in general elections: Evidence from 49 field experiments. American Political Science Review, 112, 148–166. https://doi.org/10.1017/S0003055417000363

33.

Kinder

D. R.

Kiewiet

D. R.

(1981). Sociotropic politics: The American case. British Journal of Political Science, 11, 129–161. https://doi.org/10.1017/S0007123400002544

34.

Lewicka

(2011). Place attachment: How far have we come in the last 40 years? Journal of Environmental Psychology, 31, 207–230. https://doi.org/10.1016/j.jenvp.2010.10.001

35.

Lewicka

Rowiński

Iwańczak

Bałaj

Kula

A. M.

Oleksy

Prusik

Toruńczyk-Ruiz

Wnuk

(2019). On the essentialism of places: Between conservative and progressive meanings. Journal of Environmental Psychology, 65, Article 101318. https://doi.org/10.1016/j.jenvp.2019.101318

36.

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. https://doi.org/10.1177/0146167207303016

37.

Lüders

Mühlberger

Jonas

(2020). Motivational and affective drivers of right-wing populism support: Insights from an Austrian presidential election. Social Psychological Bulletin, 15, Article e2875. https://doi.org/10.32872/spb.2875

38.

Luhtanen

Crocker

(1992). A collective self-esteem scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18, 302–318. https://doi.org/10.1177/0146167292183006

39.

Major

Blodorn

Blascovich

G. M.

(2018). The threat of increasing diversity: Why many White Americans support Trump in the 2016 presidential election. Group Processes & Intergroup Relations, 21, 931–940. https://doi.org/10.1177/1368430216677304

40.

Mason

Wronski

Kane

J. V.

(2021). Activating animus: The uniquely social roots of Trump support. American Political Science Review. https://doi.org/10.1017/S0003055421000563

41.

Montgomery

J. M.

Nyhan

Torres

(2018). How conditioning on posttreatment variables can ruin your experiment and what to do about it. American Journal of Political Science, 62, 760–775. https://doi.org/10.1111/ajps.12357

42.

Morgan

S. L.

(2018). Status threat, material interests, and the 2016 presidential vote. Socius: Sociological Research for a Dynamic World, 4, 1–7. https://doi.org/10.1177/2378023118788217

43.

Mutz

D. C.

(2018a). Response to Morgan: On the role of status threat and material interests in the 2016 election. Socius: Sociological Research for a Dynamic World, 4, 1–11. https://doi.org/10.1177/2378023118808619

44.

Mutz

D. C.

(2018b). Status threat, not economic hardship, explains the 2016 presidential vote. Proceedings of the National Academy of Sciences of the USA, 115, E4330–E4339. https://doi.org/10.1073/pnas.1718155115

45.

Osborn

H. J.

Sosa

Rios

(2020). Perceiving demographic diversity as a threat: Divergent effects of multiculturalism and polyculturalism. Group Processes & Intergroup Relations, 23, 1014–1031. https://doi.org/10.1177/1368430219880606

46.

Outten

H. R.

Schmitt

M. T.

Miller

Garcia

A. L.

(2012). Feeling threatened about the future: Whites’ emotional reactions to anticipated ethnic demographic changes. Personality and Social Psychology Bulletin, 38, 14–25. https://doi.org/10.1177/0146167211418531

47.

Outten

H. R.

Lee

Costa-Lopes

Schmitt

M. T.

Vala

(2018). Majority group members’ negative reactions to future demographic shifts depend on the perceived legitimacy of their status: Findings from the United States and Portugal. Frontiers in Psychology, 9, Article 79. https://doi.org/10.3389/fpsyg.2018.00079

48.

Perkins

K. M.

Dils

A. T.

Flusberg

S. J.

(2022). The perceived threat of demographic shifts depends on how you think the economy works. Group Processes & Intergroup Relations, 25, 227–246. https://doi.org/10.1177/1368430220951621

49.

Quillian

(1995). Prejudice as a response to perceived group threat: Population composition and anti-immigrant and racial prejudice in Europe. American Sociological Review, 60, 586–611. https://doi.org/10.2307/2096296

50.

Rickert

E. J.

(1998). Authoritarianism and economic threat: Implications for political behavior. Political Psychology, 19, 707–720. https://doi.org/10.1111/0162-895X.00128

51.

Rhodes

J. H.

Sharrow

E. A.

Greenlee

J. S.

Nteta

T. M.

(2020). Just locker room talk? Explicit sexism and the impact of the Access Hollywood tape on electoral support for Donald Trump in 2016. Political Communication, 37, 741–767. https://doi.org/10.1080/10584609.2020.1753867

52.

Sawyer

P. S.

(2020). Populist conspiracy theories and candidate preference in the U.S. Acta Politologica, 3, 1–26. https://doi.org/10.14712/1803-8220/12_2020

53.

Sawyer

P. S.

(2021). Conspiracism in populist radical right candidates: Rallying the base or mainstreaming the fringe? International Journal of Politics, Culture, and Society. https://doi.org/10.1007/s10767-021-09398-4

54.

Schrock

Dowd-Arrow

Erichsen

Gentile

Dignam

(2017). The emotional politics of making America great again: Trump’s working class appeals. Journal of Working-Class Studies, 2, 5–22. https://doi.org/10.13001/jwcs.v2i1.6039

55.

Sides

Tesler

Vavreck

(2018). Identity crisis: The 2016 presidential campaign and the battle for the meaning of America. Princeton University Press.

56.

Smith

D. N.

Hanley

(2018). The anger games: Who voted for Donald Trump in the 2016 election, and why? Critical Sociology, 44, 195–212. https://doi.org/10.1177/0896920517740615

57.

Stephan

W. G.

Renfro

C. L.

Esses

V. M.

Stephan

W. C.

Martin

(2005). The effect of feeling threatened on attitudes toward immigrants. International Journal of Intercultural Relations, 29, 1–19. https://doi.org/10.1016/j.ijintrel.2005.04.011

58.

Thompson

(2016, May 13). Donald Trump and the twilight of White America. The Atlantic. https://www.theatlantic.com/politics/archive/2016/05/donald-trump-and-the-twilight-of-white-america/482655/

59.

Van der Linden

Panagopoulos

Azevedo

Jost

J. T

. (2021). The paranoid style in American politics revisited: An ideological asymmetry in conspiratorial thinking. Political Psychology, 42, 23–51. https://doi.org/10.1111/pops.12681

60.

Wetts

Willer

(2018). Privilege on the precipice: Perceived racial status threats lead White Americans to oppose welfare programs. Social Forces, 97, 793–822. https://doi.org/10.1093/sf/soy046

61.

Whitehead

A. L.

Perry

S. L.

Baker

J. O.

(2018). Make America Christian again: Christian nationalism and voting for Donald Trump in the 2016 presidential election. Sociology of Religion, 79, 147–171. https://doi.org/10.1093/socrel/srx070

62.

Willer

Feinberg

Wetts

(2016). Threats to racial status promote Tea Party support among White Americans (SSRN working paper). Stanford University. https://doi.org/10.2139/ssrn.2770186

63.

Wnuk

Oleksy

Toruńczyk-Ruiz

Lewicka

(2021). The way we perceive a place implies who can live there: Essentialisation of place and attitudes towards diversity. Journal of Environmental Psychology, 75, Article 101600. https://doi.org/10.1016/j.jenvp.2021.101600

64.

Zou

L. X.

Cheryan

(2017). Two axes of subordination: A new model of racial position. Journal of Personality and Social Psychology, 112, 696–717. https://doi.org/10.1037/pspa0000080

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

0.14 MB