Varieties of White Working-Class Identity

Abstract

The present work demonstrates that, contrary to popular political narratives, working-class White Americans are far from monolithic in their class identities, social attitudes, and political preferences. Latent profile analysis is used to distinguish three types of identity in a nationally representative sample of working-class Whites: Working Class Patriots, who valorize responsibility, embrace national identity, and disparage the poor; Class Conflict Aware, who regard social class as a structural phenomenon and ascribe elitist attitudes to higher classes; and Working Class Connected, who embrace working-class identity, sympathize with the poor, and feel disrespected because of the work they do. This identity typology appears unique to working-class Whites and is associated with distinct patterns of attitudes regarding immigration, race, and politics, such that Class Conflict Aware and Working Class Connected Whites are considerably more progressive than are Working Class Patriots. Implications for electoral politics and race relations are discussed.

Keywords

whites social class identity race immigration politics latent profile analysis

The White working class is often credited with helping elevate Donald Trump to the presidency and elect scores of Republicans to state and local offices (Morgan & Lee, 2018; Teixeira & Rogers, 2000). Yet despite working-class Whites’ general rightward tilt, the electoral preferences and social values of this group resist easy categorization. Indeed, tens of millions of working-class White voters opposed Trump in 2016 (Tyson & Maniam, 2016), in large part because his xenophobic and exclusionary rhetoric ran counter to their more progressive social views (Smith & Hanley, 2018). Contrary to popular political narratives (e.g., Packer, 2012), the White working class is a not a monolithic political force—raising the question of why Whites who occupy a similar socioeconomic location display such a diversity of social and political views (McDermott et al., 2019).

In the present article, we argue that working-class Whites’ social and political attitudes are inextricably tied to their class identities—that is, their beliefs about their own location in the socioeconomic hierarchy, about adjacent classes, and about social values worth upholding (McDermott et al., 2019). We begin by leveraging nationally representative survey data to document the number, nature, and relative prevalence of different identity types found among working-class White adults. We then assess the degree to which the structure of class identity among working-class Whites differs from that found in other groups (i.e., non-working-class Whites and working-class Black people and Latino/as). Finally, we examine associations between White working-class identity types and attitudes toward immigrants, racial minorities, and politics. In accomplishing these goals, we seek to illuminate sources of attitudinal diversity among the White working class—a group whose views are socially and politically consequential, and yet poorly understood.

Our Conceptualization of Identity

Within psychology, social identity theory (Tajfel & Turner, 2004) and self-categorization theory (Turner et al., 1987) are the dominant frameworks for understanding people’s subjective connection to social groups (e.g., races, genders, and classes). These theoretical traditions paint self-definition in terms of a social category—one’s explicit or tacit acknowledgement of membership in the category—as the sine qua non of social identification (Ashmore et al., 2004; Deaux, 2015). Our treatment of identity diverges from these perspectives in important ways.

By “White working-class identities,” we mean the class-relevant psychological characteristics of Whites who, according to externally observable criteria, belong to the (American) working class. On this definition, a White person need not identify as working class to qualify as a member of the category—or even as having a working-class identity. This is because, in our analysis, “working class” is a material social location to which the individual’s self-concept may or may not correspond. Class-relevant characteristics, in turn, include much more than self-definition in terms of a particular social-class category (although self-definition is one of many potentially important features of one’s class identity). Rather, we take our cue from sociological approaches to identity, as well as multidimensional identity theories within social psychology, which count a wide range of characteristics—including traits, values, roles, ideologies, evaluations, and narratives that go with membership in a social category—as components of one’s identity (Burke, 2020; Leach et al., 2008; Luhtanen & Crocker, 1992; Sellers et al., 1998).

In keeping with this broad conceptualization of identity, we map White working-class identities in terms of several components distinguished in previous ethnographic work (McDermott et al., 2019). These components include feelings of regard in one’s job, beliefs about class conflict, psychological distance from the most socioeconomically disadvantaged (i.e., the poor), a de-emphasis on class identity in favor of national identity, and social values relevant to economic well-being. While these do not exhaust the possible facets of identity among the White working class, we believe they are critical to defining individuals’ sense of their location in the socioeconomic hierarchy.

Varieties of White Working-Class Identity

With important exceptions (e.g., Cramer, 2016; Gest, 2016), scholars have paid insufficient attention to systematic sources of heterogeneity in White working-class politics. Nevertheless, recent work on the link between education and politics hints at the existence of distinct sociopolitical identities within the White working class.

In the United States and other western countries, the attainment (versus lack) of a 4-year college degree has become increasingly predictive of social and political attitudes (Kriesi, 2010; Pew Research Center, 2016; Spruyt et al., 2016; Stubager, 2009, 2013), especially among White people (Pew Research Center, 2018; Zingher, 2022). Compared to the college-educated, those without college degrees tend to endorse more conservative social and political views, exhibiting greater preference for social hierarchies, traditional values, and right-wing political parties (Stubager, 2013). While some scholars take such trends to mean that education is supplanting social class (defined in terms of income, wealth, or occupational prestige) as a shaper of political attitudes (e.g., Clark & Lipset, 1991; Houtman et al., 2017), we employ a broader conception of class that includes education and the cultural, social, and symbolic resources it confers (Bourdieu, 1986; Stephens et al., 2007). From this perspective, the rise of an educational cleavage among the electorate implies that multiple dimensions of social class influence Whites’ political attitudes in complex—and potentially countervailing—ways.

The interplay between educational and non-educational dimensions of social class suggests the emergence of competing cleavages among the White working class. In the United States, education and income—though positively correlated—have independent and opposite implications for White people’s sociopolitical views, with low levels of education predicting right-wing social and electoral preferences and low income predicting left-leaning and Democratic preferences (Bartels, 2006; Zingher, 2022).¹ Consequently, the White working class may possess contradictory political affinities, their relatively low levels of wealth and income rendering liberal policies appealing and their lack of a college education rendering conservative policies attractive. We posit that the contravening forces of education and economic status result in the formation of distinct subtypes within the White working class. Specifically, while some members of this group resolve the political tension between educational attainment and economic status by retaining the more liberal values associated with lower income strata, others adopt the more conservative values increasingly common among Whites with lower educational attainment.

A Tripartite Typology of White Working-Class Identity

To directly assess the existence and nature of subtypes among the White working class, McDermott et al. (2019) conducted semi-structured interviews of non-college-educated Whites in Missouri, Kentucky, and Indiana. Rather than focusing solely on subjective class identification, or Whites’ sense of attachment to the working class (Jackman & Jackman, 1983), the interviews assessed a broader range of identity components—including moral values and attitudes toward adjacent social classes (Lamont, 2000). Analyses revealed three types of working-class identity, each exhibiting a different pattern of attitudes toward salient outgroups.

The largest group of respondents uncovered by McDermott et al. (2019) were termed Working Class Patriots. Patriots tend to embrace American identity more than class identity, emphasize the values of responsibility and hard work, derogate the poor as lazy and undeserving, express admiration for the rich, and report feeling respected because of the work they do. In turn, Patriot identity was associated with strongly negative views of immigrants and racial minorities and positive attitudes toward Donald Trump. This group thus closely resembles the proud “working men” in Michèle Lamont’s (2000) ethnography of American workers and matches popular characterizations of the White working class as relatively xenophobic and politically reactionary (Frank, 2005; Packer, 2012). Working Class Patriots also exemplify the more conservative and authoritarian values apparent in studies of non-college-educated populations in the United States and elsewhere (see, e.g., Stubager, 2013).

