Perceived Local Population Dynamics and Immigration Policy Views

Abstract

How do perceptions of local immigrant populations influence immigration policy views? Building on findings that Americans may not accurately perceive population dynamics, we argue that objective measures do not fully capture the effects of local context on public opinion. Our research uses novel subjective experimental reminders about current levels of and recent changes in local immigrant populations to explore how these perceptions impact immigration policy views. In a survey experiment, we asked 2,400 Americans to consider current levels of or recent changes in their local immigrant population. Asking subjects to consider current levels of local immigrant populations modestly increases support for pro-immigrant policies, with particularly strong effects among non-White and Republicans. These effects may be driven by positive perceptions of immigrants and have implications for understanding the role of local community frames in shaping public opinion about immigration, particularly for groups who do not typically support permissive immigration policies.

Keywords

immigration public opinion policy views local context survey experiment

Immigration is at the center of policy debates in contemporary American politics, and views on this issue play an important role in shaping political behavior. For example, 52% of respondents to a recent Pew survey cited immigration as “very important” to their vote choice in the 2020 presidential election (Pew Research Center, 2020). Immigration is also one of the most polarizing policy issues in American politics. According to another recent survey, 46% of Democrats but only 17% of Republicans believe it is important to “establish a way for immigrants who came here illegally to stay” (Daniller, 2019). Although many policy debates about immigration occur at the national level, most Americans experience its impact in their local communities. How do Americans perceive immigrant populations in their local communities? Do these perceptions shape immigration policy views?

Existing research investigates the impact of local immigrant context on public opinion, often finding a negative association between living near immigrant populations and support for immigration (Abrajano & Hajnal, 2015; Campbell et al., 2006; Enos, 2014; Ha, 2010; Hopkins, 2010; Rocha et al., 2011). While most studies focus on the political impact of immigrant population levels, there is a debate about whether standing levels of or recent changes in immigrant populations shape policy views (Hopkins, 2010; Newman, 2013). Whether they focus on standing population levels or recent changes, these studies examine the impact of objective demographic characteristics on public opinion, thereby assuming that Americans accurately perceive local population dynamics and their policy views reflect these perceptions.

This assumption is tested in a growing body of research which finds that most Americans do not correctly perceive the demographic composition of their local communities. Many Americans struggle with innumeracy, which leads them to overestimate the proportion of racial minorities both locally and nationally (Nadeau et al., 1993; Wong, 2007; Wong et al., 2012). These misperceptions may also extend to immigrant populations (Hopkins et al., 2019). Although perceptions may only be weakly correlated with actual population dynamics, they have implications for political attitude formation (Coulton et al., 2001; Newman et al., 2015; Velez & Wong, 2017; Wong, 2007; Wong et al., 2012). For example, Americans’ perceptions of local immigrant demographics mediate the effects of local context on policy views (Newman et al., 2015). Moreover, recent studies find that correcting such misperceptions is difficult and has limited impact on political views (Hopkins et al., 2019; Swire-Thompson et al., 2020).

Building on this line of work, we re-examine the debate about the effects of standing levels of and changes in local immigrant populations on policy views using subjective perceptions of local immigrant context. We argue that examining the effects of objective measures of immigrant populations offers, at best, an incomplete, and at worst, a distorted understanding of how local immigrant context shapes policy views. Does asking individuals to think about standing levels of and recent changes in local immigrant populations increase or decrease support for permissive immigration policies? This study aims to answer this question, testing the causal impact of Americans’ subjective demographic perceptions on immigration policy views.

This research offers several contributions to the literature on local immigrant context and policy views. First, we explore the extent to which Americans accurately perceive local immigrant population dynamics. We find that perceptions of local immigrant population levels are weakly correlated with objective measures, but perceptions of recent changes in immigrant populations do not reflect actual demographics. Second, we conduct a survey experiment, analyzing the impact of asking Americans to subjectively consider current levels of or recent changes in the immigrant population in their local municipality on policy views. The current immigrant population levels frame increases support for permissive immigration policies by about 3 percentage points, while the recent immigrant population changes frame does not affect policy views. These effects are substantively and statistically significant. They are also similar in magnitude to those of established predictors of immigration policy views, including race and education (Macdonald, 2021).

Since views on immigration are starkly divided across lines of race, immigrant status, and partisanship, we also explore whether individual- and community-level measures of these factors moderate the experimental treatment effects (Abrajano & Hajnal, 2015; Hawley, 2011; Nteta & Rice, 2021; Mangum, 2019; Masuoka & Junn, 2013; Reny et al., 2019). We find several differences in the effects of the current immigrant population levels frame by individual race and partisanship. After being asked to consider levels of immigrants in their community, non-Whites (but not Whites) and Republicans (but not Democrats) become more supportive of pro-immigrant policies, relative to those in the control condition. Although the recent immigrant population changes frame does not have direct effects on policy views, it increases support for pro-immigrant policies among respondents in communities with small (but not large) naturalized and college-educated immigrant populations. Finally, text analysis of open-ended responses investigates the mechanism by which views on immigration change, providing preliminary evidence that public opinion becomes more favorable towards immigrants through personal interactions with immigrants, as theories of intergroup contact predict (Allport; Ha, 2010; Hood & Morris, 1997; Oliver & Wong, 2003).

