Religion, Partisanship, and the Boundaries of Inclusion: Explaining Pro-Immigration Attitudes in the United States ()
1. Introduction
Individuals’ religious and racial group identities, political and educational factors and the values associated with those statuses strongly shape whether they perceive immigrants to be a threat or a benefit to U. S. society. While prior U.S. research largely focused on anti-immigrant attitudes among those who are in dominant racial groups or religiously fundamentalist through threat-based mechanisms (Jardina, 2019; Woods & Arthur, 2014), this study shifts the focus to the drivers of pro-immigration attitudes rather than opposition. As immigration is shaping the nation’s demographic, economic, and cultural landscape, understanding the drivers of public opinion is essential for evaluating the legitimacy and sustainability of immigration policy. In recent decades, immigration has emerged as one of the most polarizing issues in American politics, dividing individuals not only along partisan lines but also across socio-demographic, racial, and religious boundaries (Haubert & Fussell, 2006; Kaba, 2019).
This study investigates how socio-demographic, political, and religious factors shape perceptions of immigrants’ impacts on U.S. society. Specifically, the study examines whether religious orientation moderates the relationship between party identification and immigration attitudes in the United States. Additionally, it hypothesizes that individuals with more cosmopolitan orientations, characterized by younger age, more liberal political views, and higher levels of educational attainment, are more likely to express favorable attitudes toward immigration, as they may perceive immigrants as part of a shared in-group or as holding similar sociocultural status (Afrin, 2026; Citrin & Sides, 2008). This study uses data from the 2022 General Social Survey (Davern et al., 2024) to establish the theoretical framework by testing the hypothesis of whether religious orientation matters for party identification and how that influences immigration attitudes. In doing so, it shifts the focus from simple predictors of exclusion to a deeper inquiry into the drivers of pro-immigrant sentiment, and how identity-based belief systems shape the relationship between religiosity and inclusive political views.
The next section reviews existing literature that explains variations in perceptions of immigrants through the lenses of social identity theory, group threat theory and modernization theory, which have been central to understanding such attitudes in the U.S. context. The following section describes the sample, variables and analytical methods used in the study. This is followed by a presentation of the results from a series of Ordinary Least Squares (OLS) regression models predicting immigration attitudes, conducted using Stata software. The final section discusses the key findings.
2. Theoretical Framework
2.1. Group Threat Theory: Status Anxiety and Exclusion
Understanding public attitudes toward immigration is very important in the contemporary political and social debates in the United States. Immigration attitudes and policies are shaped not only by economic and demographic considerations but also by public opinion, which in turn affects electoral outcomes, policy legitimacy, and the social integration of immigrants (Burstein, 2003; Citrin & Wright, 2013; Hainmueller & Hopkins, 2014). In the United States, immigration remains a polarizing issue, and public opinion drives huge insights for policymakers, scholars, and civic leaders seeking to navigate an increasingly diverse and ideologically divided society.
Group Threat Theory (Blalock, 1967; Bobo, 1983; Bobo & Hutchings, 1996) posits that dominant groups adopt exclusionary attitudes toward immigrants when they perceive threats to their economic, political, or cultural status. Immigration, in this view, heightens anxieties about competition over scarce resources and the erosion of group dominance. This perspective has been widely applied to explain restrictive immigration attitudes among majority populations in the United States and Europe (Chandler & Tsai, 2001; Quillian, 1995; Wilson, 2001).
In the American context, white Americans, who feel that demographic change is a threat to their group status, are more likely to support strict immigration policies (Outten et al., 2012; Mutz, 2018). As the nation becomes more racially and ethnically diverse, some white individuals experience a sense of threat and relative deprivation, fearing that increasing diversity will undermine their group’s culture and political dominance. Craig and Richeson (2014) demonstrate that reminders of impending racial demographic shifts increase exclusionary immigration attitudes among whites, reinforcing the idea that perceptions of relative deprivation drive hostility. These dynamics are especially pronounced among older individuals, who may have been socialized in less diverse contexts and thus experience immigration as a more profound disruption (Semyonov et al., 2006; Stephan & Stephan, 2000).