McDermott et al.’s (2019) interviews revealed two additional types of identity that complicate prevailing narratives concerning the White working class. Relative to Patriots, Working Class Connected and Class Conflict Aware Whites feel an attachment to their social class, de-emphasize responsibility in favor of other values (e.g., generosity and compassion), display respect and sympathy for the poor, and report feeling undervalued and disrespected because of their jobs. Class Connected and Conflict Aware Whites, however, express very different conceptions of the socioeconomic hierarchy: Conflict Aware Whites regard class as a persistent structural phenomenon that enables the rich to exploit the poor, whereas this conception is absent among the Class Connected group. Compared to Patriots, Conflict Aware—and, to a slightly lesser extent, Class Connected—Whites hold positive opinions of immigrants and racial minorities and more progressive political views. To varying degrees, then, Working Class Connected and Class Conflict Aware Whites exhibit left-leaning political tendencies in spite of their lack of college education.

The Present Research

While McDermott et al.’s (2019) qualitative work illuminates the variegated nature of White working-class identity, there are limits to what can be gleaned from those data. In particular, the small number of interview sites (St. Joseph, Missouri; Frankfort, Kentucky; and Indianapolis, Indiana) provides scant information about the prevalence of identity types in the broader population—and thus limited insight into their implications for national politics. In the present research, we search for the three types of White working-class identity in a nationally representative survey of working-class White Americans. In so doing, we seek to bolster our confidence in the types, better understand their distribution in the U.S. population, assess their uniqueness to the White working class, and more systematically link them to social and political outcomes (Small, 2011).

Method

Operationalizing the Working Class

Many operational definitions of working-class status have been proposed (Diemer et al., 2013). Researchers who take a gradational approach to the measurement of class (e.g. Kraus et al., 2013) often favor definitions rooted in income, occupational prestige, educational attainment, subjective social rank, or some combination of these indices. Rather than producing a clearly bounded “working class,” this strategy instead yields a continuous measure of socioeconomic status (SES). Researchers who press for a categorical definition of class—usually because they consider class groupings to be political, social, or cultural entities (Bourdieu, 1986; Dietze & Knowles, 2016; Kraus et al., 2012; Wright, 2015)—have drawn upon various objective and subjective measures.

Occupation is commonly used to distinguish between the working class and other class positions (Weeden & Grusky, 2005). Blue-collar labor has a long history of association with the working class, and “pink collar” service workers have more recently been subsumed into the working class as well (Kefalas, 2003). A focus on occupation captures the Marxist emphasis on class as rooted in a group’s relationship to the means of production. Pragmatic difficulties stem from the fact that numerous occupations—such as police officer, nurse, and computer technician—span the boundaries of working and middle class in terms of required skills and community perceptions (Hout et al., 1995).

Income is a more problematic marker of class categories, as the boundaries around groups are arbitrary and capture a range of occupational and educational levels. For instance, Williams (2017) considered a vast middle-income range to be working class—a definition that would include some college professors but exclude many skilled tradespeople. In addition, variation in incomes across regions of the country adds complexity to an income-based definition of the working class. Income thus has limited utility as the sole index of social class.

Education has emerged as a commonly used definition of working class. Those without 4-year college degrees are considered members of this group, while the college-educated are considered at least middle class (Stephens et al., 2007). Of potential class indicators, we believe that education aligns most closely with our broad conceptualization of social class, as education is strongly correlated with income, occupation, and cultural capital (Bourdieu, 1986; Diemer et al., 2013). An important disadvantage of education as a measure of class is the heterogeneity in college degrees (e.g., graduation from either Harvard or the University of Phoenix would exclude a person from the working class), thus risking the creation of an overly inclusive middle-class category. However, as our research deals primarily with the attitudes of the (White) working class, we chose education—specifically, the lack of a 4-year college degree—as our criterion for inclusion in this category. This definition of working class also matches that used in previous related work (McDermott et al., 2019).

Survey Samples

Survey company Bovitz, Inc. was contracted to recruit a nationally representative sample of working-class White Americans (i.e., Whites lacking a 4-year college degree). (See the Supplemental Material for a detailed description of Bovitz’s sampling methodology.) The final usable sample consisted of 2,044 respondents, of whom 1,067 (52.2%) were men and 977 (47.8%) were women. Ages ranged from 18 to 95 (M = 49.3, SD = 14.5); 68.5% of the sample had at least some post-high-school education (e.g., an associate degree or vocational training); and respondents’ median income bracket was $35,000–49,999. Respondents were geographically distributed across 50 U.S. states and 672 counties.

For purposes of gauging the extent to which the structure of White working-class identity generalizes to other groups, we also recruited three additional representative samples: 571 non-working-class (i.e., college-educated) Whites, 520 working-class Black respondents, and 481 working-class Latino/as.

Survey Items

Subjective class category

Respondents’ preferred social-class label was ascertained using the following item:

If you were asked to use one of the following six names for your social class, which would you say you belong to: the poor, the lower class, the working class, the middle class, the upper-middle class, or the upper class?

Respondents could select one label from the list.

Identity-type indicators

The survey included a set of items designed to gauge respondents’ working-class identity type. These type indicators tapped themes that distinguish between the three types discovered in previous qualitative work (McDermott et al., 2019). Table 1 displays the rationale for the choice of each item. To facilitate comparisons between the indicator means associated with our identity types, we rescaled the ordinal indicators—distancing from the poor and belief in class divisions—to range from 0 (reflecting the minimum observed value) to 1 (reflecting the maximum observed value).

Table 1.

Identity-type indicators and rationales based on McDermott et al. (2019).

Indicator	Rationale for inclusion	Prediction
American identification	Working Class Patriots (WCP) embrace American identity	WCP high
Value of responsibility	WCP emphasize responsibility	WCP high
Distancing from poor	WCP disparage poor as lazy	WCP high
Class divisions	Class Conflict Aware (CCA) see class in structural terms	CCA high
Looked down on	Working Class Connected (WCC) feel job disrespect	WCC high

American identification

To measure the relative emphasis respondents place on American identity, they were asked: “Which identity is more important to you: [class category] or American?” (1 = American, 0 = [preferred class category]), where [preferred class category] is the subjective class category previously identified by the respondent.

Value of responsibility

To assess their degree of commitment to different social values, respondents rated three values, “generosity,” “responsibility,” and “following the rules,” on a 5-point scale (1 = not at all important, 2 = slightly important, 3 = moderately important, 4 = very important, and 5 = extremely important). The resulting value of responsibility indicator was assigned a value of 1 if responsibility was rated as more important than generosity and a value of 0 if it was not.²

Distancing from the poor

To measure psychological distance from the poor, respondents were asked “how close you feel” to several social groups on a 7-point scale (1 = not close at all, 4 = neutral, 7 = very close). The distancing from the poor indicator was computed by subtracting rated distance from (i.e., reverse-scored closeness to) the poor from rated distance from the middle class.

Belief in class divisions

To assess a structural conception of social class, the belief in class divisions item read as follows: “In the United States traditional divisions between owners and workers still remain. A person’s social standing depends upon whether he/she belongs to the upper or lower class.” Respondents rated their agreement on a 7-point scale (1 = strongly disagree, 7 = strongly agree).