These findings have implications for understanding how frames of immigration impact public opinion and policy views (Haynes et al., 2016; Knoll et al., 2011; Merolla et al., 2013). Our results point to the power of frames of immigration that center local community in shaping Americans’, and particularly Republicans’, views on immigration policy. In a time when frames of immigrants are increasingly negative (Abrajano et al., 2017; Farris & Silber Mohamed, 2018), and Americans are highly polarized on this issue across party lines, an alternative framing of immigration that centers localities may increase support for more permissive immigration policies, even among subgroups that do not generally favor immigration.

Literature Review

This study draws on several strands of literature, including research on subjective and objective perceptions of local population dynamics and studies about whether current levels of or changes in local immigrant populations impact public opinion. In analyzing moderating effects of individual- and community-level characteristics, this study also draws on work about the role of race, immigrant generational status, and partisanship in shaping views on immigration. Finally, the qualitative analysis of mechanisms engages with research on whether exposure to immigrants in local communities affects public opinion through intergroup contact or group threat.

Subjective Versus Objective Perceptions of Local Population Dynamics

Several studies suggest that objective demographic measures do not provide a complete picture of the political effects of local population context. First, most Americans’ perceptions of local spaces do not align neatly with objectively geographic units (Coulton et al., 2001). As Wong et al. put it, “the evidence is clear: ‘pictures in our heads’ do not resemble government administrative units in shape or content” (2012, 1153). Across several studies, the socio-economic and demographic characteristics of respondent-defined neighborhoods also differ from those of census-defined neighborhoods (Coulton et al., 2001; Velez & Wong, 2017; Wong et al., 2012).

Second, there is limited evidence that individuals accurately assess the demographic composition of objective geographic units. For example, many Americans overestimate the proportion of racial minorities in their communities and the nation (Alba et al., 2005; Baybeck & McClurg, 2005; Cho & Baer, 2011; Coulton et al., 2001; Nadeau et al., 1993). Americans are better at estimating the demographic characteristics of small geographic units, like zip codes, than counties or states, but estimates on all geographic levels are only moderately correlated with objective measures (Newman et al., 2015; Wong, 2007). Moreover, subjective group size estimates have larger effects on political attitudes than their objective counterparts (Wong et al., 2012). These findings may extend to immigrant population context (Hopkins et al., 2019).

The gap between the perceived demographics of respondent-defined communities and the objective characteristics of census-defined geographic units suggests that objective contextual measures may offer an incomplete picture of the effects of local population context on political views. Building on this literature, we first explore whether individuals accurately assess the size of immigrant populations in their local communities.

Current Levels of Versus Recent Changes in Local Immigrant Populations

Next, we re-evaluate the scholarly debate about whether standing levels of and changes in local immigrant populations shape public opinion. While many studies find that local immigrant population levels shape views on immigration (Abrajano & Hajnal, 2015; Rocha et al., 2011; Stein et al., 2000), others show that a sudden change in the immigrant population leads to hostile views (Hopkins, 2010; Newman, 2013). Hopkins (2010) argues that “sudden demographic changes” shape policy views by “generating uncertainty and attention” to immigration and related issues (40). Building on this claim, Newman (2013) develops the “acculturating contexts hypothesis,” which predicts that feelings of threat are activated when immigrants enter areas that previously had few immigrants.

Newman and Velez (2014) test the claim that Americans are more attentive to changes in group size than current levels of local immigrant populations. They find support for the “salience-of-change” hypothesis, which predicts changes in the immigrant population are salient because they capture people’s attention. By contrast, immigrant population levels (i.e., the current size of the immigrant population) have limited effects on public opinion because they do not heighten awareness of the demographic composition of a local environment. These studies are grounded in the assumption that rapid changes in the local immigrant population influence citizens’ political views by shaping the dynamics of intergroup relations in the locality.

Taken together, these findings suggest that changes in immigrant population dynamics have larger effects on public opinion than current immigrant levels. However, these studies use objective rather than perceived measures of local immigrant context. Furthermore, they largely rely on observational data and cannot rule out the possibility of confounding effects.

Individual and Contextual Drivers of Views on Immigration

In recognition of the centrality of race, partisanship, and immigrant generation to the formation of immigration attitudes, we examine whether individual- and community-level measures of these factors moderate the effects of perceived immigrant context on policy views.

