Regional Variation in Social Norms and Domestic Anti-Immigrant Hostility

Abstract

Why are some ingroup members (i.e., natives) more tolerant of, or hostile toward, outgroup others (i.e., immigrants)? A large volume of evidence exists attempting to address this issue. The bulk of findings from observational and experimental research focuses on the dual concept of ‘realistic’ and ‘symbolic’ threat, operationalized in terms of migrant stock or percentage of foreign-born population. According to the literature, a higher stock generally indicates a higher threat, which in turn implies greater xenophobic sentiment. Against this backdrop, the present study shifts the analytic emphasis to a novel explanatory mechanism in the US context: strength of social norms at the state level. Specifically, by drawing on the concept of cultural tightness-looseness, it asks whether American citizens in a geographic region characterized by strong social norms and stringent sanctioning of deviance are, on average, more xenophobic. Results from mixed effects models provide substantial support. Based on nationally representative data collected by the Pew Research Center (2018) during the height of Trump administration, the analysis reveals that cultural tightness has a contextual effect: net of controls, residency in a state with a higher index score is associated with greater levels of anti-immigrant prejudice and stronger preference for anti-immigration policy.

Keywords

cultural tightness/looseness anti-immigrant views outgroup attitude contextual effect

Introduction

Cultural Understanding of Outgroup Attitudes

Perhaps only few other social issues are more politically divisive and controversial than immigration today (Daniller, 2019). Anti-immigrant sentiments can largely be attributed to two distinct yet related sources: economic competition and cultural threat (Bloom et al., 2015; Hainmueller & Hopkins, 2015; Heath et al., 2020; Valentino et al., 2019). On the one hand, immigrants are denigrated since they reputedly steal jobs from domestic workers and lower wages. On the other, immigrants are feared because they are thought to undermine traditional values and lifestyles. As flows of transnational migration have steadily increased, in varying degree, natives across host societies have expressed hostility and even outright xenophobia at the prospects of ‘foreign invasion’ (Dennison & Geddes, 2021). The US has been no exception in this regard, with the highest proportion of immigrants according to latest statistics (McAuliffe & Triandafyllidou, 2021). While the total number of native-born Americans had fallen, the country nevertheless experienced an overall population growth due to immigrant inflows (Jordan & Gebeloff, 2022). Importantly, as has been well-publicized, the American public opinion on immigration has been fractured along partisan lines. According to a survey by the Gallup Poll, about two-thirds of (68%) Republicans, in comparison with just 17% of Democrats, preferred a restrictive immigration policy (Saad, 2022).

What accounts for such divergence? More specifically, for the main purposes of this study, why are some American citizens more critical toward and unwilling to embrace the immigrant population? Party affiliation, along with ideological conviction, certainly plays a critical role in whether natives welcome or eschew people of foreign origins (Harteveld, 2017), as do concerns of ‘realistic’ and ‘symbolic’ threat, real or imagined (Kaufmann & Goodwin, 2018; Lancee & Pardos-Prado, 2013; Schmuck & Matthes, 2017). Clearly, the bulk of existing findings is conceptually driven by these two forms of threat.¹ The debatable issue is whether they are “best understood either in terms of economic resources or cultural identities” (Helbling & Traunmüller, 2016, p. 398). According to some, immigration-induced material competition generates a subjective state of relative deprivation among natives vis-à-vis the foreign-born, resulting in outgroup prejudice (Gest et al., 2018; Meuleman et al., 2020; Mutz, 2018; Pettigrew et al., 2008). Other scholars contend that negative sentiments toward outgroup members stem from a sense of symbolic insecurity surrounding cultural or ethnic identity (Choi et al., 2019; Gorodzeisky & Semyonov, 2019; Newman et al., 2012; for review, see Alesina & Tabellini, 2022) or reflect a reactionary ‘cultural backlash’ against globalization (Hochschild, 2016; Norris & Inglehart, 2019).

Regardless of the analytic focus, there is a common denominator underlying much of the extant research: the notion of threat is operationalized by the size of migrant stock, i.e., proportion of immigrant population, as a meta-analysis points out (Kaufmann & Goodwin, 2018). That is, the larger the presence of immigrants, the greater the fear. Equipped with this guiding idea, scholars have sought to investigate the link between volumes of immigration, both in terms of size and change, and natives’ perception/reception of the foreign-born. Against this backdrop, and also building on and extending prior research, the current study introduces a novel heuristic concept in understanding anti-immigrant attitudes in the US context: strength of social norms measured at the state level. Is there a relationship between living in a more culturally strict state and xenophobic orientations and preferences among residents? Using population-based data and multiple model specifications, this paper takes up this question.

Aside from rare exceptions (e.g., Jackson, Van Egmond, et al., 2019; Leong & Ward, 2006), studies analyzing the impact of cultural environment in relation to outgroup attitudes are scarce. To the extent that culture is examined, both experimental and observational attempts focus on specific value items gauged at the individual-participant level (e.g., Choi et al., 2019; Eisentraut, 2019; Havermans & Verkuyten, 2021; Hövermann & Messner, 2019; Moss et al., 2019; Seewann, 2022; Shin & Dovidio, 2018; Visintin et al., 2020; Wollast et al., 2022). As one survey-based study demonstrates, Europeans who espouse universal versus traditional values are more likely to support a liberal immigration policy (Davidov et al., 2020). In experimental settings, subjects holding higher self-transcendence values (e.g., equality) are similarly shown to have a more favorable view of immigrants (Wolf et al., 2019), whereas conservation values (e.g., anxiety-avoidance) predict less willingness to interact with them (See et al., 2020). While such findings highlight the importance of personally held values, they fall short of considering the broader milieu in which people are culturally embedded.² The present study adds to the literature by proposing and demonstrating that the strength of social norms of a particular place, over and above existing immigration-induced threat, has a distinct impact on people’s outgroup attitudes.

Cultural Tightness as the Key Mechanism

As has been observed, outgroup prejudice and discrimination are fundamentally ‘context-dependent’ (Igarashi & Nagayoshi, 2022; Álvarez-Benjumea, 2022), i.e., they shift partly as a function of local normative contexts (Di Santo et al., 2022). Research on the relationship between terrorist attacks and xenophobia, for instance, shows that prejudiced attitudes are explicitly articulated only when they are perceived to be socially acceptable (Álvarez-Benjumea & Winter, 2020). Expressions of prejudice, in short, are not universal but ‘culture specific’ (Fiske, 2017). By moving beyond individual-level ‘value conceptualizations’ and ‘value priorities’ (Seewann, 2022), this paper offers a ‘geographical account of psychological phenomena’ (Chen et al., 2020; see also Frasure-Yokley & Wilcox-Archuleta, 2019). That is, its main objective is to probe how the boundary and enforcement of social norms at a macro level differentially impact anti-immigrant sentiments at a micro level. To that end, this study draws on the concept of cultural tightness-looseness (TL), a validated index of the strength of social norms and strictness with which deviance is socially sanctioned (Gelfand et al., 2017).

Research shows that there are significant cross-national differences along the TL dimension (Van Bavel et al., 2020). Originally proposed by Gelfand et al. (2011) based on data across 33 countries, the concept refers to “a shared cultural construct with high within-nation agreement and high between-nation variance.” This cultural dichotomy broadly refers to subjective perceptions and understandings of what is normative in society. Tightness denotes strong social norms, “implicit or explicit rules that constrain behavior without the force of laws” (Stamkou et al., 2019, p. 947), and greater conformity with relatively little latitude. In tighter societies, norms tend to be clearer and norm violations are more harshly punished. Such societies thus prioritize, above all, cultural homogeneity along with social order (Stamkou et al., 2019). In a subsequent piece, Jackson et al. (2020) corroborated earlier results by extending the argument to non-industrial societies. Cultural tightness has been found to be associated with diverse outcomes such as support for authoritarianism (Jackson, Van Egmond, et al., 2019), perceptions of effective leadership (Aktas et al., 2016), proclivity for social comparison (Baldwin & Mussweiler, 2018), country-level innovation (Deckert & Schomaker, 2022), and compliance with public health protocols (Gilliam et al., 2022).