Looked down on

A final identity-type indicator, looked down on, measured the degree to which respondents felt they had been disrespected by others because of their job. The item read: “Someone has looked down on me because of the kind of job I have” (1 = yes, 0 = no).

Social and political attitudes

Items were administered tapping an array of social attitudes and political preferences. In order to facilitate comparisons between profiles’ mean scores on these dimensions, all single-item measures and questionnaire composites were rescaled to range from 0 (reflecting the minimum observed value) to 1 (reflecting the maximum observed value).

Anti-immigrant attitudes

Two sets of items were administered to gauge respondents’ attitudes toward immigration and immigrants. In the first set, we administered Altemeyer’s (1996) “Posse” scale, adapted to measure views about illegal immigration. The items were: “I would tell the police about any illegal immigrants that I knew,” “I would help hunt down illegal immigrants and turn them over to the police,” “I would participate in attacks on illegal immigrants if supervised by the proper authorities,” “I would support the use of physical violence to make illegal immigrants reveal the identity of other immigrants,” and “I would support the execution of illegal immigrants” (1 = strongly disagree, 7 = strongly agree). A composite Posse score was created (α = .86) by averaging the items.

The second set of anti-immigration items measured concern that there is “too much immigration” in the United States today (1 = not at all concerned, 7 = very concerned); support for establishing English as the official language of the United States (1 = strongly oppose, 7 = strongly support); the belief that the time necessary to become a U.S. citizen should be increased (1 = time required should be decreased, 2 = time required should stay the same, 3 = time required should be increased); and the belief that the number of immigrants allowed into the United States each year should be increased (1 = number of immigrants allowed should be decreased, 2 = number of immigrants allowed should stay the same, 3 = number of immigrants allowed should be increased; reverse-scored). These anti-immigrant items were averaged to form a composite (α = .75).

Anti-Black attitudes

Anti-Black attitudes were assessed using four items tapping respondents’ concern about “racism against Blacks” (1 = not at all concerned, 7 = very concerned; reverse-scored); belief in the existence of anti-Black discrimination (1 = no discrimination, 4 = a lot of discrimination; reverse scored); belief that police treat Blacks better than Whites (1 = treat Whites better than Blacks, 2 = treat Black and Whites about the same, 3 = treat Blacks better than Whites); and belief that “a person’s racial/ethnic background” is important for “getting ahead in society” (1 = not at all important, 5 = essential; reverse-scored). These items were averaged to form a composite measure of anti-Black sentiment (α = .66).

Political ideology and voting preferences

Respondents’ political ideologies and voting preferences were assessed. First, participants rated their levels of social and economic conservatism using two items: “In terms of social and cultural issues, how liberal or conservative are you?” and “In terms of economic issues, how liberal or conservative are you?” (1 = liberal, 6 = neither, 11 = conservative). Voting in the 2016 presidential election was probed by having respondents select from the following options: “Donald Trump,” “Hillary Clinton,” “Other” (with option to specify), and “I did not vote.” Similarly, voting in 2012 was assessed by having respondents choose between “Barack Obama,” “Mitt Romney,” “Other” (with option to specify), and “I did not vote.” Future voting intentions were measured by asking respondents whom they planned to vote for in 2020, with options of “Donald Trump,” “The Democratic Party nominee,” and “I do not plan to vote.”

Racialization of the working class

Respondents were asked to rate the likelihood that an average member of various groups (i.e., White people, Black people, Latino/as, and Asians) is poor, working class, middle class, or upper class (1 = very unlikely, 7 = very likely). In order to assess the degree to which our respondents racialize the working class (McDermott et al., 2019)—that is, think of the category working class as overlapping with the category White—we computed a difference score by subtracting the Black likelihood from the White likelihood, such that higher scores indicate more racialization of the working class.³

Analyses and Results

Means and standard deviations of, and correlations between, variables assessed among working-class Whites are displayed in Table 2.

Table 2.

Descriptive statistics for assessed variables (working-class Whites).

Variable	Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20
1. Subjective social class	3.06	.94	—	.07	.05	.44	-.13	-.19	-.04	.07	.04	.02	-.04	-.05	-.12	.08	-.06	.16	.18	.18	.14	.12
2. American identification	.75			—	.07	.14	-.13	-.09	.07	.24	.23	.12	-.19	-.11	-.10	.09	-.14	.23	.23	.18	.23	.21
3. Value of responsibility	.40				—	.14	-.04	-.02	.07	.10	.11	.07	-.14	-.19	-.08	.08	-.09	.08	.10	.06	.11	.10
4. Distancing from the poor	.51					—	-.18	-.20	.04	.12	.07	.06	-.10	-.10	-.07	.02	-.05	.17	.20	.16	.15	.14
5. Belief in class divisions	.48	.27					—	.15	-.10	-.33	-.26	-.17	.24	.18	.20	-.22	.08	-.36	-.32	-.21	-.24	-.26
6. Looked down on	.50							—	-.04	-.07	-.09	-.05	.07	.12	.17	-.08	.09	-.17	-.17	-.10	-.09	-.10
7. Working class is White	.53	.16							—	.18	.21	.15	-.21	-.18	-.07	.15	-.09	.13	.13	.05	.12	.15
8. Immigration concern	.65	.32								—	.52	.35	-.58	-.25	-.23	.33	-.17	.52	.48	.32	.47	.50
9. English official	.82	.25									—	.27	-.45	-.31	-.24	.27	-.24	.41	.40	.25	.37	.37
10. Citizenship time	.46	.33										—	-.40	-.23	-.19	.23	-.10	.25	.21	.12	.22	.25
11. Immigration increase	.38	.36											—	.35	.25	-.29	.16	-.40	-.38	-.24	-.38	-.40
12. Anti-Black racism concern	.64	.31												—	.51	-.40	.17	-.35	-.32	-.22	-.34	-.35
13. Black discrimination	.72	.27													—	-.51	.19	-.33	-.32	-.24	-.26	-.27
14. Police treat Whites worse	.24	.27														—	-.23	.38	.36	.27	.34	.38
15. Race affects opportunity	.27	.29															—	-.18	-.20	-.13	-.19	-.21
16. Social conservatism	.50	.28																—	.84	.48	.54	.55
17. Economic conservatism	.54	.27																	—	.45	.50	.53
18. Romney 2012 vote	.26																			—	.54	.51
19. Trump 2016 vote	.41																				—	.81
20. Trump 2020 intended vote	.39																					—

Note. Variables without standard deviations are proportions. All values significant at p < .05 unless italicized

Latent Profile Analysis

As a form of mixture modeling, latent profile analysis (LPA) is premised on the notion that variables’ observed distributions may reflect unobserved subgroups (“clusters”) of individuals (Oberski, 2016). The goals of LPA are to identify the number of distinct subgroups in the data and to characterize each subgroup in terms of a unique profile of parameters—typically the means of one or more indicator variables (Pastor et al., 2007). LPA bears similarities to other clustering techniques, such as cluster analysis and latent class analysis (LCA). However, unlike cluster analysis, LPA has the advantage of being model-based, with rigorous criteria for selecting an optimal solution; and unlike LCA, in which cluster indicators must be dichotomous, LPA allows for dichotomous and non-dichotomous indicators (Pastor et al., 2007).