Religiosity can also be understood through group threat dynamics. Bloom et al. (2015) show that religious social identity plays a more crucial role than doctrinal belief in shaping anti-immigration attitudes. Their findings suggest that individuals who strongly identify with a religious group are more likely to express exclusionary views toward immigrants, particularly when immigrants are perceived as religiously dissimilar. Comparative studies of global fundamentalist movements emphasize that these groups rely on absolute moral truths and clear distinctions between the faithful and the secular world, reinforcing a protective stance toward perceived cultural threats (Almond et al., 2003; Marty & Appleby, 1991). They often emphasize cultural preservation, moral strength, and group purity (Knoll, 2009; Layman & Carmines, 1997). These orientations heighten perceptions of immigrants as cultural outsiders who threaten the cohesion of the dominant moral order. Thus, white, older, conservative, and religiously fundamentalist individuals are expected to be least supportive of immigration.
2.2. Modernization Theory: Education and Cosmopolitanism
Modernization theory was widely used in social science, especially in the post-World War II period. This theory was widely used to understand how developing nations could become economically and socially developed, often looking at Western industrialized countries as models. It emphasizes the liberalizing role of education and broader societal modernization. Education fosters cognitive complexity, tolerance, and cosmopolitan values, enabling individuals to interpret immigration as enriching rather than threatening (Hainmueller & Hiscox, 2007; Inglehart, 1997; Jardina, 2019). Research consistently shows that higher levels of educational attainment are associated with greater support for immigration, as education reduces susceptibility to ethnocentric fears and increases openness to diversity (Ceobanu & Escandell, 2010; Citrin & Wright, 2013).
Empirical findings support this perspective across multiple contexts. Educated individuals are more likely to view immigrants as economic contributors and less likely to accept narratives of cultural threat (Citrin & Wright, 2013; Hainmueller & Hiscox, 2007). Education also broadens social networks, exposing individuals to diverse peers and multicultural environments, which weakens rigid in-group preferences and promotes inclusive identities (Coenders & Scheepers, 2003).
Age and political ideology intersect here as well: younger and more politically liberal individuals tend to embrace cosmopolitan worldviews, reinforcing education’s liberalizing effect. Several studies find that older individuals are more likely to support restrictive immigration policies, while younger individuals tend to exhibit more inclusive attitudes (Hainmueller & Hopkins, 2014; Semyonov et al., 2006). Older citizens may be more sensitive about their cultural and economic threats, especially given their socialization during periods of less ethnic diversity. Thus, they identify out-groups as threatening when they believe their group’s social or economic position is at risk (Stephan & Stephan, 2000; Tajfel & Turner, 1979). Simultaneously, conservatives tend to express more restrictive immigration views, whereas liberals are more supportive of inclusive policies (Chandler & Tsai, 2001; Tesler, 2020).
2.3. Social Identity Theory: In-Group Boundaries and Out-Group Solidarity
Social Identity Theory (Tajfel & Turner, 1979; Hornsey, 2008) emphasizes that individuals derive self-concept from group memberships. These identities shape attitudes toward migrants depending on how group boundaries are drawn (Masuoka & Junn, 2022). Extending this identity-based approach, Davis (2019) introduces the concept of Christian nationalism, defined as an ideological framework that seeks to align national identity with a particular religious tradition. He shows that adherence to Christian nationalist beliefs is strongly associated with political behaviors and policy preferences that prioritize cultural homogeneity and reinforce symbolic boundaries between insiders and outsiders. Recent work further shows that Christian nationalism is associated with exclusionary attitudes toward immigrants even among the religiously inactive, suggesting that such views are embedded in broader cultural identity rather than religious practice alone (Stroope et al., 2021).
Within this framework, when boundaries are rigid and exclusionary, immigrants are more likely to be perceived as threatening out-groups. In contrast, when boundaries are more flexible and inclusive, immigrants may be seen as allies or as part of a broader, shared social community. This perspective also explains racial variation in immigration attitudes. Minority groups, particularly Black Americans, often express more inclusive immigration preferences than whites (Haubert & Fussell, 2006; Carter & Pérez, 2016; Masuoka & Junn, 2022). Their historical experiences of marginalization and exclusion can foster a sense of solidarity with immigrants, consistent with social identity theory’s expectation of cross-group alliances among disadvantaged groups.