The TL scale applies not only inter- but also intra-nationally. For example, within China significant variations along this dimension exist with profound implications, i.e., geographic differences in, among others, urbanization, economic growth, and mental health (Chua et al. (2019). Related to the present study’s main objective, social norm differences are also shown regionally across the US. There are culturally tight places such as Mississippi and Alabama and culturally loose parts of the country including New Hampshire and California, with the former (latter) having more (less) stringent normative rules/regulations with lower (higher) tolerance for deviance (Harrington & Gelfand, 2014). For historical reasons, the US states have evolved to generate and sustain adaptive cultures of varying tightness/looseness. A critical observation is that today tighter cultures are generally less inclusive, diverse, and permissive (Deckert & Schomaker, 2022; Gelfand et al., 2011; Jackson, Van Egmond, et al., 2019). By extension, this geographic distinction implies that cultural tightness operationalized at the state level, net of compositional characteristics, may be predictive of Americans’ views on unauthorized immigrants. Existing evidence is highly suggestive. Among US adults, cultural tightness is linked with Eurocentric normative standards in defining national identity (Lopez et al., 2022). Cross-nationally, it is found that variation in the strictness of social norm accounts for the variation in prejudice against ethnic minorities (Jackson, Van Egmond, et al., 2019).

Study Hypothesis

Under conditions of external threat (e.g., immigration), motivations for ingroup favoritism and outgroup exclusion often get amplified (Boyer et al., 2015). In the context of ingroup homogeneity, outsiders become criticized, feared, and rejected by the majority (e.g., natives). Put differently, threats create pressures on groups to maintain tighter normative rules and stricter sanctioning mechanisms to optimize their survival (Gilliam et al., 2022; Roos et al., 2015; Van Bavel et al., 2020). A major implication here is that in culturally tighter places such pressures would become more reinforced. A critical question for this study, then, is whether and to what extent the spatial distribution of cultural tightness may exacerbate intergroup relations and, in particular, the perception of outsiders (i.e., immigrants). A long-held assumption in the extant scholarship is that immigrants pose a viable group threat, objectively or subjectively, to the native population (Quillian, 1995). Consequently, much of research efforts has been devoted to unpacking the processes underlying this threat-xenophobia connection (for review, see Esses, 2021). As a result, despite the fact that a voluminous literature has emerged covering a multitude of explanatory factors, the role of social norms has not been fully incorporated.³ In short, previous accounts have rarely examined the ‘contextual’ role of culture. Given this limitation, this paper proposes and demonstrates that the tightness-looseness dimension offers a unique prism through which anti-immigrant views can be examined, over and above the externally imposed realistic and/or symbolic threat.

At the ecological level, tight states in the US are shown to correlate with strong disapproval of social deviance or norm violation (Gelfand et al., 2017; Jackson et al., 2020). It is also the case that those who espouse ‘binding’ values—which are based on moral beliefs prioritizing groups over individuals—prefer greater cultural tightness, i.e., they desire strong social norms and have lower tolerance for deviant behavior. Moreover, binding moral foundations imply concern for and preoccupation with preserving collective entities such as communities and national identities “through duties, loyalty, and (social and physical) purity” (Di Santo et al., 2022, p. 2). According to Harrington and Gelfand (2014, p. 7992), “tightness describes an external social reality that exists independent of any one individual.” In other words, the concept operates fundamentally at an environmental level. Taken together, it is thus conjectured that ingroup members in geographic contexts with strong social norms are more resistant to outgroup members, that is, those perceived to be unfit or incompatible with the status quo.

Hence, narrowly focusing on immigrant-induced threat in and of itself may provide a skewed, if not biased, analysis of why some natives are more xenophobic. Surely, the rise of foreign-born population can elicit negative responses. Yet, the central underlying mechanism, as this paper argues, may reside with a broader cultural context in which immigration takes place, more specifically, at the level of cultural tightness. Based on previous discussions, the following main hypothesis is presented and tested: all else equal, residency in a region (state) with a higher score on the cultural tightness scale is associated with greater xenophobic prejudice toward unauthorized immigrants among the US native population. Below, based on nationally representative data, statistical analyses are carried out and findings discussed for the hypothesis testing.

Methodology

Participants and Procedure

Data are drawn from the June Political Survey 2018 conducted by Pew Research Center (2018), which are publicly available at its online repository (https://www.people-press.org/dataset/september-2018-political-survey). The Center maintains eight broad areas of survey research including “U.S. Politics and Policy” under which the Political Survey series falls. Its primary goal is to gauge the trends in American public opinion on a host of domestic and international issues. The June 2018 version was completed between June 5-12 among a nationally representative sample of 2,002 American adults living in the 50 States and the District of Columbia. Data were collected via phone interviews in English and Spanish. The (500 landline and 1,500 cell phone) combined sample was weighted using an iterative technique that matches gender, age, education, race, and region to parameters from the 2016 Census Bureau’s American Community Survey one-year estimates and population density to parameters from the Decennial Census.

Measures: Dependent Variables

There are two outcomes: Xenophobic Sentiment and Restrictive Immigration. Both measure the study participants’ views on people of foreign origins residing in the US specifically without authorization, an important distinction as natives react to immigrants differently based on their legal status (Hainmueller & Hopkins, 2015). They are also among the most frequently examined dependent variables in the scholarship under the rubric of ‘prejudice and discrimination’ (Esses, 2021; see also Kaufmann & Goodwin, 2018). Xenophobic Sentiment is based on the following: “When it comes to undocumented immigrants currently in the U.S which comes closer to your view — even if neither is exactly right? 1) Undocumented immigrants mostly fill jobs that American citizens would like to have [OR] Undocumented immigrants mostly fill jobs American citizens don’t want; 2) Undocumented immigrants living in the U.S. are more likely than American citizens to commit serious crimes [OR] Undocumented immigrants living in the U.S. are no more likely than American citizens to commit serious crimes.” For both statements, agreement with the former phrase is coded 1 Given the skewed distribution, responses are summed and then dichotomized.⁴ Restrictive Immigration is measured using a single survey item (‘As you may know, many immigrants who came illegally to the U.S. when they were children now have temporary legal status that may be ending. Would you favor or oppose Congress passing a law granting them permanent legal status?’; coded 1 if oppose).⁵

Measures: Predictor Variable

The main predictor, Cultural Tightness, is based on the state-level index created by Harrington and Gelfand (2014), which comprises 9 subcomponents. Four gauge the strength of punishment (legality of corporal punishment in schools, percentage of students hit/punished, rate of executions, and severity of punishment for violating state laws), two tap latitude and permissiveness (access to alcohol and legality of same-sex civil union), two additional items measure aggregate religiosity and percentage of people without religious affiliation, and the last one, as a proxy for cultural diversity, is the percentage of total population that is foreign-born (reverse coded). All items are standardized and summed into a composite tightness score for each state (for more detail, see Online Supplement in Harrington & Gelfand, 2014).

Individual-Level Covariates

To adjust for potential confounding, the statistical models include multiple covariates at individual and regional units of analysis. At the former level, following socioeconomic and demographic variables are included, which have been conventionally shown to be associated with outgroup attitudes (for review, see Esses, 2021; Hainmueller & Hopkins, 2015; Kaufmann & Goodwin, 2018): Age (in years), Male (coded 1; ref.: female), Non-Hispanic White (coded 1 if self-identified as such; 0 otherwise),⁶ Education coded on an 8-point scale (e.g., 1 = Less than high school 3 = Some college, no degree, 5 = Four year college or university degree/Bachelor’s degree, 8 = Postgraduate or professional degree), and Household Income coded on a 9-point ordinal scale (e.g., 1 = ‘Less than $10,000,’ 3 = ’20 to under $30,000,’ 5 = ’40 to under $50,000,’ 7 = ’75 to under $100,000,’ 9 = ‘$150,000, or more’). Importantly, given its strong association with attitudes toward immigrants/immigration (Saad, 2022), self-described political affiliation is adjusted for (Republican = 1; 0 otherwise). Also included are two dummy indicators (Most Dense and Dense) for population density at the residential level as a proxy for cosmopolitan identity linked with outgroup orientation (Maxwell, 2020). Finally, to control for intergroup contact, two additional binary indicators are added (Often Contact and Sometimes Contact) that tap the frequency of social interaction with foreign-born populations (Boin et al., 2021; Meleady et al., 2017).