Mplus 8.5 software (Muthén & Muthén, 2020) was used to test LPA solutions for five cluster indicators—three of which were dichotomous (American identification, value of responsibility, and looked down on) and two of which were ordinal (distancing from the poor and belief in class divisions. Restricted maximum likelihood estimation was specified. Many sets of random starting values (64,000) were tested for each model, increasing our confidence that the estimation algorithm found global (rather than local) log likelihood maxima and thus the most probable latent clusters (Vermont & Magidson, 2004).

Following Pastor et al. (2007), we tested models specifying different numbers of clusters and alternative variance–covariance matrices (Table 3). Our models specified from two to five clusters. For each number of clusters, we tested a simple model constraining the variance of the ordinal indicators (distancing from the poor and belief in class divisions) to equality across clusters and constraining the covariance between these indicators to zero within and across clusters (A models). More complex sets of models were then fitted: B models, which retained the A models’ single variance estimates across clusters but freed the ordinal indicators to covary equally across clusters; and C models, which retained the A models’ zero-covariance constraint but freed the ordinal indicators’ variances to differ between clusters. Finally, two progressively more complex models were fitted: D models, which freely estimated variances in each cluster and allowed a single covariance across clusters; and E models, which freely estimated both the ordinal indicators’ variances and their covariance in each cluster.

Table 3.

Model fit (Bayes Information Criterion) for working-class Whites; preferred model (3C) in bold.

Variance–covariance structure	Number of clusters
Variance–covariance structure	2	3	4	5
A. Equal variances; no covariance	6630	6532	6505	6451
B. Equal variances; one covariance	6634	6539	6513	6458
C. Unique variances; no covariance	6572	6521	6533	6608
D. Unique variances; one covariance	6580	6528	6538	6555
E. Unique variances and covariances	6587	6541	6552	6621

Selection of Preferred Solutions

We identified our preferred LPA solutions by inspecting the Bayes Information Criterion (BIC) and conducting X² difference tests, which compare the fit of alternative models specifying the same number of clusters (Pastor et al., 2007). We next report how we arrived at our preferred solutions for each of the samples.

Working-class Whites

One model, 5D, was immediately rejected because the solution could not be replicated even with a very large number of random starts, suggesting a poor fit to the data (Geiser et al., 2014). Models 4A, 5A, 4B, and 5B have the lowest (best) BICs. However, we rejected these models upon inspection of descriptive statistics for the resulting clusters (Pastor et al., 2007). Specifically, we observed a number of significant between-cluster differences in the variances of distancing from the poor and belief in class divisions, despite the fact that the models in question constrained these variances to equality. Model 3C—specifying three clusters with different variances and no covariance—had the next-lowest BIC and is therefore our preferred solution. Corroborating this choice, X² difference tests showed that model 3C fit the data significantly better than the simpler model 3A and did not fit significantly worse than the more complex model 3D. Figure 1a depicts each cluster’s means on the indicator variables, and Table 4 reports the average posterior probabilities associated with each of the three clusters.

Figure 1.

Indicator means and 95% confidence intervals for preferred latent profile analysis (LPA) solutions.

Table 4.

Classification table for preferred solution (working-class Whites).

Mean probability associated with cluster
	n	1	2	3
1. Working Class Connected	216	724	.019	.257
2. Class Conflict Aware	344	.080	.792	.128
3. Working Class Patriots	1473	.124	.019	.856

Non-working-class Whites

Models 2A and 2B had the lowest BICs (Table 5). However, X² difference tests showed that 2B fit significantly better than 2A and that 2D fit significantly better than 2B. We therefore retained model 2D—specifying two clusters, different variances between clusters, and a single covariance across clusters—as the preferred solution for non-working-class Whites. Bolstering this choice, X² difference tests showed that model 2D fit significantly better than the simpler model 2C and did not fit significantly worse than the more complex model 3E. Inspection of descriptive statistics for the two clusters further corroborated our choice of solutions. Each cluster’s means on the indicator variables are shown in Figure 1b.

Table 5.

Model fit (Bayes Information Criterion) for non-working-class Whites; preferred model (2D) in bold.

Variance–covariance structure	Number of clusters
Variance–covariance structure	2	3	4	5
A. Equal variances; no covariance	1920	1936	1950	1962
B. Equal variances; one covariance	1922	1940	1956	1968
C. Unique variances; no covariance	1928	1936	1966	2007
D. Unique variances; one covariance	1927	1941	1970	1999
E. Unique variances and covariances	1933	1945	1976	2004

Working-class Black people

Model 2C—specifying two clusters, different variances, and no covariance—had the lowest BIC and was retained as the preferred solution for working-class Black respondents (Table 6). Corroborating its selection, X² difference tests showed that model 2C fit the data significantly better than the simpler model 2A and did not fit significantly worse than the more complex model 2D. Inspection of descriptive statistics for the two clusters further supported our preferred solution. Each cluster’s means on the indicator variables are shown in Figure 1c.

Table 6.

Model fit (Bayes Information Criterion) for working-class Black people; preferred model (2C) in bold.

Variance–covariance structure	Number of clusters
Variance–covariance structure	2	3	4	5
A. Equal variances; no covariance	1919	1934	1942	1951
B. Equal variances; one covariance	1922	1937	1949	1957
C. Unique variances; no covariance	1906	1927	1955	1986
D. Unique variances; one covariance	1908	1930	1960	1990
E. Unique variances and covariances	1914	1943	1979	2015

Working-class Latino/as

Model 2A—specifying two clusters, one variance across clusters, and no covariance—had the lowest BIC and was retained as the preferred solution for working class Latino/a respondents (Table 7). Corroborating its selection, a X² difference test showed that model 2A did not fit significantly worse than the more complex models 2B or 2C. Inspection of descriptive statistics for the two clusters further corroborated our choice of solutions. Each cluster’s means on the indicator variables are shown in Figure 1d.

Table 7.

Model fit (Bayes Information Criterion) for working-class Latino/as; preferred model (2A) in bold.

Variance–covariance structure	Number of clusters
	2	3	4	5
A. Equal variances; no covariance	1786	1801	1812	1821
B. Equal variances; one covariance	1792	1940	1956	1968
C. Unique variances; no covariance	1796	1816	1820	1838
D. Unique variances; one covariance	1799	1815	1826	1844
E. Unique variances and covariances	1801	1820	1841	1864

Nature of the Working-Class White Clusters

Having selected a final LPA solution for working-class Whites, we proceeded to examine the three resulting clusters in terms of their associated indicator means. Figure 1a displays each cluster’s mean levels of American identification, value of responsibility, distancing from the poor, belief in class divisions, and looked down on. Clusters were labeled according to the identity type they most closely resembled. One cluster identified strongly as American, placed high value on responsibility, reported high levels of psychological distance from the poor, and tended to feel respected in their jobs; these individuals thus displayed characteristics associated with Working Class Patriot identity. Another cluster tended to identify strongly with their social class, placed relatively little emphasis on responsibility, felt close to the poor, and tended to feel disrespected in their jobs; these respondents therefore matched the characteristics of Working Class Connected identity. Finally, a third cluster displayed middling levels of American identification, responsibility values, and distancing from the poor, but possessed the highest levels of belief in structural class divisions; hence, these respondents exemplified the Class Conflict Aware identity. Overall, we found that we could readily associate each of the clusters recovered by the LPA with one of the profiles found in previous qualitative work (McDermott et al., 2019) and outlined in Table 1.⁴

Prevalence of the Working-Class White Clusters

There are two ways to determine the sizes of clusters based on an LPA. First, one can assign each respondent to the cluster that he or she has the highest posterior probability of belonging to; according to this method, our sample consisted of 1,473 (72.5%) Working Class Patriots, 344 (10.6%) Class Conflict Aware individuals, and 216 (16.9%) Working Class Connected individuals (see Table 8). Alternatively, one can average the posterior probabilities associated with each cluster and multiply this by the total sample size; this “model-based” approach suggests our sample consisted of 1,361 (67.0%) Working Class Patriots, 367 (18.0%) Working Class Connected individuals, and 305 (15.0%) Class Conflict Aware individuals (see Table 4). Regardless of how the clusters sizes are estimated, our nationally representative survey suggests that a substantial majority of working-class White Americans are Working Class Patriots, with the remaining population roughly split between Working Class Connected and Class Conflict Aware. These cluster sizes tend to corroborate McDermott et al.’s (2019) interview results.