Existing research finds that public opinion on immigration varies by race and generational status. White and native-born Americans are more likely to support restrictive immigration policies while foreign-born and non-White Americans are more likely to support permissive immigration policies (Mangum, 2019; Masuoka & Junn, 2013). Public opinion on immigration also varies by the racial and immigrant demographics of local communities, often following similar patterns. For example, backlash effects against local immigrant populations are typically found in studies of White Americans and may not extend to people of color (Abrajano & Hajnal, 2015; Reny et al., 2019; Rocha et al., 2011).

Partisanship is another well-known predictor of immigration policy views (Oskooii et al., 2018; Pearson-Merkowitz et al., 2016). The two major political parties and their supporters are increasingly polarized over a host of issues, including immigration (Mason, 2016; McCarty et al., 2016). The contemporary Republican Party strongly opposes any loosening of immigration restrictions and increased benefits for immigrants currently in the country, while the Democratic Party generally supports immigration and immigrant inclusion in the social safety net (Tichenor, 2009). Votes on immigration policy in Congress have also closely followed party lines in recent years (Casellas & Leal, 2013; McCarty et al., 2016).

Although public opinion on immigration is highly polarized along party lines, existing research suggests that frames of immigration have the potential to impact policy views among members of both parties (Alamillo et al., 2019; Knoll et al., 2011; Merolla et al., 2013; Reich & Mendoza, 2008). Moreover, certain frames can shift Americans’ policy views in the same direction regardless of their party affiliation. For example, negative frames of immigration that center crime lead both Democrats and Republicans to oppose a pathway to legalization for undocumented immigrants (Haynes et al., 2016). In fact, partisan differences in views on specific immigration policies may not be very large to begin with since Democrats, Independents, and Republicans largely agree on the policy consequences of immigration (Neiman et al., 2006). However, other work suggests that frames of immigration may have a stronger impact among Republicans, who are more likely than Democrats to rank immigration as an important policy issue (Knoll et al., 2011).

Identifying the Mechanism: Group Threat Versus Intergroup Contact

The final section of the paper explores the mechanisms by which immigration policy views change. We consider the positive mechanism of intergroup contact and the negative mechanism of group threat. While our experiment does not directly these test theories, the textual analysis offers insight into the role that each may play in shaping policy views.

Theories of group threat predict that when members of a group feel threatened by the presence of another group in their local environment, it leads to backlash against out-group members (Key, 1949). This perspective focuses on intergroup relations in local communities. While theories of group threat were originally used to described relations between African Americans and Whites (Baybeck, 2006; Enos, 2016; Giles & Hertz, 1994; Hero, 1998; Taylor, 1998), they also apply to White Americans’ attitudes about immigration (e.g. Abrajano & Hajnal, 2015; Campbell et al., 2006; Rocha et al., 2011). These studies find that living in areas where there are many opportunities for contact between immigrants and non-immigrants lead to feelings of threat, as well as opposition to immigration and policies that help immigrants.

In contrast, theories of intergroup contact predict that exposure to members of racial out-groups reduces race-based stereotypes and leads to positive intergroup relations. In a seminal study, Allport ([1954] 1988) argues that for intergroup contact to lead to positive outcomes, members of both groups must share “equal status and common goals…” supported by “institutions” and “the perception of common interests and common humanity” (281). This theoretical perspective finds strong support across the social sciences (Christ et al., 2014; Dyck & Pearson-Merkowitz, 2014; Glasford & Calcagno, 2012; Ha, 2010; Hewstone & Swart, 2011; Hood & Morris, 1997; Pettigrew, 1997; Schmid et al., 2014; Tausch et al., 2010) and in several studies focused on immigrant context (Oliver & Wong, 2003; Stein et al., 2000).

Although much of the existing literature finds support for group threat as an explanation for immigration policy attitudes, several features of existing studies may lead to a hasty rejection of intergroup contact. First, differences in the geographic unit of analysis may explain variation in the political effects of objective contextual measures on public opinion across studies (Cho & Baer, 2011). Second, many of the geographic contexts examined by existing research are large and lack the social significance of smaller subjective contexts like neighborhoods, which are not clearly geographically bounded (Enos, 2017). Finally, many of these studies rely on observational data and cannot fully rule out the confounding effects of other local characteristics.

Theory and Hypotheses

Are Americans’ Perceptions of Local Immigrant Context Accurate?

In line with existing work, the immigrant population perception hypothesis predicts that Americans’ incorrect perceptions of local demographic characteristics extend to immigrant populations. At best, perceptions of local immigrant populations will be only weakly correlated with objective demographic characteristics.

Current Levels Versus Changes in Immigrant Population Levels

Does asking respondents to think about current levels of or recent changes in local immigrant populations affect immigration policy views? Building on findings that current immigrant population levels influence public opinion, the immigrant population levels hypothesis predicts that asking respondents to think about current levels of the immigrant population in their subjectively defined local community will affect immigration policy views. In contrast, the immigrant population change hypothesis predicts that asking respondents to think about changes in the immigrant population shapes public opinion. As discussed above, the literature is divided over these predictions. We test both possibilities.