State-Level Covariates

At the regional level, following macroeconomic measures are included: GDP (gross domestic product), Poverty Level, and Unemployment Rate based on data from the Migration Policy Institute (https://www.migrationpolicy.org/programs/migration-data-hub). Inclusion of these controls serves to minimize alternative explanations (related to realistic or economic threat) other than the one advanced in this study. To that end, as a proxy for symbolic threat (see Kaufmann & Goodwin, 2018), a further adjustment is made to account for the effect of diversity by adding a generalized index of within-state ethnolinguistic fractionalization (GELF). Inclusion of this variable is critical “because heterogeneity prevents the consolidation of strong group norms” (Jackson et al., 2020, p. 2). GELF is based on a Herfindahl-type formula estimating the probability that two randomly selected individuals in a geographic location belong to different sociodemographic subgroups (Bossert et al., 2011). In the literature, migrant stock is typically included. But since the TL scale already incorporates the proportion of foreign-born population in its calculation, GELF offers a methodologically sound alternative that avoids the issue of multicollinearity. Descriptive statistics for all the variables are shown in Table 1. Table 2 provides state-by-state scores and rank orders for cultural tightness.

Table 1.

Descriptive Statistics.

	Mean/Proportion	SD	Min	Max
Outcome measure
Xenophobic Sentiment	36%	1.46	0	1
Restrictive Immigration	22%		0	1
Anti-Immigrant Index	1.3		0	5
Individual-level covariates
Age	51.94	18.08	18	94
Male	55%		0	1
Non-Hispanic White	70%		0	1
Education	5.04	1.81	0	8
Household Income	5.73	2.31	1	9
Republican	27%		0	1
Often Contact	34%		0	1
Sometimes Contact	23%		0	1
Most Dense	19%		0	1
Dense	23%		0	1
State-level covariates
Cultural Tightness	0	1	−1.81	2.28
GDP	0	1	−.71	4.94
Poverty Level	0	1	−1.87	2.33
Unemployment Rate	0	1	−1.66	3.28
GELF	0	1	−2.10	1.90

Data source: Political Survey 2018 (Pew Research Center)

Notes: GDP = gross domestic product, GELF = generalized ethnolinguistic fractionalization. All state-level variables have been standardized as z-scores.

Table 2.

Cultural Tightness Variation Across US States.

Rank	State	Score
1	Mississippi	78.86
2	Alabama	75.45
3	Arkansas	75.03
4	Oklahoma	75.03
5	Tennessee	68.81
6	Texas	67.54
7	Louisiana	65.88
8	Kentucky	63.91
9	South Carolina	61.39
10	North Carolina	60.67
11	Kansas	60.36
12	Georgia	60.26
13	Missouri	59.60
14	Virginia	57.37
15	Indiana	54.57
16	Pennsylvania	52.75
17	West Virginia	52.48
18	Ohio	52.30
19	Wyoming	51.94
20	North Dakota	51.44
21	South Dakota	51.14
22	Delaware	51.02
23	Utah	49.69
24	Nebraska	49.65
25	Florida	49.28
26	Iowa	49.02
27	Michigan	48.93
28	Minnesota	47.84
29	Arizona	47.56
30	Wisconsin	46.91
31	Montana	46.11
32	Illinois	45.95
33	Idaho	45.50
34	Maryland	45.50
35	New Mexico	45.43
36	Rhode Island	43.23
37	Colorado	42.92
38	New Jersey	39.48
39	New York	39.42
40	Alaska	38.43
41	Vermont	37.23
42	New Hampshire	36.97
43	Hawaii	36.49
44	Connecticut	36.37
45	Massachusetts	35.12
46	Maine	34.00
47	Nevada	33.61
48	Washington	31.06
49	Oregon	30.07
50	California	27.37

Data source: Harrington and Gelfand (2014).

Analytic Strategy

Data contained in Political Survey 2018 consist of individual respondents nested in regional units. To examine the contextual effect of social norms (i.e., cultural tightness) on American citizens’ anti-immigrant attitudes—and to accommodate the hierarchical data structure—multilevel modeling is employed. The value for Cultural Tightness, as well as for other state-level covariates, is not available for the District of Columbia. As a result, the survey participants residing there (N = 11) are excluded. With the listwise deletion of case with missing values, the analytic samples contain 1,726 and 1,744 respondents across 50 states for the two outcomes (Xenophobic Sentiment and Restrictive Immigration, respectively).⁷ Respondent-level weights provided in the dataset are applied to account for sample selection and to compensate for nonresponse rate. Both outcome measures are dichotomous. However, since coefficients from logistic regression cannot be compared directly across different model specifications, mixed-effects linear probability models (LPMs) are fitted as conventionally practiced (Abascal, 2020; for an overview, see Breen et al., 2018).⁸

LPMs allow for comparison of results across nested models in terms of probability changes (Mood, 2010). As summarized in Tables 3 and 4, findings are presented with different model specifications. A parameter estimate from multivariate LPMs reflects a corresponding shift in the probability that Y = 1. In the present study, this refers to the probability of having xenophobic attitude or supporting restrictive immigration for a unit increase in the explanatory variable of interest (i.e., cultural tightness), while holding other covariates constant.⁹ Based on this, running LPMs makes it possible to directly compare coefficients for cultural tightness estimated across multiple models. All statistical analyses use the restricted maximum likelihood (REML) estimator to address inferential deficiencies associated with small numbers of higher-level clusters. Though the number of clusters in the Political Survey data (N = 50) exceeds the critical threshold of 20, this precautionary measure was taken to minimize potential anti-conservative statistical inferences. Using appropriate degrees of freedom is also necessary since the standard (normal distribution) assumption regarding the t-statistics may be violated. As a remedy, the suggested “m – l – 1 rule” is followed (Elff et al., 2021) using the specialist package HLM 8.1 (Raudenbush et al., 2019).

Table 3.

Mixed-Effects LPMs Predicting the American Public Opinion on Unauthorized Immigrants (Full Sample).

DV	Model 1 Xenophobic Sentiment		Model 2 Xenophobic Sentiment		Model 3 Xenophobic Sentiment		Model 4 Restrictive Immigration		Model 5 Restrictive Immigration		Model 6 Restrictive Immigration
DV	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)
(State-level)
Cultural tightness	.050^**	(.015)	.053^**	(.020)	.038^*	(.015)	.043^***	(.012)	.035^*	(.016)	.029^**	(.010)
GDP			−.025^*	(.011)	−.016†	(.008)			−.023^**	(.008)	−.008†	(.005)
Poverty level			.017	(.025)	−.023	(.023)			.013	(.017)	−.015	(.012)
Unemployment rate			.031	(.021)	.036†	(.019)			.004	(.022)	.011^**	(.010)
GELF			−.058^*	(.022)	−.021	(.022)			−.035†	(.017)	−.011	(.013)
(Individual-level)
Age					.002^**	(.001)					.001	(.001)
Male					.080^**	(.025)					.091^***	(.020)
Non-hispanic white					−.051	(.032)					−.010	(.029)
Education					−.029^***	(.006)					−.008	(.005)
Household income					−.023^***	(.006)					−.001	(.004)
Republican					.291^***	(.031)					.236^***	(.026)
Often contact					−.140^***	(.025)					−.142^***	(.024)
Sometimes contact					−.147^***	(.032)					−.164^***	(.030)
Most dense					−.030	(.038)					−.039	(.025)
Dense					−.087^**	(.028)					−.053^*	(.026)
Constant	.387^***	(.015)	.398^***	(.015)	.370^***	(.014)	.231^***	(.012)	.245^***	(.011)	.233^***	(.012)
Random effects
Individual-level variance	.209		.209		.165		.155		.155		.125
State-level variance	.004^**		.002^*		.001^*		.002^*		.000		.000^*
Deviance	2558.737		2545.914		2065.137		1955.017		1942.461		1510.138
Sample size (L1/L2)	1,726/50		1,726/50		1,726/50		1,744/50		1,744/50		1,744/50

Data source: Political Survey 2018 (Pew Research Center). Notes: Robust standard errors (SE) are in parentheses; DV = dependent variable; LPMs = linear probability models; L1 = individual-level; L2 = state-level.

†p < .10 *p < .05 **p < .01 ***p < .001.

Table 4.

Robustness Checks Using Mixed-Effects LPMs Based on the Non-hispanic White Subsample.