Table 8.

Frequency and demographic characteristics of identity types.

	WCC	CCA	WCP
n of respondents (model-based)	366	305	1361
% of respondents	18.0	15.0	67.0
Predicted age	38.6	44.4	53.4
Likelihood female	.53^a	.52^a	.40^a
Likelihood post-HS educated	.59	.74^a	.70^a
Predicted income bracket	< $25,000	$25,000-34,999	$35,000-49,999

Note. WCC: Working Class Connected; CCA: Class Conflict Aware; WCP: Working Class Patriots; HS: high school. Means sharing superscript in row are not different at p < .05.

Uniqueness of the Clusters Found Among Working-Class Whites

Because we conceive of working-class identity types as different responses to a shared working-class status, we did not expect the same profiles to emerge for non-working-class Whites. Indeed, in contrast to the three types of identity found among working-class Whites, we observed only two identity types among non-working-class Whites. As seen in Figure 1b, one type, consisting of 76% of respondents, resembles Working Class Patriots (Cluster 1) and another resembles Class Conflict Aware (Cluster 2). These groups, however, do not show the substantial gap in distancing from the poor found among working-class Whites. As might be expected in a sample containing no working-class individuals, we do not see a cluster resembling the Working Class Connected identity type, members of which are theorized to feel a close personal connection to the working class (McDermott et al., 2019).

We did not have strong a priori expectations about whether the clusters found among working-class Whites would also emerge among non-White working-class respondents. Our latent profile analyses of working-class Black people and Latinos/as suggest that class identity among these groups differs considerably from that found among working-class Whites. Among these ethnoracial minority groups, we see two clusters differentiated almost entirely by belief in class divisions (i.e., whether or not respondents regard social classes in the United States as highly distinct and conflictual) and looked down on (i.e., whether or not respondents feel disparaged because of the work they do). It thus appears that disagreements about American identification and the value of responsibility, while helping form the basis for different types of White working-class identity, do not function in this way for working-class Black people and Latino/as.

Demographic Correlates of White Working-Class Identity Types

We next examined demographic correlates of the White working-class identity types. To this end, we followed Pastor et al.’s (2007, p. 26) recommendation to regress variables of interest simultaneously on the clusters’ posterior probabilities while constraining the intercept term to zero. On this approach, the resulting regression coefficients represent means of the variables of interest in each cluster, weighted by the accuracy with which individuals can be classified. To obtain significance tests of the difference between weighted cluster means, a Wald test is used to compare the fit of an unconstrained model to the fit of a model in which a pair of coefficients is fixed to equality. The results of this analysis are shown in Table 8. Table 9 displays estimates for each demographic characteristic, this time adjusting for the all remaining demographic characteristics (e.g., predicted ages are adjusted for the associations between age and likelihood female, likelihood post-high-school educated, and predicted income bracket).

Table 9.

Adjusted demographic characteristics of identity types.

	WCC	CCA	WCP
Predicted age	37.9^a	44.2^b	53.5^c
Likelihood female	.57^a	.54^a	.43
Likelihood post-HS educated	.66^ab	.76^a	.67^b
Predicted income bracket	< $25,000^a	$25,000–34,999^b	$35,000–49,999^c

Note WCC: Working Class Connected; CCA: Class Conflict Aware; WCP: Working Class Patriots; HS: high school. Means sharing superscripts in row are not different at p < .05. Estimates for each demographic characteristic are adjusted for all remaining characteristics.

We found that Working Class Patriots were significantly older than Class Conflict Aware Whites, and that Class Conflict Aware Whites were significantly older than Working Class Connected. A significant difference in gender composition emerged in the adjusted estimates, such that Patriots are less likely to be female than are Class Connected and Conflict Aware Whites. The adjusted estimates also show that Class Conflict Aware Whites tend to be more highly educated than Class Connected Whites or Working Class Patriots; although all respondents lacked a 4-year college degree, a relatively large number of Conflict Aware Whites had completed some college or received 2- or 3-year (e.g., associates) degrees or vocational training. Finally, Working Class Connected Whites had lower incomes than Conflict Aware individuals, who in turn had lower incomes than Patriots.

Social and Political Views by Identity Type

Using the regression approach (Pastor et al., 2007), we examined means for social and political attitudes by identity type. In addition to regressing outcomes on the posterior probabilities of membership in each cluster, we also adjusted for our demographic variables (i.e., age, gender, income, and education). Thus, these findings reflect the mean outcome for individuals in each identity type after accounting for demographic differences between the types.

Class self-perceptions

We first compared the identity types in terms of social-class self-identification. Although we considered all respondents to be working class, in the sense of lacking a college education, individuals varied in self-perceived social class. As shown in Table 10, a majority or substantial plurality of respondents in each identity type labeled themselves “working class.” However, members of different identity types varied in their likelihood of identifying with other class categories. Specifically Working Class Connected respondents were significantly more likely to identify as “poor” than were Class Conflict Aware respondents, who in turn were more likely to identify as “poor” than were Working Class Patriots. Conversely, Patriots were significantly more likely to see themselves as “middle class” than were Conflict Aware respondents, who in turn were more likely to label themselves “middle class” than were Class Connected Whites. This pattern corroborates the notion that the Working Class Connected identity type is distinguished by a strong subjective connection to the working class.

Table 10.

Subjective class identification by identity type.

	WCC	CCA	WCP
Poor	.23	.15	.02
Lower class	.17^a	.17^a	.12^a
Working class	.55^a	.46^a	.46^a
Middle class	.04	.19	.35

Note. WCC: Working Class Connected; CCA: Class Conflict Aware; WCP: Working Class Patriots. Means not sharing superscripts in row are different at p < .05. Adjusting for age, gender, income, and education.

Racialization of the working class

We next compared the identity types in terms of the extent to which they racialized the working class as White—that is, assumed that Whites are more likely to be working class than are Black people. Noting that a score of .5 on our racialization measure indicates that a respondent sees White and Black people as equally likely to be working class, we found that only Working Class Patriots racialized the working class as White (see Table 11). These findings corroborate McDermott et al’s (2019) ethnographic observation that only Working Class Patriots made being White a prerequisite for inclusion in the working class.⁵

Table 11.

Social and political attitudes by identity type.