Direction of Change: Do Subjective Reminders Increase or Decrease Support for Immigration?

We are also interested in the direction of change in policy views. Do subjective reminders of local immigrant context increase or decrease support for immigration? The positive immigrant perception hypothesis predicts that thinking about local immigrant constituencies, whether those populations are perceived as large or small, will lead to positive attitudes about immigrants and increase support for permissive immigration policies. In contrast, the negative immigrant perception hypothesis predicts that thinking about immigrant populations will lead to backlash against immigrants and increase support for restrictive immigration policies.

Moderating Effects of Race, Immigrant Status, and Partisanship

Finally, it is possible that that the effects of immigrant population cues are moderated by race, immigrant generation, and partisanship on the individual and community levels. The centrality of race, immigration, status, and partisanship to the development of immigration policy views suggests that the immigrant context treatments may have differential effects among Whites and non-Whites, immigrants and non-immigrants, and Democrats and Republicans. Consistent with this claim, the individual-level moderators hypothesis predicts that individual-level race, immigrant generation, and partisanship moderate the treatment effects. Given that these factors also operate on the contextual level, the community-level moderators hypothesis predicts that the treatment effects differ by the immigrant and partisan composition of respondent’s local communities.

Data and Methods

To test our hypotheses, we analyze a survey experiment fielded on a national online panel of 2,400 American adults recruited to participate by the survey research firm CINT. Approximately 70% of the respondents were White and 30% were ethnic or racial minorities. Conducted in the fall of 2018, the experiment was embedded in an omnibus survey that included questions about respondents’ demographics, health, news consumption patterns, political interest, as well as several other embedded experiments. The use of random assignment and a pure control condition in our experiment should mitigate any potential bias from treatment spillover effects (Gaines et al., 2007; Transue et al., 2009). While our sample was drawn from an opt-in online panel, it is similar to the nationally representative 2018 Cooperative Election Study (CES) sample on several key demographic characteristics (Appendix A).

After answering standard demographic questions (race, age, state of residence, zip code, education, gender, religion, income, partisanship, and ideology), each participant in our study was randomly assigned to one of three experimental conditions: an immigrant population levels treatment condition (n = 803), an immigrant population change treatment condition (n = 799), or a no-information control condition (n = 798). Subjects in the immigration population levels and immigration population change treatment conditions answered two “questions-as-treatments” that cued aspects of their municipality’s local immigrant population, before they were asked several immigration policy items that we analyzed as our dependent variables. By contrast, subjects in the no-information control condition were not exposed to any cues about their municipality’s local immigrant population and skipped directly to answering the immigration policy items.

The immigrant population levels treatment asked respondents: “Thinking about the local municipality where you currently live, about how large would you say the immigrant population is today?” Closed-ended response options included: “Not large at all,” “not too large,” “somewhat large,” “very large,” and “extremely large.” To ensure that participants internalized the treatment and thought about the size of the immigrant population in the municipality where they live, a follow-up question asked respondents to “please write one or two sentences about how the current size of the immigrant population in your local municipality makes you feel.” We use these open-ended responses later to explore possible mechanisms for our experimental results.

The immigrant population change treatment asked respondents: “Thinking about the local municipality where you currently live, about how much has the immigrant population increased or decreased over the last 15 years?” Closed-ended response options included: “Decreased a lot,” “decreased somewhat,” “stayed about the same,” “increased somewhat,” and “increased a lot.”¹ Here, again to ensure that participants internalized the treatment and thought about changes in the immigrant population in the municipality where they live, a follow-up question asked respondents to “please write 1 or 2 sentences about how the increase or decrease over the last 15 years in the size of the immigrant population in your local municipality makes you feel.” We also use these open-ended responses to explore possible mechanisms for our experimental results.

Table 1 presents information about our experimental design visually. The three experimental conditions are shown across the columns, with the number of respondents in each condition reported in the first row of the table. The second and third rows show the exact order and wording of, respectively, the closed- and open-ended questions-as-treatments that we used to cue respondents with information about the immigration population in their local municipality, or “NA” for subjects in the control condition. These are followed by the immigration policy items asked of subjects in the treatment conditions after they completed the open-ended question-as- treatment task, or immediately after demographic items for subjects in the control condition.

Table 1.

Immigration Levels and Change Experimental Design.