Political party identification operates as a social identity that structures individuals’ perceptions of in-groups and out-groups, thereby shaping attitudes toward immigration (Tajfel & Turner, 1979). Rather than reflecting only policy preferences or ideological positions, partisanship functions as a durable group attachment that structures perceptions of different groups, shaping attitudes toward politically relevant social categories, including immigrants (Layman & Green, 2006). In recent decades, partisan divisions over immigration have intensified, with Republicans generally expressing more restrictive attitudes and Democrats exhibiting more supportive positions (Hainmueller & Hopkins, 2014; Jardina, 2019). Similarly, party identification has increasingly taken on the characteristics of a social identity, producing affective polarization in which individuals evaluate social issues and groups through partisan lenses (Iyengar et al., 2012).
However, partisanship alone does not fully explain variation in immigration attitudes. Individuals within the same party may hold divergent views depending on how they interpret social boundaries and moral obligations. Here, religious orientation plays a critical role. Rather than functioning solely as a demographic characteristic, religious orientation reflects a moral and symbolic framework that shapes how individuals define in-groups and out-groups. More liberal religious groups, such as Jewish and mainline Protestant communities, have historically supported immigrant rights, including refugee resettlement and opposition to restrictive policies (Alba et al., 2009; Sherkat, 2011). In contrast, religious fundamentalists prioritize in-group cohesion and moral boundaries, often resulting in exclusionary immigration attitudes (Knoll, 2009; Layman & Carmines, 1997).
2.4. Integrating the Frameworks
Taken together, these theoretical perspectives suggest that immigration attitudes are shaped by distinct yet interrelated mechanisms rather than a single unified process. Group threat theory primarily accounts for opposition to immigration by emphasizing how perceived economic, cultural, and political threats generate exclusionary preferences among dominant groups. In contrast, modernization theory highlights the emergence of supportive attitudes, arguing that education, generational change, and exposure to diversity foster cosmopolitan values that encourage more favorable views toward immigrants. However, these liberalizing effects are not uniform across individuals.
Social identity theory provides a critical bridge between these perspectives by explaining how group-based identities condition the translation of these underlying mechanisms into concrete attitudes. It suggests that individuals interpret immigration through the lens of their social identity and that the flexibility or rigidity of group boundaries determines whether immigrants are perceived as risky outsiders or incorporated into an expanded in-group. As a result, the same structural factors, such as education or political ideology, may produce different attitudinal outcomes depending on how individuals define their social and symbolic boundaries.
Building on this integrated framework, this study advances the key arguments that religious orientation moderates how partisan identities are translated into immigration attitudes. Rather than exerting only a direct effect, religious worldviews shape the moral and symbolic boundaries through which individuals interpret political issues, thereby influencing whether partisan attachments produce exclusionary or more inclusive preferences. In particular, more liberal religious orientations may soften rigid group boundaries and attenuate partisan polarization, whereas religiously fundamentalist may reinforce boundary maintenance and amplify negative attitudes towards immigration. Using recent data from the 2022 General Social Survey, this study offers a more nuanced and interaction-based understanding of public opinion on immigration in the contemporary United States.
3. Hypotheses
Building on prior empirical and theoretical work on immigration attitudes, this study advances two testable hypotheses. H1: Religious orientation moderates the relationship between political party identification and immigration attitudes. Liberal religious orientations are associated with more positive immigration attitudes among the political parties. This expectation follows from literature suggesting that partisanship is a strong predictor of immigration attitudes, but that its effects may be conditioned by deeper belief systems, particularly religious worldviews. Religious orientations vary in their openness to cultural diversity and boundary drawing, which may amplify or attenuate partisan differences in immigration attitudes.
The second hypothesis draws on modernization theory and the literature on cosmopolitanism, which emphasizes the liberalizing effects of education, generational change, and political ideology. H2: Individuals with more cosmopolitan orientations—indicated by younger age, more liberal political views, and higher levels of educational attainment—will express more favorable attitudes toward immigration.
4. Data and Measures
For this study, General Social Survey (Davern et al., 2024) weights and survey design data have been utilized. The GSS, conducted by the National Opinion Research Center (NORC) at the University of Chicago since 1972 (with some exceptions), and it is one of the most respected and long-standing national surveys in the United States. It collects data on a broad range of topics including demographics, political and social attitudes, religious beliefs, and opinions from a nationally representative sample of adults living in U.S. households. The data is widely used by researchers across many disciplines including sociology, political science, psychology, and economics. The GSS data is publicly accessible and can be downloaded through the GSS data explorer portal at https://gssdataexplorer.norc.org.