	WCC	CCA	WCP
Racialization of working class	.50^a	.50^a	.54
Anti-immigrant Posse scale	.14	.08	.28
Anti-immigrant composite	.57	.46	.70
Anti-Black attitudes	.31^a	.27^a	.46
Social conservatism	.36	.29	.59
Economic conservatism	.38^a	.36^a	.62
Trump in 2016	.16^a	.21^a	.52
Obama 2012 to Trump 2016	.11^a	.10^a	.30
Trump in 2020 (intended)	.17^a	.16^a	.50
Trump 2016 to Not Trump 2020 (intended)	.15^a	.33	.11^a

Anti-immigrant sentiment

Table 11 reports results for our two measures of anti-immigrant sentiment. On both the anti-immigrant Posse scale and the anti-immigrant opinion composite, Working Class Patriots scored significantly higher than Working Class Connected respondents, who in turn scored significantly higher than Class Conflict Aware Whites.

Anti-Black sentiment

As shown in Table 11, Working Class Connected and Class Conflict Aware respondents did not differ significantly in terms of anti-Black attitudes; both groups, however, were less anti-Black than were Working Class Patriots. Here we see a discrepancy with previous qualitative work (McDermott et al., 2019), which found that the Class Connected identity was associated with progressive views on immigration but less-progressive views on race. The present findings appear to suggest the opposite—that Class Connected Whites are as progressive as Conflict Aware individuals on race but less progressive than Conflict Aware individuals on immigration.

Political preferences

Next, we examined the political ideologies and voting preferences of members of the three identity types. As shown in Table 11, Working Class Patriots were drastically more socially and economically conservative than Working Class Connected or Class Conflict Aware Whites. While Working Class Connected respondents were more socially conservative than their Class Conflict Aware counterparts, these groups did not differ in their levels of economic conservatism. Interestingly, only Working Class Patriots fell above the midpoint on either dimension of ideology; both Class Connected and Conflict Aware Whites tended to report being left-of-center.

Turning to electoral behavior, 52% of Working Class Patriots supported Donald Trump in 2016, compared to only 16% of Class Connected Whites and 21% of Conflict Aware Whites. Suggesting that this discrepancy reflected resonance with Trump’s candidacy—rather than pre-existing ideological differences between the identity types—Patriots reported having switched from voting for Obama in 2012 to Trump in 2016 significantly more often than Class Connected Whites (30% vs. 11%) and Conflict Aware Whites (30% vs. 10%). The types’ voting preferences remained largely stable between 2016 and 2020, with 50% of Working Class Patriots intending to vote for Trump in 2020 versus only 17% of Class Connected and 16% of Conflict Aware respondents. However, suggesting that Class Conflict Aware Trump voters in 2016 were often disappointed with his presidency, a relatively large percentage (33%) of this group intended not to vote for Trump again in 2020.

Discussion

Despite often being characterized as a monolithic social and political force, members of the White working class display considerable diversity in their intergroup attitudes and voting behavior (Smith & Hanley, 2018; Teixeira & Rogers, 2000; Tyson & Maniam, 2016). In an ethnographic study of working-class Whites in Kentucky, Missouri, and Indiana, McDermott et al. (2019) identified three identity types among White working-class interviewees: Working Class Patriots, who identify strongly as American, emphasize responsibility, disparage the poor, and report feeling respected in their jobs; Class Conflict Aware Whites, who see the working class as locked in a conflictual relationship with socioeconomic elites; and Working Class Connected Whites, who identify strongly as members of the working class, feel compassion toward the poor, and report feeling looked down on because of the work they do. These researchers found that the three identity types were associated with different patterns of social attitudes—with Patriots tending to disparage Black people and Latino immigrants, Conflict Aware Whites displaying progressive attitudes toward these groups, and Class Connected Whites exhibiting a combination of tolerant attitudes toward immigrants and hostile attitudes toward Black people.

The present research represents a quantitative extension of these qualitative findings. In a nationally representative sample of working-class (non-college-educated) White Americans, we measured five themes emerging from previous qualitative work: American identification, the value placed on responsibility, psychological distance from the poor, the belief in stark divisions between social classes, and the tendency to feel looked down on by members of higher classes. LPA was then used to assess whether the White American population contains discrete types resembling the Working Class Patriot, Class Conflict Aware, and Working Class Connected groups. Indeed, the best LPA solution yielded three identity types based on our five indicators, and these types could be readily matched to those found in McDermott et al.’s (2019) qualitative work (Figure 1a). The representation of the types in our survey sample broadly matched the breakdown in the ethnographic study—with Patriots making up the majority of respondents and the remaining sample split roughly between Class Conflict Aware and Working Class Connected Whites.

We also examined the social and political attitudes of the three identity types recovered by our LPA. We found that Working Class Patriots held strongly negative views of both Latino immigrants and Black people and showed the strongest preference for Donald Trump in the 2016 and 2020 presidential elections. Class Conflict Aware Whites displayed the most tolerant attitudes toward Latino immigrants and Black people and strong opposition to Donald Trump. However, the results for Working Class Connected Whites diverged somewhat from previous ethnographic results (McDermott et al., 2019). Whereas in interviews this group displayed more tolerant views of immigrants than of Black people, in the present study they were more tolerant of Black people than of immigrants. While we cannot say with certainty why this discrepancy emerged, it may reflect the specific survey items that we used or the relatively small samples in our interview work. Future research is needed to resolve this inconsistency.⁶

Origins of the Three Identity Types

As discussed at the outset, education and income are dimensions of social class that, while positively correlated, nonetheless carry opposite implications for White people’s sociopolitical affinities (Bartels, 2006; Pew Research Center, 2016, 2018; Stubager, 2009, 2013). Working-class Whites may react to these countervailing forces by cleaving into subtypes that reconcile these forces in different ways—either by adopting the right-leaning sociopolitical views typical of Whites who lack a college education or by retaining the left-leaning views long associated with lower income strata. Our data are consistent with this dynamic.

Working Class Patriots—comprising about two-thirds of our White working-class sample—deny class divisions, feel distant from the poor, report esteem in their work, and identity strongly with America. This group is socially and economically conservative, expresses negative attitudes toward immigrants and Black people, and tended to support Donald Trump in 2016 and 2020. In short, this segment of the White working class exhibits the populist social and political views increasingly found among Whites lacking a college education. The fact that Patriots make up a majority of working-class Whites may attest to the dominance of conservative media among non-college-educated Americans. Indeed, Fox News attracts fewer college-educated viewers than do CNN and MSNBC—even though income differences in the networks’ viewerships are slight (Martin, 2020).

A sizable minority within the White working class—roughly equally divided between Working Class Connected and Class Conflict Aware Whites—embraces more progressive politics. Despite lacking 4 years of college and being the least likely to have any post-high-school education, Working Class Connected Whites in our sample displayed generally progressive racial and immigration attitudes and staunchly anti-Trump voting patterns. Class Conflict Aware Whites, distinguished by their strong belief in class conflict and exploitation, are similarly liberal in their orientation. It thus appears that the Class Connected and Conflict Aware groups resist the right-leaning affinities typical of non-college-educated Whites and instead hew to a more progressive politics consistent with their low levels of income. Despite being minorities, these segments of the White working class are large enough to be pivotal in determining the direction of U.S. politics in the coming decades.