N	Immigrant levels treatment	Immigrant change treatment	No-information control
N	803	799	798
Question- as- treatment wording	Thinking about the local municipality where you currently live, about how large would you say the immigrant population is today?	Thinking about the local municipality where you currently live, about how much has the immigrant population increased or decreased over the last 15 years?	NA
Question- as- treatment wording	Response options: Extremely large, very large, somewhat large, not too large, not large at all	Response options: Increased a lot, increased somewhat, stayed about the same, decreased somewhat, decreased a lot	NA
Open- ended question- as- treatment wording	Now please write one or two sentences about how the current size of the immigrant population in your local municipality makes you feel	Now please write one or two sentences about how the increase or decrease over the last 15 years in the size of the immigrant population in your local municipality makes you feel.	NA
Dependent variables	Immigration policy items	Immigration policy items	Immigration policy items
Additional post- treatment questions	Immigration CHANGE question- as-treatment & open- ended follow-up	Immigration LEVELS question-as-treatment & open-ended follow-up	Immigration CHANGE question-as-treatment & open-ended follow-up
Additional post- treatment questions			Immigration LEVELS question-as-treatment & open-ended follow-up

After answering the immigration policy questions, all subjects then completed some additional post-treatment questions: those in the immigrant levels condition answered the immigration change close-ended question-as-treatment, and those in the immigrant change condition answered the immigration levels close-ended question-as-treatment. Those in the control condition answered both the immigrant levels and immigrant change close-ended questions-as- treatments. All respondents answered these closed-ended questions at the end of the experiment to generate baseline subjective population perception measures for all subjects.

Table 2 shows mean values of various characteristics for the respondents randomly assigned to each of our experimental conditions. Balance tests confirm that randomization to treatment and control conditions was successful, with no major differences in sample characteristics by condition.

Table 2.

Mean Demographic Values and Balance Tests Across Experimental Conditions.

Demographic variable	Control condition	Immigrant levels treatment	Immigrant changes treatment	One-way ANOVA test
Race (White)	0.73	0.72	0.69	F = 1.13, p = .32
Age	47.07	46.66	46.00	F = 0.94, p = .39
Gender (Female)	0.53	0.55	0.54	F = 0.62, p = .54
Education	0.70	0.69	0.69	F = 0.56, p = .57
Income	0.52	0.52	0.53	F = 0.45, p = .64
Married	0.48	0.49	0.48	F = 0.09, p = .91
Foreign born	0.05	0.05	0.04	F = 0.11, p = .90
Partisanship	0.48	0.46	0.45	F = 0.90, p = .41

Note. Age is a numeric measure. All other variables are scaled from 0 to 1. Race and gender are binary indicators for “White” and “female.” Partisanship is coded so higher values reflect Republican identification.

The immigrant population levels and change treatments ask respondents to think about the local immigrant population in self-defined communities. While the term ‘municipality’ formally refers to “a town, city, or district with its own local government” (Oxford Learner’s, 2022), the flexible nature of the term allows respondents to draw from experiences in a variety of local spaces rather than rigidly defining which geographic entities they may consider. In encouraging subjects to think of the immigrant population in a locality defined in their own terms, these treatments vary a potential mediator of the effects of local context on public opinion – individuals’ perceptions of that local immigrant population.²

We used this design to elicit perceptions of local immigrant population levels and changes, whatever those perceptions may be. We also considered using vignettes to manipulate perceptions of local immigrant populations (e.g., informing respondents that many or very few immigrants live in their communities). However, we did not think this alternative design would be effective given that misperceptions about immigrant populations are hard to correct, especially in the context of survey experiments (Hopkins et al., 2019). Although our design does not manipulate respondents’ perceptions of immigrant population levels and changes, it encourages respondents to consider local population dynamics prior to reporting policy views. This is a useful first step in exploring the effects of perceptions of immigrant populations on policy views. Future work might randomize perceptions of local immigrant context and identify the causal effects of perceiving large or small population levels and recent increases or decreases in immigrant populations on policy views.

To examine whether respondents accurately perceive population dynamics, we compare the closed-ended responses to objective measures of the immigrant population. We conceptualize objective local context on the zip code level, following previous research (Newman et al., 2015; Velez & Wong, 2017). We argue that zip codes generally map onto respondents’ perceptions of municipalities because they are small enough to be considered local but large enough to incorporate many of the places that individuals visit in daily life, including schools, parks, shops, restaurants, and offices. We use the percentage of foreign-born residents in a respondent’s zip code as an objective measure of current immigrant population levels. The change in the percentage of foreign-born residents in a respondent’s zip code over a 7-year period³ serves as an objective measure of changes in immigrant population levels. Both are drawn from the 5-year ACS.