DV	Model 1 Xenophobic Sentiment		Model 2 Xenophobic Sentiment		Model 3 Xenophobic Sentiment		Model 4 Restrictive Immigration		Model 5 Restrictive Immigration		Model 6 Restrictive Immigration
DV	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)
(State-level)
Cultural tightness	.069^***	(.018)	.095^***	(.020)	.058^***	(.016)	.054^***	(.015)	.065^**	(.019)	.040^**	(.013)
GDP			−.027^*	(.011)	−.025^**	(.009)			−.029^**	(.008)	−.020^**	(.006)
Poverty level			−.009	(.027)	−.038	(.027)			−.019	(.022)	−.035†	(.019)
Unemployment rate			.067^**	(.024)	.069^***	(.019)			.050^*	(.025)	.047^**	(.019)
GELF			−.081^**	(.024)	−.047†	(.025)			−.038^a	(.021)	−.016	(.018)
(Individual-level)
Age					.002^**	(.001)					.001	(.001)
Male					.085^**	(.029)					.099^***	(.028)
Non-hispanic white					___	___					___	___
Education					−.025^**	(.008)					−.015†	(.008)
Household income					−.025^**	(.008)					.001	(.006)
Republican					.297^***	(.033)					.208^***	(.027)
Often contact					−.129^***	(.027)					−.148^***	(.035)
Sometimes contact					−.132^***	(.036)					−.179^***	(.037)
Most dense					.004	(.041)					−.048	(.032)
Dense					−.086^**	(.028)					−.054	(.034)
Constant	.389^***	(.017)	.396^***	(.016)	.383^***	(.015)	.254^***	(.015)	.266^***	(.013)	.253^***	(.012)
Random effects
Individual-level variance	.209		.209		.167		.166		.166		.138
State-level variance	.005^**		.001		.000		.003^*		.000		.000^*
Deviance	1743.954		1724.700		1418.027		1429.031		1415.836		1165.799
Sample size (L1/L2)	1,260/50		1,260/50		1,260/50		1,262/50		1,262/50		1,262/50

†p < .10 *p < .05 **p < .01 ***p < .001.

Results

Concerning the outcomes, as descriptive statistics show in Table 1, slightly more than a third (36%) of the sample agreed with at least one of the two xenophobic statements describing unauthorized immigrants (i.e., that they are in direct economic competition with American laborers and/or pose a physical security threat). A little less than a quarter (22%) expressed a desire to deport unauthorized immigrants rather than allow them to remain legally in the country. With respect to the main predictor, cultural tightness, Table 2 provides a breakdown of state-by-state variation. At the top is Mississippi (with the score of 78.86), immediately ahead of Alabama in second place. The least culturally tight state, listed at the bottom, is California (with 27.37), right below Oregon. A critical implication here is that the residents of, for example, Mississippi (with more stringent social norms) exhibit greater anti-immigrant attitudes on average than those living in, say, California (with less stringent social norms). To further investigate this possibility, the output from estimating mixed-effects linear probability models is presented in Table 3.

According to the null or unconditional model (not shown), the intraclass correlation (ICC) indicates that less than 5% of the total variance in both outcomes are due to clustering. However, along with the significant higher-level variance, the size of design effect (DEFF) exceeds the threshold, validating the use of multilevel analysis despite the relatively small ICC value (Maas & Hox, 2004; see also Musca et al., 2011).¹⁰ Models 1 – 3 are based on Xenophobic Sentiment as the outcome; Models 4 – 6 correspond to Restrictive Immigration. According to Model 1, based on bivariate analysis without any of the controls, the coefficient for Cultural Tightness is significantly positive (ß = .05, p < .01). A similar finding surfaces in Model 4 (ß = .043, p < .001). Substantively, this indicates initial evidence for the proposed hypothesis: living in a tighter state implies more negative sentiments toward the (unauthorized) foreign-born outgroup. This bivariate relationship, however, could be spurious. The shown associations between cultural tightness and the two outcomes may be confounded by state-level characteristics, such as macroeconomic indicators or ethnic diversity. To check for spuriousness, Model 2 (for Xenophobic Sentiment) and Model 5 (for Restrictive Immigration) incorporate measures for economic size, poverty level, unemployment rate, and ethnolinguistic heterogeneity. Conditioning on these contextual variables, consistent results are found, although both the effect size and the significance level diminish with respect to Restrictive Immigration (ß = .053, p < .01, for Model 2; ß = .035, p < .05, for Model 5). But these extended analyses are still limited because they do not account for compositional effects or individual-level covariates. Omitting them would be critical since natives’ attitudes toward immigrants, unauthorized or not, are powerfully driven by socioeconomic and demographic factors. Models 3 and 6 thus contain fully specified results by including them.

As visible in Model 3, except for perceived racial classification and one of the dummy proxies for residential density, the rest are significantly predictive of Xenophobic Sentiment. The state’s GDP is negatively associated with the outcome while unemployment rate is positively associated, albeit marginally (p < .1). After adjusting for a host of confounders across individual and contextual levels, the parameter estimate for Cultural Tightness becomes reduced in terms of size but remains statistically robust (ß = .038, p < .05). Proceeding with Model 6, two additional covariates at the individual level fail to predict Restrictive Immigration: age and education. In contrast to the previous model (Model 3), there is now a significantly positive relationship between state-level unemployment rate and the outcome (p < .05). Importantly, net of controls, it is shown once again that living in a culturally tighter state is positively associated with stronger preference for anti-immigration policy (ß = .029, p < .01).

Robustness Checks

Political Survey 2018 does not contain information on the immigrant status of respondents; they can only be distinguished based on subjectively perceived ethnic/racial affiliation. Given this data limitation, the models discussed above were re-estimated using a subsample of self-identifying non-Hispanic whites (i.e., by excluding all ethnic minorities who were not US-born). Findings are displayed in Table 4. Looking at the main result for Cultural Tightness across six models, the parameter estimates are greater in size vis-à-vis those discussed earlier from Table 3 using the full sample of natives and potential immigrants alike. In other words, associations between the strength of social norms and anti-immigrant prejudice are more pronounced when assessed in the narrower context of the majority subsample. For example, according to Model 3 in Table 3 based on the full sample, the coefficient for Cultural Tightness predicting Xenophobic Sentiment is .038 (p < .05). In comparison, as shown in Model 3 in Table 4 using the subsample, it is estimated at .058 (p < .001). With respect to Restrictive Immigration, Model 5 in Table 3 based on the full sample initially shows the coefficient to be .035 (p < .05). Similarly, a subsequent analysis of the subsample (Model 5 in Table 4) yields a higher estimate of .065 (p < .01).

It should be noted, however, that two sets of coefficients from the two different samples are not distinguishable at the conventional (i.e., 95%) level of statistical significance. Running an interaction term between cultural tightness and race using the original (full) sample confirmed this. Hence, supplementary findings based on the subsample consisting of those who identify themselves as being non-Hispanic white should be interpreted simply as a general confirmation of those presented earlier using the entire dataset. As a further robustness check, an entirely new dependent variable (Anti-Immigrant Index) was constructed by merging five separate survey items: the three original ones (used to operationalize Restrict Immigration and Xenophobic Sentiment) and two new ones (based on whether the participants support Trump’s plans to build a US-Mexico border wall and whether they reject granting citizenship status to the children of unauthorized immigrants in the US). Original answers were summed to create an index, which varies from 0 to 5, with relatively high internal consistency (Cronbach’s α = .7). Table 5 summarizes the alternative results. According to the unconditional model (Model 1), the ICC is shown to be .08. Model 2 contains only the individual-level controls, or compositional effects. Based on alternative analytic strategies—multilevel linear regression in Model 3 and multilevel Poisson regression in Model 4), parameter estimates with respect to the key variable (Cultural Tightness) provide further evidence in support of the main thesis: all else equal, the presence of stronger social norms shows greater outgroup prejudice.

Table 5.

Additional Robustness Check Using the New Outcome Variable.

DV = Anti-Immigrant Index	Model 1		Model 2		Model 3		Model 4
DV = Anti-Immigrant Index	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)	Coef.	(SE)
(Individual-level)
Age			.008^***	(.002)	.008^***	(.002)	.006^***	(.001)
Male			.376^***	(.068)	.376^***	(.068)	.306^***	(.064)
N-H White			−.059	(.089)	−.057	(.094)	−.064	(.086)
Education			−.079^***	(.023)	−.078^***	(.023)	−.070^***	(.019)
Household Income			−.054^***	(.016)	−.054^***	(.016)	−.039^**	(.013)
Republican			1.327^***	(.099)	1.313^***	(.101)	.082^***	(.073)
Often Contact			−.512^***	(.068)	−.515^***	(.070)	−.377^***	(.062)
Sometimes Contact			−.631^***	(.092)	−.628 ^***	(.091)	−.482^***	(.078)
Most Dense			−.299^***	(.086)	−.272^**	(.086)	−.254^**	(.084)
Dense			−.353^***	(.066)	−.322^***	(.073)	−.267	(.063)
(State-level)
Cultural Tightness					.148^***	(.041)	.141^***	(.032)
GDP					−.028	(.022)	−.024	(.018)
Poverty Level					−.039	(.052)	−.045	(.042)
Unemployment Rate					.095	(.056)	.089	(.046)
GELF					−.049	(.056)	−.056	(.048)
Constant	1.441^***	(0.066)	1.356^***	(.034)	1.362^***	(.036)	.123^***	(.032)
Between-country variance	.129^***		.031^***		.011^*		.004
Within-country variance	1.568		1.021		1.021		__
-2logL	6603.457		5722.153		5708.121		5498.570

Data source: Political Survey 2018 (Pew Research Center)

Notes: Results in Model 4 are from estimating multilevel Poisson regression. The rest (M1 – M3) are based on multilevel linear regression.