Further research should explore the factors that allow some members of the White working class to maintain progressive views in the face of a powerful new educational cleavage. Since these groups earn lower incomes than Working Class Patriots, and low income tends to be associated with progressivism (e.g., Bartels, 2006), it may be that Class Connected and Conflict Aware Whites are simply adopting a politics concordant with their economic circumstances. However, given the right-wing media ecosystem to which non-college-educated Whites are disproportionately exposed, we suspect that other experiential factors may help further dissuade these groups from embracing the politics of Working Class Patriots. For instance, both Class Connected and Conflict Aware Whites tend to report being “looked down on” because of the jobs they do—a form of class adversity that might increase empathy for other groups experiencing marginalization. It may also be that Class Connected and Conflict Aware Whites have more experience (either personally or in their social networks) with the scourge of opioid addiction, and that this form of adversity could act as an “empathy engine” that fosters compassion for a wide range of disadvantaged groups. Finally, Class Connected and Conflict Aware Whites might come to their liberal politics through different routes. For Conflict Aware Whites uniquely high in consciousness of class conflict, progressivism may flow from a politicized conceptualization of society. In contrast, Class Connected Whites, perhaps due to particularities of their educational experiences, may lack an elaborated understanding of class structure but come to liberal politics through experiences more interpersonal in nature. Although speculative, these possibilities are amenable to empirical study.

Advantages of Our Quantitative Approach

This quantitative study—in combination with previous qualitative work (McDermott et al., 2019)—represents a mixed-methods investigation of working-class identity among White Americans. As such, the present study both confirms and complements aspects of the interview work (Small, 2011). McDermott et al. (2019) proposed three discrete types of White working-class identity; our LPA analysis of survey results from a representative sample of White Americans confirms—quantitatively—that these types exist in the wider population. Moreover, the present findings suggest that associations seen in interviews between identity types and social and political attitudes are, for the most part, robust. For example, Working Class Patriots display the least tolerant attitudes toward immigrants and Black people of any of the three identity types, whereas Class Conflict Aware Whites display the most tolerant attitudes toward both of these groups. At the same time, this present work complements our interview study by yielding information that cannot be derived from a small number of geographically bounded interviews. For instance, our LPA results provide—based, as they are, on a nationally representative sample—a far more reliable estimate of the prevalence of the three identities in the American population.

Implications

The present results belie the notion that working-class Whites in the United States constitute a monolithic sociopolitical bloc. Instead, this population can be subdivided into three distinct groups, each with a different understanding of its place in the socioeconomic hierarchy. These understandings, in turn, are associated with very different views on social issues and politics.

The existence of three types of White working-class identity (Working Class Patriots, Working Class Connected, and Class Conflict Aware) carries implications for the future of ethnoracial relations and politics in the United States. Indeed, scholars have long sought to understand factors that determine working-class Whites’ stances with respect to racial and ethnic minorities. On the one hand, exposure to members of minority groups can trigger feelings of status threat among Whites, thereby exacerbating their prejudicial attitudes and conservative political leanings (Craig & Riche- son, 2014a, 2014b). Alternatively, however, it is possible that some working-class Whites might adopt a coalitional posture vis-à-vis minorities—seeking common cause on the basis of shared socioeconomic disadvantages (Cortland et al., 2017; Craig & Richeson, 2012, 2016). We suggest that working-class Whites’ choice between a competitive versus coalitional posture might depend largely on their identity type, with relatively prejudiced Working Class Patriots tending to adopt a competitive posture and relatively tolerant Class Conflict Aware tending to assume a coalitional stance. Working Class Connected Whites, for their part, are more tolerant than Patriots but lack Conflict Aware Whites’ appreciation of class structures—and thus might be susceptible to either posture depending on cues they receive within their social environments. Encouraging this pivotal group to ally—rather than compete—with minorities is a crucial step in the pursuit of intergroup harmony, economic justice, and racial equality.

The political implications of this work flow directly from those regarding race. Given the close ties between Americans’ views on race, immigration, and politics, our working-class White identity types are likely to be associated with receptivity to very different political appeals. As the present results show, Working Class Patriots were particularly drawn to Donald Trump’s candidacies. However, Working Class Connected and Class Conflict Aware Whites—representing a full third of our sample—reported breaking hard against Trump in 2016 and 2020. We therefore believe it would be unwise for progressive politicians to “write off” the White working class as potential allies. Instead, politicians and candidates should speak to the concerns and values that define Conflict Aware and Class Connected identities—perhaps by emphasizing communal values and awareness of structural socioeconomic factors—in order to create political appeals that resonate with these segments of the White working class.

Supplemental Material

sj-docx-1-gpi-10.1177_13684302221144735 – Supplemental material for Varieties of White Working-Class Identity

Supplemental material, sj-docx-1-gpi-10.1177_13684302221144735 for Varieties of White Working-Class Identity by Eric D. Knowles, Monica McDermott and Jennifer A. Richeson in Group Processes & Intergroup Relations

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Russell Sage Foundation [grant number #87-16-01].

ORCID iD

Eric D. Knowles

Supplemental material

Supplemental material for this article is available online.

Notes

References

Altemeyer

(1996). The authoritarian specter. Harvard University Press.

Ashmore

R. D.

Deaux

McLaughlin-Volpe

(2004). An organizing framework for collective identity: Articulation and significance of multidimensionality. Psychological Bulletin, 130(1), 80.

Bartels

L. M.

(2006). What’s the matter with What’s the Matter with Kansas? Quarterly Journal of Political Science, 1(2), 201–226. https://doi.org/10.1561/100.00000010

Bourdieu

(1986). The forms of capital. In Richardson

J. G.

(Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Greenwood Press.

Burke

(2020). Identity. In Kivisto

(Ed.), The Cambridge handbook of social theory (pp. 63–78). Cambridge University Press.

Clark

T. N.

Lipset

S. M.

(1991). Are social classes dying? International Sociology, 6(4), 397– 410.

Cortland

C. I.

Craig

M. A.

Shapiro

J. R.

Richeson

J. A.

Neel

Goldstein

N. J.

(2017). Solidarity through shared disadvantage: Highlighting shared experiences of discrimination improves relations between stigmatized groups. Journal of Personality and Social Psychology, 113(4), 547. https://doi.org/10.1037/pspi0000100

Craig

M. A.

Richeson

J. A.

(2012). Coalition or derogation? How perceived discrimination influences intraminority intergroup relations. Journal of Personality and Social Psychology, 102(4), 759–777. https://doi.org/10.1037/a0026481

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(6), 750–761. https://doi.org/10.1177/0146167214524993

10.

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(6), 1189–1197. https://doi.org/10.1177/0956797614527113

11.

Craig

M. A.

Richeson

J. A.

(2016). Stigma-based solidarity: Understanding the psycho-logical foundations of conflict and coalition among members of different stigmatized groups. Current Directions in Psychological Science, 25(1), 21–27. https://doi.org/10.1177/0963721415611252

12.

Cramer

K. J.

(2016). The politics of resentment: Rural consciousness in Wisconsin and the rise of Scott Walker. University of Chicago Press.

13.

Deaux

(2015). Social identity in sociology. In Wright

J. D.

(Ed.), International encyclopedia of the social behavioral sciences (2nd ed., pp. 319–324). Elsevier.

14.

Diemer

M. A.

Mistry

R. S.

Wadsworth

M. E.

López

Reimers

(2013). Best practices in conceptualizing and measuring social class in psychological research. Analyses of Social Issues and Public Policy, 13(1), 77–113. https://doi.org/10.1111/asap.12001

15.