To compare perceptions to observed measures, we present scatterplots and bivariate correlation coefficients. Following Newman et al. (2015), we also regress the perceived measures onto observed measures and demographic controls using standard OLS and hierarchical linear models that nest respondents into the geographic unit of analysis.⁴ Since some respondents may conceptualize their municipality on a higher level, we conduct robustness checks at the Census-designated place and county levels.⁵

Next, we test for treatment effects on the following dependent variables: (1) opposition to a law increasing regulation of immigration, (2) support for a politician who proposes a pathway to citizenship for undocumented immigrants, (3) support for a law increasing the number of refugees allowed to enter the U.S., (4) belief that local sanctuary policies rarely obstruct federal immigration investigations, and (5) a scaled index of the four pro-immigration views. These items capture a wide range of immigration policies that are implemented on the local, state, and national levels. Items 1–3 were measured on identical five-point Likert scales,⁶ while item 4 was also measured on a five-point scale.⁷ We created an index of these four items because they scale well together (alpha = .76) and collectively capture greater variation of views on a range of immigration policies than any single item. We also create a modified version of the index, which omits item 4 (alpha = .73). All dependent variables are scaled from zero to one and coded so that higher values reflect greater support for immigration for pro-immigration policies.

We calculated average treatment effects for (1) the immigrant population levels treatment versus control and (2) the immigrant population change treatment versus control for the five outcomes. The relevant comparison for each treatment is the no-information control condition. To calculate average treatment effects, we conducted weighted difference-in-means tests for the full sample and binary subsets for the individual and community level moderators. The survey weights approximate a nationally representative sample.

As described above, respondents provided textual responses after viewing the experimental treatments. We use this data to further explore differences in public opinion on immigration across conditions and explore potential mechanisms driving the results. We use dictionary methods to identify the sentiments linked to the open-ended responses. This method uses dictionaries, or pre-sorted groupings of words which correspond with particular sentiments and match them with words in the textual responses. The result is a count of how many words in each response correspond to sentiments such as “positive,” “negative,” “fear,” and “anger.” We use the 2015 Lexicoder Sentiment Dictionary through the quanteda R package (Benoit et al., 2018) to find the average number of positive and negative terms in each open-ended response.

We also coded each open-ended response qualitatively for valence and subject matter. The valence was coded as “positive” or “negative” (see Appendix F for coding details). Responses were coded as positive if they were generally positive or indicated that the subject was happy to have immigrants in their community. Responses were coded as negative if they were generally negative or indicated that the subject was unhappy with the presence of immigrants in their community. The qualitative subject matter coding allows us to analyze these responses with more nuance than quantitative methods allow. We coded responses for mentions of economic issues, diversity, documentation status, crime, and government assistance. A single coder coded every open-ended response. To check for inter-coder reliability, another coder coded a random subset of 20% of the responses (see Appendix F). Any discrepancies were discussed and resolved.

Results

Perceived Versus Objective Immigrant Population Levels and Changes

To test the immigrant population perception hypotheses, we compare respondents’ perceptions of levels and changes in local immigrant populations to objective measures of the immigrant population in their zip codes. In line with the hypothesis, most respondents did not have an accurate sense of immigrant population dynamics. However, respondents had a better sense of standing levels of than recent changes in immigrant populations.

Figure 1 shows a weak positive association between the perceived size of the immigrant population in a respondent’s municipality and the proportion foreign born in their zip code (Pearson’s correlation coefficient = .34). The association between the perceived change in the immigrant population in a respondent’s municipality and changes in the proportion foreign born in their zip code is negative and near zero (Pearson’s correlation coefficient = −.006) (Figure 2). There is also a high degree of variance in objective population estimates within each subjective category. These findings are robust to the Census-designated place and county levels (Appendix Figures B1 and B2).

Figure 1.

Perceived versus observed local immigrant population levels (zip code level).

Figure 2.

Perceived versus observed change in local immigrant population levels (zip code level).

To offer a more nuanced analysis of the relationship between these measures, we regress the perceived measures onto the observed measures and demographic controls. In line with the bivariate results, the proportion foreign born in respondents’ zip codes has a positive and statistically significant association with perceived immigrant levels, but the OLS regression model only explains about a fifth of the variation in the five-point perceived measure, suggesting that many other factors shape perceptions of local immigrant populations (Appendix Table B3). In contrast, the change in the proportion foreign born in respondents’ zip codes is not associated with perceive changes in immigrant levels (Appendix Table B4). These results extend to the Census-designated place and county levels (Appendix Tables B3 and B4).

Interpreting Subjective Measures of Local Immigrant Context

In keeping with the subjective nature of perceptions of population dynamics, respondents’ post-treatment open-ended reflections suggest that they conceptualize “local context” on a variety of geographical levels. Some respondents referred to the region or city they live in while others spoke about towns and neighborhoods. For other subjects, perceptions of immigrants involved immediate social ties, including the individuals their children grew up with or the owners and patrons of local stores. Yet others mention hearing languages other than English spoken while out shopping or knowing that there are undocumented immigrants in their schools or neighborhoods. These responses indicate that objective measures of local immigrant context pass through subjective perceptual filters.

Experimental Results

Next, we present the results of the survey experiment. Table 3 presents the weighted mean values of each immigration policy outcome across treatment conditions. It also displays the average treatment effects of the immigrant levels and change treatments, relative to the control, with p-values extracted from t-tests to identify statistically significant differences. Unweighted mean values are included in Appendix Table C1. Figure 3 presents these treatment effects visually for the immigrant levels treatment versus the control (panel 1) and the immigrant changes treatment versus the control (panel 2). The effects are plotted with 95% confidence intervals to approximate statistically significant differences from zero.