After excluding seven observations (0.2% of the original sample) with missing or non-response values on one or more variables included in the analysis, the final analytic sample consisted of 3346 respondents (n = 3346). This cleaned dataset includes only those respondents who provided complete data on the dependent and independent variables, allowing for valid statistical inference in the multivariate analyses.
This study draws its variables from two primary categories of the General Social Survey (Davern et al., 2024): the replicating core and the topical modules. The dependent variable in this analysis is attitude toward immigration (letin1a), which measures whether respondents believe the number of immigrants entering the U.S. should be increased, decreased, or kept the same. Respondents were given five response options ranging from (5) “Decrease a lot” to (1) “Increase a lot”. To ensure consistent interpretation, the measure was reverse-coded (letin1a2), where higher values indicate stronger support for increasing immigration. The key independent variables in the analysis are party identification and religious orientation. To account for potential confounding factors, the analysis includes age, education, race, sex, nativity, political ideology, and religious affiliation as controls. This structure enables a focused analysis of the effects of political identification and religious orientation on immigration attitudes, while controlling for the denominational background on effects.
In the General Social Survey (GSS), the key independent variable partyid is measured on a seven-category scale based on the question: “What is your political party identification?” This variable was recoded into four categories: 1) Democrat, 2) Independent (neither party), 3) Republican, and 4) Other party. Additionally, religious orientation is measured using the question: How fundamentalist are respondents currently, where religious orientation is categorized into three groups: 1) Fundamentalist, 2) Moderate, and 3) Liberal.
Religious affiliation was determined by asking the question: What is your religious preference? The original responses were recoded into five groups: 1) Protestant, 2) Catholic, 3) Jewish, 4) Christian (Protestant and Catholic), and 5) None (including those reporting no religious affiliation). Age was assessed through the standard GSS item that records respondents’ age at the time of the survey, with options from 18 years to 89 years or older. Additionally, education is measured using respondents’ reported highest year of school completed. Education ranges from (0) “Less than High School” to (20) “Doctorate or equivalent”. Political Ideology is captured using the GSS question: “We hear a lot of talk these days about liberals and conservatives. Where would you place yourself on this scale?” Responses range from (1) “Extremely liberal” to (7) “Extremely conservative” with higher values indicating more conservative political ideology. Race is assessed with the question: “What race do you consider yourself?” Respondents were categorized into three groups: 1) White, 2) Black, and 3) Other (including Asian, Native American, and multiracial respondents), and sex is coded as 1 (male) and 2 (female).
Nativity was measured based on the GSS question “Was respondent (R) born in this country?” Responses were coded as 1 (yes, U.S.-born) and 0 (no, foreign-born). This variable distinguishes between individuals born in the United States and those born abroad. To examine potential moderating effects, an interaction term between political party identification and religious orientation was created. This interaction assesses whether the effect of religiosity on immigration attitudes varies depending on the respondent’s political party identification. In this study, both bivariate and multivariate regression analysis techniques have been employed to inspect the relationships between the variables. All the analyses were conducted using Stata 18.5 software, following comprehensive data cleaning procedures, including addressing missing data and reverse-coding. As a robustness check, I also estimated an ordered logit model because the dependent variable is ordinal. The ordered logit model produced substantively identical conclusions to the OLS model.
5. Results and Discussion
At the initial stage of the analysis, descriptive properties of all the studied variables have been explored. Table 1 presents the descriptive statistics (i.e., mean, standard deviation, and minimum-maximum values) for the variables in this study. The average score for immigration attitudes among respondents is 2.86 (SD = 1.20), these suggests, on average, individuals lean slightly toward decreasing immigration rather than increasing it.
Among the key predictors, the average age of respondents is approximately 48.77 years (SD = 17.79), reflecting a diverse age distribution ranging from young adults to elderly individuals. The average education level is 14.19 years (SD = 2.87), indicating that most respondents completed some college education or held a bachelor’s degree. Regarding political ideology, the mean score is 3.97 (SD = 1.52), suggesting that, on average, respondents identify as moderate, slightly leaning toward conservative positions (where higher scores represent more conservative views). For party identification, 44.35% (n = 1484) of respondents identified as Democrats, 20.95% (n = 701) as Independents, 32.00% (n = 1072) as Republicans, and 2.66% (n = 89) as affiliating with other political parties. This distribution indicates substantial partisan diversity within the sample, providing an appropriate basis for examining partisan variation in immigration attitudes.