**p < .01; ***p < .001.

Lastly, in multilevel analysis with a relatively small number of higher-level units (clusters), influential cases can pose a problem since the main findings could be driven by them (Van der Meer et al., 2010). As recommended, Cook’s D values are calculated as a diagnostic tool to detect any potential outliers (Helbling & Traunmüller, 2016).¹¹ Appendix 1 (for Xenophobic Sentiment) and Appendix 2 (for Restrictive Immigration) contain results from plotting the Cook’s D values (on the Y-axis) for each of the fifty states. Those that exceed the critical threshold (4/N – k – 1, where N = number of cases and k = number of variables), i.e., influential cases, are shown in boldface letters. In Appendix 1 (in the order of Cook’s D value) Vermont, Mississippi, Nevada, and Massachusetts qualify as influential cases. In Appendix 2, they include South Carolina, Vermont, and Arkansas. Deleting these influential cases (states) from the analysis would mean dropping a significant number of observations (respondents) nested in them, leading to a loss of statistical power. A remedy is to add a fixed-effect dummy variable for each influential case in the model (Van der Meer et al., 2010). As a conservative test, findings from simultaneously including all the state-level fixed effects for influential cases are displayed in Table 6 using a subsample of self-identified non-Hispanic whites. Across all four models (with and without adjusting for compositional effects), coefficients for Cultural Tightness emerge as consistently robust with respect to both Xenophobic Sentiment (ß = .04, p < .001 in Model 2) and Restrictive Immigration (ß = .025, p < .05 in Model 4).

Table 6.

Sensitivity Analysis Adjusting for Influential Cases.

DV	Model 1 Xenophobic entiment	Model 2 Xenophobic Sentiment	Model 3 Restrictive Immigration	Model 4 Restrictive Immigration
Highest Cook’s D value	0.228	0.228	0.099	0.099
Critical threshold^a	0.086	0.086	0.085	0.085
State with the highest value	VT	VT	SC	SC
State-level (L2) fixed effects	VT	VT	SC	SC
	MS	MS	VT	VT
	NV	NV	AR	AR
	MA	MA
Individual-level (L1) covariates	No	Yes	No	Yes
Cultural Tightness	0.076^*** (.014)	.040^*** (.013)	.055^*** (.012)	.025^* (.011)
Constant	0.368^*** (0.014)	0.370^*** (0.013)	0.242^*** (0.013)	0.241^*** (0.011)
L2 variance (τ)	0.000^*	0.000^*	0.000^*	0.000^*
L1 variance (σ²)	0.225	0.185	0.177	0.150
Deviance (-2LL)	1706.449	1472.502	1409.563	1226.287
N (L1/L2)	1,260/50	1,224/50	1,262/50	1,226/50

Data source: Political Survey 2018 (Pew Research Center)

Notes: The analysis is based on the non-Hispanic white subsample. Standard errors are in parentheses.

Models 2 and 4 adjust for the individual-level controls as shown in Tables 3 and 4 DV = dependent variable.

VT = Vermont, SC = South Carolina, NV = Nevada, MS = Mississippi, MA = Massachusetts, AR = Arkansas.

^aThe threshold formula for Cook’s D is 4/N – k - 1 (N = number of cases, k = number of variables).

*p < .05; **p < .01, ***p < .001.

Discussion

According to a poll by Pew Research Center, conducted in the same year as the Political Survey data analyzed herein, slightly more than eight out of ten registered Republican voters (83%) in the US preferred to deport undocumented immigrants (Daniller, 2019). In comparison, less than one out of three Democratic counterparts (31%) held the same view. A large body of scholarship has attempted to explain this partisan split as “symptomatic of a broader American identity crisis”. It is argued that xenophobia does not necessarily emerge out of economic hardship. Rather, it reflects a rejection of societal changes brought about by, among others, outsiders with foreign values and practices that are seemingly incompatible with those of host societies (Norris & Inglehart, 2019). Also, as the argument goes, identification with and support for anti-immigrant political movements are driven by a perceived decline in social status among the majority (Mutz, 2018), or what Gest et al. (2018) call “nostalgic deprivation.” While these and other explanations offer valuable insights, the vast majority fall short of considering the important role of broader cultural context, i.e., social norms. The current research was designed to help bridge this gap in the literature.

Using population-based data on American citizens, this study demonstrated a direct, or an independent, connection between cultural tightness (strict social norms) and negative attitudes toward unauthorized immigrants. To empirically validate this contextual effect, a series of multilevel models was estimated conditioning on individual- and state-level confounders. Net of socioeconomic, demographic, macroeconomic, and other controls, cultural tightness was found to shape how US natives view ‘illegal’ foreigners in their country, thus highlighting the causal importance of “the role of place” (Frasure-Yokley & Wilcox-Archuleta, 2019). Aside from the main analysis using original sample consisting of all survey participants, a robustness check was conducted based on a subset of respondents who self-identified as being a member of the non-Hispanic white majority. Results from this supplementary investigation confirmed the spatially varying influence of state-level social norms. As previously noted, however, given the fact that a relatively low proportion of the total variance in the outcome is due to group-level differences, readers should exercise caution in interpreting the magnitude or contextual relevance of cultural tightness on individual-level differences in anti-immigrant attitudes.

A further sensitivity analysis was carried out by addressing the issue of influential cases. This was done by including state-level fixed effects, which produced consistent findings in support of this paper’s thesis: ceteris paribus, living in a culturally tighter state with stronger social norms and lower tolerance for deviance is associated with greater xenophobic attitudes toward unauthorized immigrants. As an indirect test of potential mechanisms underlying this relationship, additional models were estimated to check for within-level and cross-level interaction terms (models not shown). No significant findings emerged concerning cross-level interactions between sociodemographic variables and cultural tightness. But a significant within-level term was found at the 95% level between GELF (a measure of ethnic diversity) and cultural tightness for one of the outcomes (Restrictive Immigration). Specifically, the magnitude of the focal association was stronger in states that are more ethnically heterogeneous. In other words, there is partial evidence suggesting that in culturally tighter places with a higher proportion of ethnic minorities, natives exhibit even greater anti-immigrant sentiments—a result consistent with the theoretical notion that stronger social norms connote greater outgroup prejudice.

An implicit consensus in the literature is that anti-immigrant prejudice is “by and large a ‘cultural’ phenomenon” (Choi et al., 2019, p. 16274; see also Alesina & Tabellini, 2022). That is, whether natives welcome or reject people of foreign origins hinges more on symbolic (nonmaterial), rather than realistic (material), considerations. Despite this wide recognition, related findings remain scarce or, at best, fragmented. Researchers, for the most part, have identified cultural differences or distinctions, i.e., “value dissimilarity” (Wolf et al., 2019), between native and non-native groups as the primary mechanism. To the extent that outgroup members are seen as ‘different from us,’ they are more likely shunned or approached with scepticism by the ingroup. As such, the emphasis has been on the relative fit or congruence between domestic and foreign values espoused by natives and immigrants, respectively (Choi et al., 2019; Havermans & Verkuyten, 2021; Moss et al., 2019; Seewann, 2022; Wolf et al., 2019). The purpose of this paper was to advance the scholarship by conceptualizing and operationalizing culture explicitly at the macro level, independent of individual value orientations. To that end, consistent with Hofstede’s work on aggregate-level cultural variation, it relied on the TL scores across regions (Harrington & Gelfand, 2014) that are irreducible to actor-level attributes.