Table 3.

Weighted Mean Outcome Values and Average Treatment Effects (Full Sample).

	Mean values			Treatment effects
Outcome	Control	Immigrant levels treatment	Immigrant change treatment	Immigrant levels - Control	Immigrant change - Control
Oppose law restricting immigration	0.39	0.43	0.42	0.04*	0.03
Support proposing pathway to citizenship	0.52	0.53	0.53	0.01	0.01
Agree sanctuary policies rarely obstruct	0.38	0.40	0.37	0.02^	−0.01
Support increasing refugees to the US	0.43	0.46	0.45	0.03	0.02
Pro-immigration scale	0.43	0.46	0.44	0.03*	0.01
Modified pro- immigration scale	0.45	0.47	0.47	0.02^	0.02

Note. ^p < .10, *p < .05. Mean values are rounded to 2^nd decimal. Modified pro-immigrant scale drops “sanctuary city” item.

Figure 3.

Average treatment effects for the full sample.

In line with the immigrant population levels hypothesis, asking respondents to consider current local immigrant population levels modestly but consistently affects immigration policy views, relative to the control. These results also offer insight on the direction of attitude change. As the positive immigrant perception hypothesis predicts, asking respondents to think about the standing levels of the local immigrant population increases support for pro-immigrant policies.

As the first panel of Figure 3 shows, exposure to the immigrant population levels treatment has a positive and statistically significant effect on support for several pro-immigration policies among the full sample. Asking people to think about current levels of the local immigrant population increases opposition to laws reducing immigration (“oppose restrictions”) by 4 percentage points (p < .05) and support for a scaled measure of all four policy outcomes (“pro-immigration scale”) by about 3 percentage points (p < .05).⁸ The immigrant population levels treatment also increases belief that sanctuary cities rarely obstruct federal investigation (“support sanctuary”) of immigration by about 2 percentage points (p < .10) and support for a law increasing the number of refugees admitted to the U.S. (“support refugees”) by about 3 percentage points (p = .104), although these effects are more uncertain.

In contrast, we find no support for the immigrant population change or the negative immigrant perception hypotheses. The second panel of Figure 3 shows the immigrant population change treatment does not have statistically significant treatment effects on any of the policy outcomes. Also, neither treatment decreases support for permissive immigration policies.⁹

Figure 4 shows the results of the sentiment analysis, comparing the number and valence of terms in open-ended responses to the immigrant levels and changes treatments. Few differences emerge in the total number of terms, sentiment terms, or the nature of the sentiments that the two treatments elicit. Among the full sample, the immigrant population change treatment elicited about 1.79 more total words than the immigrant population levels treatment. This difference may be due to respondents needing more words to discuss changes than standing population levels.

Figure 4.

Key sentiment differences across treatments.

Moderating Effects of Individual and Community Level Characteristics

To examine whether the experimental results are moderated by race, immigrant generation, and partisanship, we re-estimate the treatment effects for relevant subsets of respondents. The individual-level moderator subgroups include White and non-White respondents (race), first-, second-, and third-generation immigrants (immigrant generation), and Democrats and Republicans (partisanship). Community-level partisanship subgroups include respondents who live in counties where Democratic or Republican U.S. House candidates won by a margin of at least 10 points in 2018. The community-level immigrant demographic subgroups include binary groupings (cut off at the median) of (1) the percentage of foreign-born residents, (2) the percentage of non-citizens (3) the percentage of naturalized citizens, (4) and the percentage of foreign-born residents with a bachelor’s degree in the respondent’s zip code in 2018. These measures, drawn from the ACS, pick up on various characteristics of local immigrant communities, including the recency of immigrant flows, levels of education, and acculturation.

As the individual-level moderators hypothesis predicts, individual-level race and partisanship moderate the effects of the immigrant levels treatment.¹⁰ We identified statistically significant differences in treatment effects between subgroups with regression models that interact the immigrant levels treatment with the race and partisanship moderator measures. We plot the average treatment effects separately for each subgroup of the race and partisanship moderators in Figure 5 and Figure 6. The figures visually approximate statistically significant differences across moderator subgroups through non-overlapping 83.4% confidence intervals (Goldstein & Healy, 1995). Additional information, including subgroup sample sizes, average treatment effects for each subgroup, and full interactive regression results, is included in Appendix E.

Figure 5.

Average treatment effects for White and non-White racial subgroups.

Figure 6.

Average treatment effects for Democratic and Republican partisan subgroups.