In terms of racial composition, 68.35% (n = 2287) of the respondents identified as white, 18.05% (n = 604) as black, and 13.60% (n = 455) as belonging to other racial categories (including Asian, Native American, and multiracial). For religious orientation, approximately 17.39% (n = 582) of respondents were categorized as fundamentalists, 41.87% (n = 1401) as moderates, and 40.74% (n = 1363) as liberals in religious orientation. For nativity, about 87% (n = 2911) of respondents reported being born in the United States, while 13% (n = 435) were foreign-born. Similarly, for sex, 46.60% (n = 1559) of respondents identified as male and 53.40% (n = 1787) as female. Finally, for religious affiliation, 41.27% (n = 1381) identified as Protestant, 23% (n = 770) as Catholic, 1.76% (n = 59) as Jewish, 3.23% (n = 108) as Christian (Protestant or Catholic), and 30.72% (n = 1028) reported no religious affiliation.
Table 1. Descriptive statistics.
Variables |
Mean/Percentage |
Standard Deviation |
Min |
Max |
Dependent Variable |
|
|
|
|
Number of Immigrants Nowadays Should Be |
2.86 |
1.20 |
1 |
5 |
Independent Variables |
|
|
|
|
Political Party Identification |
|
|
|
|
Democrat |
44.35% |
- |
- |
- |
Independent |
20.95% |
- |
- |
- |
Republican |
32% |
- |
- |
- |
Other Party |
2.66% |
- |
- |
- |
Religiosity (Fund) |
|
|
|
|
Fundamentalist |
17.39% |
- |
- |
- |
Moderate |
41.87% |
- |
- |
- |
Liberal |
40.74% |
- |
- |
- |
Control Variables |
|
|
|
|
Age |
48.80 |
17.77 |
18 |
89 |
Education |
14.19 |
2.87 |
0 |
20 |
Political Ideology |
3.97 |
1.52 |
1 |
7 |
Sex |
|
|
|
|
Male |
46.60% |
|
|
|
Female |
53.40% |
|
|
|
Race |
|
|
|
|
White |
68.35% |
- |
- |
- |
Black |
18.05% |
- |
- |
- |
Other |
13.60% |
- |
- |
- |
Nativity (Born in U.S) |
|
|
|
|
Yes |
87% |
|
|
|
No |
13% |
|
|
|
Religious Affiliation |
|
|
|
|
Protestant |
41.27% |
- |
- |
- |
Catholic |
23% |
- |
- |
- |
Jewish |
1.76% |
- |
- |
- |
Christian |
3.23% |
- |
- |
- |
None |
30.72% |
- |
- |
- |
n = 3346; Dependent variable: Immigration Attitude Coded as 5 = Increase a lot, 1 = Decrease a lot; Political Ideology Coded as 1 = Extremely liberal to 7 = Extremely conservative.
After the initial exploration of the data and examination of the descriptive statistics, a series of regression analyses were conducted to evaluate all the hypotheses using Ordinary Least Squares (OLS) Regression in STATA. Primarily, a bivariate regression (Model 1) was developed to explore the relationship between the focal predictor (Immigration Attitudes) and respondents’ party identification. In the second phase, a multivariate regression analysis was conducted to assess the independent associations between all predictors and the dependent variable (Model 2).
In the third and final stage, the potential moderating effect of religious orientation on the relationship between immigration attitudes (the dependent variable) and respondents’ party identification was tested using Model 3. This model included an interaction term (i.e., Party identification × Religiosity) alongside the other nine predictors to provide insight into the anticipated moderation dynamic (Table 2).
Table 2. Multiple regression results.