Measuring culture as an integral aspect of the social environment, not at the level of its constituents, makes it possible to examine its impact across contextual units. In this study, the motivating question was whether and how residency in a particular area of the country characterized by a unique degree of cultural tightness relates to anti-immigrant sentiments at the individual level, over and above measures of symbolic and realistic threat.¹² Clearly, personal characteristics (e.g., age, gender, income, education, intergroup contact, political affiliation) play a critical role. What the main findings of this study indicate is that, adjusting for such confounders, the tightness index has profound implications for intergroup dynamics. If tighter cultures tend to emerge under conditions of greater ecological uncertainty (Gelfand et al., 2017; Jackson et al., 2020; Roos et al., 2015), then it may be said that immigrant influx contributes to stricter social norms by raising the native population’s level of economic uncertainty and cultural anxiety, real or imagined. That is, there may be an unintended consequence of immigration such that native members of “tight” (i.e., culturally homogeneous) context grow even more anti-outgroup in the increasing presence of foreign-born coresidents, who are deemed disruptive of the existing value system and hence social order (Jackson et al., 2019; van Bavel et al., 2020).

Threats of realistic competition and symbolic concerns, i.e., the twin pillars of heuristic mechanism in extant scholarship, do not breed outgroup prejudice in a cultural vacuum. Rather, they are embedded in the web of prevailing social norms that set and enforce the limits of what is desirable and/or possible. And it is only within this broader context that micro-level (material and nonmaterial) factors interactively influence whether and to what extent immigrants are welcomed (for a related view, see Álvarez-Benjumea & Winter, 2020). Previous studies examining contextual-level predictors have typically focused on “structural conditions” (Ceobanu & Escandell, 2010), such as migrant stock, unemployment, poverty, macroeconomic boom/bust, and the like. Others have investigated national immigration policies, role of the media, or presence of right-wing parties (Heath et al., 2020). Still, others have considered the impact of international relations and historical contingencies (Esses, 2021). While it is thus generally understood that “contexts really matter” (Heath et al., 2020, p. 479), they are rarely gauged from a macro-cultural perspective.

The present paper is one of the very few that places the concept of cultural tightness at the center of analysis with respect to why some Americans are more anti-immigrant. The answer backed by the statistical evidence presented above is that cultural environment is a major, yet largely overlooked, determinant. Because tighter cultures are less open to foreign values and external influences, people embedded in them tend to be less permissive of, among others, outgroups including immigrants (Gelfand et al., 2017; Harrington & Gelfand, 2014). Culturally tight places where “social norms are pervasive, clearly defined, and reliably imposed” (Gelfand et al., 2011, p. 2) can certainly be adaptive in promoting coordination and facilitating cooperation (Roos et al., 2015). On the other hand, such places are conducive to anti-outgroup threat, fear, and hostility. While this thesis was previously highlighted cross-nationally using ecological (aggregate-level) data (Jackson, Van Egmond, et al., 2019), the present study is the first to reveal a comparable pattern based on a microlevel survey of US adults.

Conclusions

It has been observed that the psychological discipline is “witnessing a cultural revolution” (Gelfand, 2019, p. 6522). As a review of cross-national research on prejudice suggests, however, that conclusion may be overstated (see Fiske, 2017). Concerning anti-immigrant attitudes, suffice it to say that a paradigm shift toward culturally rooted analysis seems quite a distance away. Currently, many methodological hurdles stand in the way of shedding proper cultural light on when and why natives feel reluctant or refuse to accept the foreign born as compatriots. Perhaps most urgent is the issue of causal identification, i.e., selection bias. Individuals naturally sort themselves into geographic places of residence based on their own preferences. That is, for example, those whose personality traits and value orientations are more congruent with culturally tight areas are prone to move and stay there. Hence, even after accounting for confounding variables, inferring the precise effect size of culture becomes problematic. A comprehensive examination of large-scale surveys finds that when it comes to intergroup contact, what matters is not personal but vicarious experience. Specifically, the context-level contact reduces individual-level prejudice via tolerant norms that encourage intergroup interaction in the first place, implying that people are influenced by known or generalized others’ public behavior (Christ et al., 2014).

This argument can be extended to conjecture that residents of a culturally tighter state may be more xenophobic—in addition to idiosyncratic reasons—because of how their socially meaningful others think and act toward immigrants. Absent experimental research designs, future studies should provide better causal identification by ideally using panel data, with a particular emphasis on the role of “social influence” (Cialdini & Goldstein, 2004) in facilitating, or hampering, the diffusion of anti-immigrant sentiments/preferences as well as other forms of outgroup prejudice. According to one source, over time cultures have become less tight or more open across societies, including the US. Yet, “the loosening of America” (Jackson, Gelfand, et al., 2019) should not be taken as a given. We need not search hard to confirm the continuing significance of racial prejudice and discrimination (unequal treatment ethnic minorities) throughout both American history and contemporary US. Irrespective of cultural trends unfolding across the globe, more scholarly efforts are needed to probe the powerful role of cultural milieu (i.e., the strength of social norms) in shaping outgroup-ingroup dynamics and concomitant outcomes.

Footnotes

Declaration of conflicting interests

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

Funding

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

ORCID iD

Harris Hyun-soo Kim

Notes

Appendix

Appendix 1.

Distribution of cook’s D values for xenophobic sentiment. Note: Four influential cases (states) are shown in boldface.

Appendix 2.

Distribution of cook’s D values for restrictive immigration. Note: Three influential cases (states) are shown in boldface.

Author Biography

Harris Hyun-Soo Kim is Professor of Sociology at Ewha Womans University in Seoul, Republic of Korea. His primary research interests are immigration and immigrants, social capital, intergroup relations, and political participation & civic engagement.

References

Abascal

(2020). Contraction as a response to group threat: Demographic decline and whites’ classification of people who are ambiguously white. American Sociological Review, 85(2), 298–322. https://doi.org/10.1177/0003122420905127

Aktas

Gelfand

M. J.

Hanges

P. J.

(2016). Cultural tightness–looseness and perceptions of effective leadership. Journal of Cross-Cultural Psychology, 47(2), 294–309. https://doi.org/10.1177/0022022115606802

Alesina

Tabellini

(2022). The political effects of immigration: Culture or economics? NBER working. Paper No. 30079.

Álvarez-Benjumea

(2022). Uncovering hidden opinions: Social norms and the expression of xenophobic attitudes. European Sociological Review, 39(3), 449–463. https://doi.org/10.1093/esr/jcac056

Álvarez-Benjumea

Winter

(2020). The breakdown of antiracist norms: A natural experiment on hate speech after terrorist attacks. Proceedings of the National Academy of Sciences of the United States of America, 117(37), 22800–22804. https://doi.org/10.1073/pnas.2007977117

Baldwin

Mussweiler

(2018). The culture of social comparison. Proceedings of the National Academy of Sciences of the United States of America, 115(39), Article E9067–E9074. https://doi.org/10.1073/pnas.1721555115

Bavel

J. J. V.

Baicker

Boggio

P. S.

Capraro

Cichocka

Cikara

Crockett

M. J.

Crum

A. J.

Douglas

K. M.

Druckman

J. N.

Drury

Dube

Ellemers

Finkel

E. J.

Fowler

J. H.

Gelfand

Han

Haslam

S. A.

Jetten

Willer

(2020). Using social and behavioural science to support COVID-19 pandemic response. Nature Human Behaviour, 4(5), 460–471. https://doi.org/10.1038/s41562-020-0884-z

Ben-Nun Bloom

Arikan

Lahav

(2015). The effect of perceived cultural and material threats on ethnic preferences in immigration attitudes. Ethnic and Racial Studies, 38(10), 1760–1778. https://doi.org/10.1080/01419870.2015.1015581

Boin

Rupar

Graf

Neji

Spiegler

Swart

(2021). The generalization of intergroup contact effects: Emerging research, policy relevance, and future directions. Journal of Social Issues, 77(1), 105–131. https://doi.org/10.1111/josi.12419

10.

Bossert

d'Ambrosio

La Ferrara

(2011). A generalized index of fractionalization. Economica, 78(312), 723–750. https://doi.org/10.1111/j.1468-0335.2010.00844.x

11.

Boyer

Firat

van Leeuwen

(2015). Safety, threat, and stress in intergroup relations: A coalitional index model. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 10(4), 434–450. https://doi.org/10.1177/1745691615583133

12.

Breen

Karlson

K. B.

Holm

(2018). Interpreting and understanding logits, probits, and other nonlinear probability models. Annual Review of Sociology, 44(1), 39–54. https://doi.org/10.1146/annurev-soc-073117-041429

13.

Ceobanu

A. M.

Escandell

(2010). Comparative analyses of public attitudes toward immigrants and immigration using multinational survey data: A review of theories and research. Annual Review of Sociology, 36, 309–328.

14.

Chen

Lai

(2020). Where you are is who you are? The geographical account of psychological phenomena. Frontiers in Psychology, 11, 536. https://doi.org/10.3389/fpsyg.2020.00536

15.