Dietze

Knowles

E. D.

(2016). Social class and the motivational relevance of other human beings. Psychological Science, 27(11), 1517–1527. https://doi.org/10.1177/0956797616667721

16.

Frank

(2005). What’s the matter with Kansas? How conservatives won the heart of America. Metropolitan Books.

17.

Geiser

Okun

M. A.

Grano

(2014). Who is motivated to volunteer? A latent profile analysis linking volunteer motivation to frequency of volunteering. Psychological Test and Assessment Modeling, 56(1), 3–24.

18.

Gest

(2016). The new minority: White working class politics in an age of immigration and inequality. Oxford University Press.

19.

Hout

Brooks

Manza

(1995). The democratic class struggle in the United States, 1948–1992. American Sociological Review, 60(6), 805. https://doi.org/10.2307/2096428

20.

Houtman

Achterberg

Derks

(2017). Farewell to the leftist working class. Routledge.

21.

Jackman

M. R.

Jackman

M. R.

(1983). Class awareness in the United States. University of California Press.

22.

Kefalas

(2003). Working-class heroes: Protecting home, community, and nation in a Chicago neighborhood. University of California Press.

23.

Kraus

M. W.

Rheinschmidt

M. L.

Piff

P. K.

(2012). The intersection of resources and rank: Signaling social class in face-to-face encounters. In Fiske

S. T.

Markus

H. R.

(Eds.), Facing social class: How societal rank influences interaction (pp. 152–172). Russell Sage Foundation.

24.

Kraus

M. W.

Tan

J. J. X.

Tannenbaum

M. B.

(2013). The social ladder: A rank-based perspective on social class. Psychological Inquiry, 24(2), 81–96. https://doi.org/10.1080/1047840X.2013.778803

25.

Kriesi

(2010). Restructuration of partisan politics and the emergence of a new cleavage based on values. West European Politics, 33(3), 673–685.

26.

Lamont

(2000). The dignity of working men: Morality and the boundaries of race, class, and immigration. Russell Sage Foundation.

27.

Leach

C. W.

Van Zomeren

Zebel

Vliek

M. L.

Pennekamp

S. F.

Doosje

Ouwerkerk

J. W.

Spears

(2008). Group-level self-definition and self-investment: A hierarchical (multicomponent) model of in-group identification. Journal of Personality and Social Psychology, 95(1), 144.

28.

Luhtanen

Crocker

(1992). A collective self-esteem scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18(3), 302–318.

29.

Martin

C. R.

(2020, February 10). Is there a working-class cable news channel? https://workingclassstudies.wordpress.com/2020/02/10/is-there-a-working-class-cable-news-channel/

30.

McDermott

Knowles

E. D.

Richeson

J. A.

(2019). Class perceptions and attitudes toward immigration and race among working-class whites. Analyses of Social Issues and Public Policy, 19(1), 349–380. https://doi.org/10.1111/asap.12188

31.

Morgan

S. L.

Lee

(2018). Trump voters and the white working class. Sociological Science, 5, 234–245. https://doi.org/10.15195/v5.a10

32.

Muthén & Muthén. (2020). Mplus 8.5 (Version 8.5).

33.

Oberski

(2016). Mixture models: Latent profile and latent class analysis. In Robertson

Kaptein

(Eds.), Modern statistical methods for HCI (pp. 275–287). Springer.

34.

Packer

(2012, February 13). Poor, White, and Republican. The New Yorker.

35.

Pastor

D. A.

Barron

K. E.

Miller

B. J.

Davis

S. L.

(2007). A latent profile analysis of college students’ achievement goal orientation. Contemporary Educational Psychology, 32(1), 8–47. https://doi.org/10.1016/j.cedpsych.2006.10.003

36.

Pew Research Center. (2016). A wider ideological gap between more and less educated adults. https://www.pewresearch.org/politics/2016/04/26/a-wider-ideological-gap-between-more-and-less-educated-adults/

37.

Pew Research Center. (2018). Wide gender gap, growing educational divide in voters’ party identification. https://www.pewresearch.org/politics/2018/03/20/wide-gender-gap-growing-educational-divide-in-voters-party-identification/

38.

Sellers

R. M.

Smith

M. A.

Shelton

J. N.

Rowley

S. A.

Chavous

T. M.

(1998). Multidimensional model of racial identity: A reconceptualization of African American racial identity. Personality and Social Psychology Review, 2(1), 18–39.

39.

Small

M. L.

(2011). How to conduct a mixed methods study: Recent trends in a rapidly growing literature. Annual Review of Sociology, 37(1), 57–86. https://doi.org/10.1146/annurev.soc.012809.102657

40.

Smith

D. N.

Hanley

(2018). The anger games: Who voted for Donald Trump in the 2016 election, and why? Critical Sociology, 44(2), 195–212. https://doi.org/10.1177/0896920517740615

41.

Spruyt

Keppens

Van Droogenbroeck

(2016). Who supports populism and what attracts people to it? Political Research Quarterly, 69(2), 335–346.

42.

Stephens

N. M.

Markus

H. R.

Townsend

S. S. M.

(2007). Choice as an act of meaning: The case of social class. Journal of Personality and Social Psychology, 93(5), 814–830. https://doi.org/10.1037/0022-3514.93.5.814

43.

Stubager

(2009). Education-based group identity and consciousness in the authoritarian–libertarian value conflict. European Journal of Political Research, 48(2), 204–233.

44.

Stubager

(2013). The changing basis of party competition: Education, authoritarian–libertarian values and voting. Government and Opposition, 48(3), 372–397.

45.

Tajfel

Turner

J. C.

(2004). The social identity theory of intergroup behavior. In Jost

J. T.

Sidanius

(Eds.) Political psychology: Key readings (pp. 276–293). Psychology Press.

46.

Teixeira

R. A.

Rogers

(2000). America’s forgotten majority: Why the white working class still matters. Basic Books.

47.

Turner

J. C.

Hogg

M. A.

Oakes

P. J.

Reicher

S. D.

Wetherell

M. S.

(1987). Rediscovering the social group: A self-categorization theory. Basil Blackwell.

48.

Tyson

Maniam

(2016). Behind Trump’s victory: Divisions by race, gender, education. Pew Research Center. http://www.pewresearch.org/fact-%20tank/2016/11/09/behind-trumps-victory-divisions-by-race-gender-education/

49.

Vallejo

(2021). The “diploma divide”: Does it exist for racial and ethnic minorities? Michigan State University, Institute for Public Policy and Social Research. https://ippsr.msu.edu/public-policy/michigan-wonk-blog/diploma-divide-does-it-exist-racial-and-ethnic-minorities

50.

Vermont

J. K.

Magidson

(2004). Latent class analysis. In Lewis-Beck

M. S.

Bryman

Liao

T. F.

(Eds.), The SAGE encyclopedia of social science research methods. SAGE Publications, Inc. https://dx.doi.org/10.4135/9781412950589

51.

Weeden

K. A.

Grusky

D. B.

(2005). The case for a new class map. American Journal of Sociology, 111(1), 141–212. https://doi.org/10.1086/428815

52.

Williams

(2017). White working class: Overcoming class cluelessness in America. Harvard University Press.

53.

Wright

E. O.

(2015). Understanding class. Verso.

54.

Zingher

J. N.

(2022). Trends: Diploma divide: Educational attainment and the realignment of the American electorate. Political Research Quarterly, 75(2), 263–277.

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

1.58 MB