Variables |
Model 1 |
Model 2 |
Model 3 |
b (S.E) |
b (S.E) |
b (S.E) |
Independent Variables |
|
|
|
Political Party Identification (Ref: Democrat) |
|
|
|
Independent |
−0.380*** (0.051) |
−0.239*** (0.052) |
−0.021 (0.389) |
Republican |
−1.060*** (0.045) |
−0.567*** (0.054) |
−0.498*** (0.106) |
Other Party |
−0.182 (0.121) |
−0.121 (0.116) |
−0.669* (0.341) |
Religiosity (Ref: Fundamentalist) |
|
|
|
Moderate |
|
0.133* (0.063) |
0.157 (0.093) |
Liberal |
|
0.118 (0.079) |
0.268** (0.107) |
Control Variables |
|
|
|
Age |
|
−0.010*** (0.001) |
−0.010*** (0.001) |
Education |
|
0.030*** (0.007) |
0.029*** (0.006) |
Political Ideology (polviews) |
|
−0.199*** (0.015) |
−0.191*** (0.015) |
Sex (Ref: Male) |
|
|
|
female |
|
−0.058 (0.037) |
−0.060 (0.037) |
Race (Ref: White) |
|
|
|
Black |
|
0.013 (0.053) |
0.36 (0.055) |
Other |
|
−0.009 (0.059) |
−0.003 (0.059) |
Nativity (Ref: Yes) |
|
|
|
No |
|
0.243*** (0.059) |
0.243*** (0.059) |
Religious Affiliation (Ref: Protestant) |
|
|
|
Catholic |
|
0.092 (0.060) |
0.097 (0.060) |
Jewish |
|
0.355* (0.151) |
0.325* (0.151) |
Christian |
|
−0.086 (0.109) |
−0.099 (0.109) |
None |
|
0.102 (0.075) |
0.069 (0.077) |
Party ID × Religiosity (Fund) |
|
|
|
Independent × Moderate |
|
|
−0.172 (0.157) |
Independent × Liberal |
|
|
−0.323 (0.155) |
Republican × Moderate |
|
|
0.010 (0.117) |
Republican × Liberal |
|
|
−0.260* (0.127) |
Other Party × Moderate |
|
|
0.319 (0.390) |
Other Party × Liberal |
|
|
0.775 (0.373) |
R-squared |
0.145 |
0.246 |
0.250 |
n = 3346. All entries are unstandardized regression coefficients. Dependent variable: Immigration Attitude Coded as 5 = Increase a lot, 1 = Decrease a lot; *p < 0.05, **p < 0.01, ***p < 0.001.
The bivariate regression model (Model 1) examines the relationship between party identification and immigration attitudes, where the dependent variable (letin1a2) is coded such that higher values indicate stronger support for increasing immigration. Using Democrats as the reference category, the results show substantial and statistically significant partisan differences in immigration attitudes. Independents express significantly lower support for immigration compared to Democrats (b = −0.380, S.E. = 0.051, p < 0.001), while Republicans exhibit a much larger negative association (b = −1.060, S.E. = 0.045, p < 0.001). This implies that, on average, Republicans’ support for increasing immigration is more than one full point lower on the five-point scale than that of Democrats. In contrast, respondents identifying with other political parties do not differ significantly from Democrats.
The magnitude and statistical significance of these coefficients indicate that party identification is a strong predictor of immigration attitudes even when considered independently. The R2 value for this model (0.145) suggests that party identification alone explains approximately 14.5% of the variance in immigration attitudes. This finding highlights the central role of partisan alignment in shaping public opinion toward immigration in the United States and underscores the extent of political polarization surrounding immigration policy.
Model 2 incorporates a set of additional covariates into the baseline specification to account for the broader sociodemographic, political, and religious context influencing immigration attitudes. The results show that party identification continues to exhibit a strong and statistically significant association with immigration attitudes after accounting for age, education, political ideology, race, sex, nativity, religious orientation, and religious affiliation. Using Democrats as the reference category, Independents express significantly lower support for increasing immigration (b = −0.239, S.E. = 0.052, p < 0.001), while Republicans show an even larger negative association (b = −0.567, S.E. = 0.054, p < 0.001). These coefficients indicate that, net of other sociodemographic, ideological, and religious factors, both Independents and Republicans are less supportive of immigration than Democrats. In terms of religiosity, compared to religious fundamentalists, individuals classified as religious moderates exhibit higher support for increasing immigration (b = 0.133, S.E. = 0.063, p < 0.05), while religious liberals also show a positive association, though the effect is smaller and not statistically significant at conventional levels. These results suggest that less fundamentalist religious orientations are generally associated with more favorable views toward immigration.