Choi

D. D.

Poertner

Sambanis

(2019). Parochialism, social norms, and discrimination against immigrants. Proceedings of the National Academy of Sciences of the United States of America, 116(33), 16274–16279. https://doi.org/10.1073/pnas.1820146116

16.

Christ

Schmid

Lolliot

Swart

Stolle

Tausch

Al Ramiah

Wagner

Vertovec

Hewstone

(2014). Contextual effect of positive intergroup contact on outgroup prejudice. Proceedings of the National Academy of Sciences of the United States of America, 111(11), 3996–4000. https://doi.org/10.1073/pnas.1320901111

17.

Chua

R. Y.

Huang

K. G.

Jin

(2019). Mapping cultural tightness and its links to innovation, urbanization, and happiness across 31 provinces in China. Proceedings of the National Academy of Sciences of the United States of America, 116(14), 6720–6725. https://doi.org/10.1073/pnas.1815723116

18.

Cialdini

R. B.

Goldstein

N. J.

(2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55, 591–621. https://doi.org/10.1146/annurev.psych.55.090902.142015

19.

Daniller

(2019). Americans’ immigration policy priorities: Divisions between – and within – the two parties. Pew Research Center. https://www.pewresearch.org/fact-tank/2019/11/12/americans-immigration-policy-priorities-divisions-between-and-within-the-two-parties/

20.

Davidov

Seddig

Gorodzeisky

Raijman

Schmidt

Semyonov

(2020). Direct and indirect predictors of opposition to immigration in europe: Individual values, cultural values, and symbolic threat. Journal of Ethnic and Migration Studies, 46(3), 553–573. https://doi.org/10.1080/1369183X.2018.1550152

21.

Deckert

Schomaker

R. M.

(2022). Cultural tightness–looseness and national innovativeness: Impacts of tolerance and diversity of opinion. Journal of Innovation and Entrepreneurship, 11(1), 19–29. https://doi.org/10.1186/s13731-022-00219-2

22.

Dennison

Geddes

(2021). Thinking globally about attitudes to immigration: Concerns about social conflict, economic competition and cultural threat. The Political Quarterly, 92(3), 541–551. https://doi.org/10.1111/1467-923X.13013

23.

Di Santo

Gelfand

M. J.

Baldner

Pierro

(2022). The moral foundations of desired cultural tightness. Frontiers in Psychology, 13, 739579. https://doi.org/10.3389/fpsyg.2022.739579

24.

Eisentraut

(2019). Explaining attitudes toward minority groups with human values in Germany-What is the direction of causality? Social Science Research, 84, 102324. https://doi.org/10.1016/j.ssresearch.2019.06.015

25.

Elff

Heisig

J. P.

Schaeffer

Shikano

(2021). Multilevel analysis with few clusters: Improving likelihood-based methods to provide unbiased estimates and accurate inference. British Journal of Political Science, 51(1), 412–426. https://doi.org/10.1017/S0007123419000097

26.

Esses

V. M.

(2021). Prejudice and discrimination toward immigrants. Annual Review of Psychology, 72, 503–531. https://doi.org/10.1146/annurev-psych-080520-102803

27.

Fiske

S. T.

(2017). Prejudices in cultural contexts: Shared stereotypes (gender, age) versus variable stereotypes (race, ethnicity, religion). Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 12(5), 791–799. https://doi.org/10.1177/1745691617708204

28.

Frasure-Yokley

Wilcox-Archuleta

(2019). Geographic identity and attitudes toward undocumented immigrants. Political Research Quarterly, 72(4), 944–959. https://doi.org/10.1177/1065912919843349

29.

Gelfand

M. J.

(2019). Universal and culture-specific patterns of tightness-looseness across the 31 Chinese provinces. Proceedings of the National Academy of Sciences of the United States of America, 116(14), 6522–6524. https://doi.org/10.1073/pnas.1902401116

30.

Gelfand

M. J.

Harrington

J. R.

Jackson

J. C.

(2017). The strength of social norms across human groups. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 12(5), 800–809. https://doi.org/10.1177/1745691617708631

31.

Gelfand

M. J.

Raver

J. L.

Nishii

Leslie

L. M.

Lun

Lim

B. C.

Duan

Almaliach

Ang

Arnadottir

Aycan

Boehnke

Boski

Cabecinhas

Chan

Chhokar

D'Amato

Subirats Ferrer

Fischlmayr

I. C.

Yamaguchi

(2011). Differences between tight and loose cultures: A 33-nation study. Science, 332(6033), 1100–1104. https://doi.org/10.1126/science.1197754

32.

Gest

Reny

Mayer

(2018). Roots of the radical right: Nostalgic deprivation in the United States and Britain. Comparative Political Studies, 51(13), 1694–1719. https://doi.org/10.1177/0010414017720705

33.

Gilliam

Schwartz

D. B.

Godoy

Boduroglu

Gutchess

(2022). Does state tightness-looseness predict behavior and attitudes early in the COVID-19 pandemic in the USA? Journal of Cross-Cultural Psychology, 53(5), 522–542. https://doi.org/10.1177/00220221221077710

34.

Gorodzeisky

Semyonov

(2019). Unwelcome immigrants: Sources of opposition to different immigrant groups among Europeans. Frontiers in Sociology, 4, 24. https://doi.org/10.3389/fsoc.2019.00024

35.

Hainmueller

Hopkins

D. J.

(2015). The hidden American immigration consensus: A conjoint analysis of attitudes toward immigrants. American Journal of Political Science, 59(3), 529–548. https://doi.org/10.1111/ajps.12138

36.

Harrington

J. R.

Gelfand

M. J.

(2014). Tightness–looseness across the 50 United States. Proceedings of the National Academy of Sciences of the United States of America, 111(22), 7990–7995. https://doi.org/10.1073/pnas.1317937111

37.

Harteveld

Kokkonen

Dahlberg

(2017). Adapting to party lines: The effect of party affiliation on attitudes to immigration. West European Politics, 40(6), 1177–1197. https://doi.org/10.1080/01402382.2017.1328889

38.

Havermans

Verkuyten

(2021). Positive and negative behavioural intentions towards immigrants: A question of ethnic categorisation or worldview conflict? International Journal of Psychology: Journal International de Psychologie, 56(5), 633–641. https://doi.org/10.1002/ijop.12748

39.

Heath

Davidov

Ford

Green

E. G. T.

Ramos

Schmidt

(2020). Contested terrain: Explaining divergent patterns of public opinion towards immigration within Europe. Journal of Ethnic and Migration Studies, 46(3), 475–488. https://doi.org/10.1080/1369183X.2019.1550145

40.

Helbling

Traunmüller

(2016). How state support of religion shapes attitudes toward Muslim immigrants: New evidence from a sub-national comparison. Comparative Political Studies, 49(3), 391–424. https://doi.org/10.1177/0010414015612388

41.

Hochschild

A. R.

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

42.

Hövermann

Messner

S. F.

(2019). Marketization and anti-immigrant attitudes in cross-national perspective. Social Science Research, 84, 102326. https://doi.org/10.1016/j.ssresearch.2019.06.017

43.

Igarashi

Nagayoshi

(2022). Norms to be prejudiced: List experiments on attitudes towards immigrants in Japan. Social Science Research, 102, 102647. https://doi.org/10.1016/j.ssresearch.2021.102647

44.

Jackson

J. C.

Gelfand

Fox

(2019). The loosening of American culture over 200 years is associated with a creativity–order trade-off. Nature Human Behaviour, 3(3), 244–250. https://doi.org/10.1038/s41562-018-0516-z

45.

Jackson

J. C.

Gelfand

Ember

C. R.

(2020). A global analysis of cultural tightness in non-industrial societies. Proceedings. Biological sciences, 287(1930), 20201036. https://doi.org/10.1098/rspb.2020.1036

46.

Jackson

J. C.

Van Egmond

Choi

V. K.

Ember

C. R.

Halberstadt

Balanovic

Basker

I. N.

Boehnke

Buki

Fischer

Fulop

Fulmer

Homan

A. C.

van Kleef

G. A.

Kreemers

Schei

Szabo

Ward

Gelfand

M. J.

Gelfand

M. J.

(2019). Ecological and cultural factors underlying the global distribution of prejudice. PLoS One, 14(9), Article e0221953. https://doi.org/10.1371/journal.pone.0221953

47.

Jordan

Gebeloff

(2022). Amid slowdown, immigration is driving U.S. Population growth. The New York Times. https://www.nytimes.com/2022/02/05/us/immigration-census-population.html

48.