Consistent with prior research, older and politically more conservative individuals express significantly lower support for immigration, whereas higher levels of education and foreign-born status are associated with more favorable pro-immigration attitudes. More broadly, recent evidence indicates that nativity can also shape other forms of public opinion, particularly through its interaction with education (Islam, 2026). These patterns remain evident even after controlling for party identification, religious orientation, and other sociodemographic characteristics. This suggests that the relationships between age, political ideology, education, nativity, and immigration attitudes are robust.
After controlling for political, ideological, and other sociodemographic factors, racial differences in immigration attitudes are no longer statistically significant. This suggests that previously observed racial gaps are largely explained by correlated political and ideological characteristics rather than race itself (Hainmueller & Hopkins, 2014; Hopkins, 2010). The coefficient for sex is not statistically significant, indicating no meaningful difference in immigration attitudes between men and women once other covariates are taken into account. Most religious affiliations do not differ significantly from Protestants in their immigration attitudes after controlling for religiosity and political factors. An exception is Jewish respondents, who exhibit significantly higher support (b = 0.355, S.E = 0.151, p < 0.05) for immigration, consistent with prior research emphasizing the role of historical experience and minority group identity.
In summary, the bivariate regression results (Model 1) indicate that party identification is strongly associated with immigration attitudes, with Independents and, in particular, Republicans expressing significantly lower support for increasing immigration compared to Democrats. This partisan gap remains statistically significant in the multivariate regression (Model 2), even after adjusting for age, education, political ideology, race, sex, nativity, religious orientation, and religious affiliation, although the magnitude of the coefficients is attenuated. In Model 2, age and political conservatism are negatively associated with support for immigration, while higher education and foreign-born status are positively associated with pro-immigration attitudes. Overall, the multivariate model explains approximately 24.6% of the variance in immigration attitudes (R2 = 0.246), underscoring the importance of partisan, ideological, and sociodemographic factors in shaping public opinion on immigration.
Religious Orientation as a Potential Moderator
Consistent with the hypothesis under consideration, religious orientation is expected to moderate the relationship between party identification and immigration attitudes. This suggests that the strength and direction of association between the focal independent variable (Party ID) and the dependent variable (immigration attitude) may vary across different levels of religious orientation. To test the expected moderation effect, another multiple regression model (Model 3) has been used in this research. This model includes interaction terms (Party ID × Religiosity) along with the other predictors considered in Model 2. The central variables of interest in this model are the focal independent variable (Party ID), the dependent variable (Immigration attitudes), and the potential moderator (religiosity), along with their interaction effects.
From Table 2 (Model 3), the interaction terms indicate that religious orientation moderates the association between party identification and immigration attitudes. Specifically, the interaction between Republican identification and religious liberalism is negative and statistically significant (b = −0.260, S.E. = 0.127, p < 0.05), indicating that the association between party identification and immigration attitudes varies across religious orientations. However, post-estimation analyses show that, within the Republican Party, religious liberals do not differ significantly from religious fundamentalists in their support for immigration (dy/dx = 0.017, p = 0.882). Thus, the significant interaction reflects differences in the partisan association across religious orientations rather than higher immigration support among Republican religious liberals.
By contrast, the interaction between Republican and religious moderation is not statistically significant, implying that partisan differences in immigration attitudes among religious moderates do not differ meaningfully from those observed among religious fundamentalists. Taken together, these findings indicate that the interaction between party identification and religious orientation is statistically significant for some groups. However, post-estimation analyses suggest that these interaction effects should be interpreted cautiously because the predicted differences between religious liberals and religious fundamentalists among Republicans are not statistically significant. The findings provide partial support for the moderating role of religious orientation, showing that religious liberalism attenuates the negative association between Republican identification and support for immigration. This suggests that religious belief systems may, under certain conditions, soften partisan polarization by reshaping how political identities are expressed in attitudes toward immigrants. However, this moderating effect is not uniform across all groups, indicating that the interaction between religiosity and partisanship is limited to specific ideological contexts.
![]()
Figure 1. Predicted probabilities of interaction effects.