Kaufmann

Goodwin

M. J.

(2018). The diversity wave: A meta-analysis of the native-born white response to ethnic diversity. Social Science Research, 76, 120–131. https://doi.org/10.1016/j.ssresearch.2018.07.008

49.

Lancee

Pardos-Prado

(2013). Group conflict theory in a longitudinal perspective: Analyzing the dynamic side of ethnic competition. International Migration Review, 47(1), 106–131. https://doi.org/10.1111/imre.12015

50.

Leong

C. H.

Ward

(2006). Cultural values and attitudes toward immigrants and multiculturalism: The case of the Eurobarometer survey on racism and xenophobia. International Journal of Intercultural Relations, 30(6), 799–810. https://doi.org/10.1016/j.ijintrel.2006.07.001

51.

Lopez

Devos

Somo

(2022). State-level cultural tightness–looseness accounts for implicit associations between American and White identities. Current Research in Ecological and Social Psychology, 3, 100033. https://doi.org/10.1016/j.cresp.2021.100033

52.

Maas

C. J. M.

Hox

J. J.

(2004). Robustness issues in multilevel regression analysis. Statistica Neerlandica, 58(2), 127–137. https://doi.org/10.1046/j.0039-0402.2003.00252.x

53.

Maxwell

(2020). Geographic divides and cosmopolitanism: Evidence from Switzerland. Comparative Political Studies, 53(13), 2061–2090. https://doi.org/10.1177/0010414020912289

54.

McAuliffe

Triandafyllidou

(eds), (2021) World migration report 2022. International Organization for Migration (IOM).

55.

Meleady

Seger

C. R.

Vermue

(2017). Examining the role of positive and negative intergroup contact and anti‐immigrant prejudice in Brexit. British Journal of Social Psychology, 56(4), 799–808. https://doi.org/10.1111/bjso.12203

56.

Meuleman

Abts

Schmidt

Pettigrew

T. F.

Davidov

(2020). Economic conditions, group relative deprivation and ethnic threat perceptions: A cross-national perspective. Journal of Ethnic and Migration Studies, 46(3), 593–611. https://doi.org/10.1080/1369183X.2018.1550157

57.

Migration Policy Institute (n.d.) . Migration Data Hub. https://www.migrationpolicy.org/programs/migration-data-hub.Retrieved from.

58.

Mood

(2010). Logistic regression: Why we cannot do what we think we can do, and what we can do about it. European Sociological Review, 26(1), 67–82. https://doi.org/10.1093/esr/jcp006

59.

Moss

A. J.

Blodorn

Van Camp

A. R.

O’Brien

L. T.

(2019). Gender equality, value violations, and prejudice toward Muslims. Group Processes & Intergroup Relations, 22(2), 288–301. https://doi.org/10.1177/1368430217716751

60.

Musca

S. C.

Kamiejski

Nugier

Méot

Er-Rafiy

Brauer

. (2011). Data with hierarchical structure: Impact of intraclass correlation and sample size on type-I error. Frontiers in Psychology, 2, 74. https://doi.org/10.3389/fpsyg.2011.00074

61.

Mutz

D. C.

(2018). Status threat, not economic hardship, explains the 2016 presidential vote. Proceedings of the National Academy of Sciences of the United States of America, 115(19), Article E4330–E4339. https://doi.org/10.1073/pnas.1718155115

62.

Newman

B. J.

Hartman

T. K.

Taber

C. S.

(2012). Foreign language exposure, cultural threat, and opposition to immigration. Political Psychology, 33(5), 635–657. https://doi.org/10.1111/j.1467-9221.2012.00904.x

63.

Norris

Inglehart

(2019). Cultural backlash: Trump. Brexit, and authoritarian populism. Cambridge University Press.

64.

Pepinsky

T. B.

(2018). A note on listwise deletion versus multiple imputation. Political Analysis, 26(4), 480–488. https://doi.org/10.1017/pan.2018.18

65.

Pettigrew

T. F.

Christ

Wagner

Meertens

R. W.

Van Dick

Zick

(2008). Relative deprivation and intergroup prejudice. Journal of Social Issues, 64(2), 385–401. https://doi.org/10.1111/j.1540-4560.2008.00567.x

66.

Pettigrew

T. F.

Tropp

L. R.

(2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751–783. https://doi.org/10.1037/0022-3514.90.5.751

67.

Pew Research Center . (2018). Political survey 2018 (conducted on sept. 18–24, 2018). https://www.pewresearch.org/politics/dataset/september-2018-political-survey/

68.

Quillian

(1995). Prejudice as a response to perceived group threat: Population composition and anti-immigrant and racial prejudice in Europe. American Sociological Review, 60(4), 586–611. https://doi.org/10.2307/2096296

69.

Raudenbush

Anthony

Richard

(2019). HLM 8 for windows [computer software]. Scientific Software International, Inc.

70.

Roos

Gelfand

Nau

Lun

(2015). Societal threat and cultural variation in the strength of social norms: An evolutionary basis. Organizational Behavior and Human Decision Processes, 129, 14–23. https://doi.org/10.1016/j.obhdp.2015.01.003

71.

Saad

(2022). U.S. Immigration views remain mixed and highly partisan. Gallup. https://news.gallup.com/poll/395882/immigration-views-remain-mixed-highly-partisan.aspx

72.

Schmuck

Matthes

(2017). Effects of economic and symbolic threat appeals in right-wing populist advertising on anti-immigrant attitudes: The impact of textual and visual appeals. Political Communication, 34(4), 607–626. https://doi.org/10.1080/10584609.2017.1316807

73.

See

Y. H. M.

Lim

A. W. Q.

Pauketat

J. V. T.

(2020). Values predict willingness to interact with immigrants: The role of cultural ideology and multicultural acquisition. Journal of Cross-Cultural Psychology, 51(1), 3–24. https://doi.org/10.1177/0022022119883018

74.

Seewann

(2022). My values, their values: How value conceptualisations influence attitudes towards immigration. Journal of Ethnic and Migration Studies, 48(9), 2091–2114. https://doi.org/10.1080/1369183X.2021.2004100

75.

Shin

Dovidio

J. F.

(2018). Differences, threats, values, and country‐specific prejudice toward immigrants and foreign workers in three major receiving countries: The United States, Germany, and Australia. Journal of Social Issues, 74(4), 737–755. https://doi.org/10.1111/josi.12296

76.

Stamkou

van Kleef

G. A.

Homan

A. C.

Gelfand

M. J.

van de Vijver

F. J.

van Egmond

M. C.

Boer

Phiri

Ayub

Kinias

Cantarero

Efrat Treister

Figueiredo

Hashimoto

Hofmann

E. B.

Lima

R. P.

Lee

I. C.

Lee

I. C.

(2019). Cultural collectivism and tightness moderate responses to norm violators: Effects on power perception, moral emotions, and leader support. Personality and Social Psychology Bulletin, 45(6), 947–964. https://doi.org/10.1177/0146167218802832

77.

Valentino

N. A.

Soroka

S. N.

Iyengar

Aalberg

Duch

Fraile

Hahn

K. S.

Hansen

K. M.

Harell

Helbling

Jackman

S. D.

Kobayashi

(2019). Economic and cultural drivers of immigrant support worldwide. British Journal of Political Science, 49(4), 1201–1226. https://doi.org/10.1017/S000712341700031X

78.

Van der Meer

Te Grotenhuis

Pelzer

(2010). Influential cases in multilevel modeling: A methodological comment. American Sociological Review, 75(1), 173–178. https://doi.org/10.1177/0003122409359166

79.

Visintin

E. P.

Green

E. G.

Falomir-Pichastor

J. M.

Berent

(2020). Intergroup contact moderates the influence of social norms on prejudice. Group Processes and Intergroup Relations, 23(3), 418–440. https://doi.org/10.1177/1368430219839485

80.

Wolf

L. J.

Weinstein

Maio

G. R.

(2019). Anti-immigrant prejudice: Understanding the roles of (perceived) values and value dissimilarity. Journal of Personality and Social Psychology, 117(5), 925–953. https://doi.org/10.1037/pspi0000177

81.

Wollast

Lüders

Nugier

Khamzina

de la Sablonnière

Guimond

(2022). Social dominance and anti‐immigrant prejudice: A cross‐national and prospective test of the mediating role of assimilation, multiculturalism, colour blindness, and interculturalism. European Journal of Social Psychology, 53(1), 167–182. https://doi.org/10.1002/ejsp.2895