The hypothesis that religiosity moderates the relationship between party identification and immigration attitudes receives partial support (Figure 1). Although one interaction term was statistically significant, the remaining interaction terms were not, and post-estimation analyses indicated no significant difference between Republican religious liberals and Republican religious fundamentalists. On the other hand, the results indicate that liberal religious individuals are not associated with more positive immigration attitudes among all the political parties. These suggest that the relationship between party identification and immigration attitudes varies across religious orientations, but the magnitude of this variation should be interpreted cautiously.
This research can potentially contribute to the literature on perceptions of immigrants by demonstrating that religious orientation is theoretically and empirically relevant to views of immigrants in the United States. Additionally, by applying social identity theory, it found that members of marginalized groups demonstrate greater positivity towards immigration than other majority populations. These findings can also contribute robust empirical support for the idea that immigration attitudes in the United States are shaped by a combination of socio-demographic, ideological, and religious factors. Consistent with group threat theory (Blalock, 1967; Bobo & Hutchings, 1996), age and political conservatism were still significantly associated with less favorable views toward increasing immigration. Older individuals and those with more conservative political ideologies expressed greater resistance to immigration, likely due to perceived cultural and economic threats posed by demographic changes.
Previous research consistently finds that Republican political identification is associated with lower openness to immigration and more restrictive attitudes toward immigration in the United States (Valentino et al., 2018; Iyengar et al., 2012; Mason, 2018). However, this study finds that this effect depends on an individual’s religious orientation, indicating that the impact of religiosity is conditional on party identity in support of immigration. This suggests that the influence of party identification may be constrained or overridden by deeply held religious worldviews, consistent with prior research showing that strong ideological commitments can limit the explanatory power of partisan identity (Imanol & Narayani, 2026; Layman, 2001; Margolis, 2018). Religiosity stood out as a strong ideological orientation shaping public opinion towards pro-immigration attitudes.
The results provide strong support for the 2nd hypothesis that individuals with more cosmopolitan orientations, characterized by younger age, more liberal political views, and higher levels of educational attainment, are more likely to express favorable attitudes toward immigration. This study found that individuals with more cosmopolitan orientations are more likely to express favorable attitudes toward immigration. Younger respondents, those with more liberal political views, and individuals with higher levels of educational attainment consistently exhibit greater support for immigration. These associations remain statistically significant after controlling for party identification, religiosity, and other sociodemographic characteristics, indicating that cosmopolitan orientations exert an independent and robust influence on immigration attitudes.
The findings of this study suggest that immigration attitudes are not simply reflections of surface-level demographics or partisan alignment, but are shaped by deeper ideological belief systems and identity-based orientations. Future research should continue to examine these moderating processes using longitudinal or experimental designs to better identify causal mechanisms, particularly in increasingly polarized societies where education, religion, and political identity intersect in complex and sometimes contradictory ways.
6. Conclusion
This study set out to examine some specific predictors of immigration attitudes in the United States, with a particular focus on whether religious orientation moderates the relationship between party identification and support for immigration. Drawing on recent data from the 2022 General Social Survey (GSS), the analysis confirmed that political identity, nativity, age, political ideology, and education remain significant predictors of public opinion toward immigration. In line with previous findings, younger, more educated, and politically liberal individuals were still more likely to support increasing immigration. Racial identity remains insignificant in this analysis, suggesting other factors are largely mediated rather than race itself. A central contribution of this paper lies in its exploration of interaction effects, particularly how religious orientation conditions the influence of party identity towards pro-immigration attitudes. The results suggest that the relationship between party identification and immigration attitudes varies across religious orientations. However, post-estimation analyses indicate that the predicted differences within the Republican subgroup are not statistically significant. Therefore, the findings should be interpreted as evidence of variation in partisan associations across religious orientations rather than as evidence that religious liberalism directly increases Republicans’ support for immigration. This finding underscores the importance of moving beyond additive models to examine how political attitudes vary across belief systems and consider how political attitudes can change depending on a person’s beliefs, especially in today’s divided society.
By integrating perspectives from group threat theory, modernization theory, and social identity theory, this study offers a more nuanced framework for understanding public opinion on immigration. This finding highlights the need for future research to examine how fundamental identity-based beliefs, such as religion, interact with structural influences like party identification in shaping public opinions. In the era of modern democracy, such analyses are essential for understanding the limits of civic education and the role of ideology in shaping public attitudes on immigration.