As Figure 5 shows, the immigrant levels treatment has differential effects on White and non-White respondents on support for a pathway to citizenship. Non-White respondents are more likely to support a pathway to citizenship when reminded of the level of immigrants in the local population while White respondents are not (Appendix Table E2). This difference in treatment effects across racial subgroups is statistically significant at the p < .05 level (Appendix Table E3). This indicates that the positive effects of considering immigrant population levels on support for pro-immigrant policies may be larger for racial minorities, as existing research suggests (Masuoka & Junn, 2013). In a robustness check, we also explore whether college education moderates that treatment effects for White respondents. No significant differences emerge in treatment effects between White respondents with and without bachelor’s degrees (Appendix Table E12).

As Figure 6 shows, the immigrant levels treatment also has differential effects among Democrats and Republicans.¹¹ Unexpectedly, Republican respondents are more likely support sanctuary cities when reminded of the level of immigrants in their locality, while Democrats are not (Appendix Table E5). This difference across partisan subgroups is statistically significant at the p < .05 level (Appendix Table E6). The immigrant levels treatment also increases opposition to policies restricting immigration among Republicans but not Democrats. However, this difference is only marginally significant (Appendix Tables E5 and E6).

It is important to note that baseline levels of support for permissive immigrant policies differ across partisan subgroups. As we might expect, mean levels of support for the four permissive immigrant policies and the scaled index differ substantially among Democrats and Republicans across all experimental conditions, with Democrats expressing higher levels of support for pro-immigrant policies (Appendix Table E5). The fact that Democrats’ policy views are not as easily changed by the immigrant levels treatment as Republicans may indicate a ceiling effect since many Democrats already support permissive immigration policies. None the less, these results suggest that reminders of immigrant population levels can increase support for pro-immigrant policies, even among groups that generally favor restricting immigration. This is an important finding in relation to research on the effects of immigration frames on policy views in a highly polarized political environment. Our findings suggest that local community frames can increase support among Republicans, a subgroup of Americans who are unlikely to support immigration.

To further understand these differences across partisan subgroups, we explore whether textual responses to the treatments varied between Democrats and Republicans. Figure 7 shows that among respondents exposed to the immigrant population levels treatment, Republicans use significantly more terms overall, more negative terms, and more anger and fear sentiment terms than Democrats. When using a dictionary to identify valence along with other sentiments, we find that among respondents exposed to the immigrant population change treatment, Republicans use significantly more terms overall and more negative terms than Democrats.

Figure 7.

Key sentiment differences within treatments (Democrats - Republicans).

Turning to the community-level moderators, none of the measures of immigrant demographics or partisan context have significant moderating effects for the immigrant levels treatment. However, several community level measures of immigrant demographics shape responses to the immigrant changes treatment, which does not affect policy views for the full sample.

We find that some characteristics of immigrant communities moderate the immigrant changes treatment effects. First, the immigrant changes treatment increases support for a pathway to citizenship and for refugees among those who live in zip codes with a low (but not high) percentage of naturalized citizens (Appendix Table E20). The immigrant changes treatment also increases support for a pathway to citizenship among those who live in zip codes with a low (but not high) percentage of foreign-born residents with a bachelor’s degree (Appendix Table E26). These subgroup differences are significantly different at p < .05 (Appendix Tables E21 and E27). Community-level measures of percent foreign born and non-citizen do not moderate the treatment effects (Appendix Tables E18 and E24). These results support the community-level moderators hypothesis and suggest that the immigrant change treatment shapes the policy views of people who live in areas with less acculturated and educated immigrant populations.

Exploring the Mechanism: Intergroup Contact Versus Group Threat

The experimental findings indicate that thinking about the local immigrant population – however large or small respondents perceive the population to be – boosts support for policies that help immigrants. However, the results do not offer direct evidence of the mechanisms driving this positive effect. Next, we examine the open-ended textual responses to assess whether positive experiences of contact with immigrants drive these effects. We asked subjects to write one or two sentences about how the immigrant population in their locality makes them feel. From these responses, we gain some insight on what may drive changes in immigration policy views.

Results from coding the open-ended responses offer evidence that positive experiences with diversity in local contexts may drive pro-immigration policy views (see Appendix F). Based on the qualitative coding, about 27% of respondents spoke positively and 21% spoke negatively of immigrants and immigration (Figure 8).¹² About 19% of all responses mention diversity, 16% economic issues, 10% legality issues, 8% crime, and 6% public resources. Of the positive valence responses, 43% discussed diversity, 21% economic issues, 5% issues of legality, 2% crime, and 2% public resources (see Figure 8).¹³ Out of the negative valence responses, about 19% mention diversity, 24% economic issues, 18% legality issues, 21% crime, and 20% public resources. The fact that a larger percentage of positive than negative responses mention diversity suggests that social interactions with diverse populations may lead to positive sentiments about immigrants.

Figure 8.

Distribution of valence and topics in open-ended responses.

Many of the positive comments discussed how experiencing new cultures and getting to know immigrants personally increased their support for immigration. We offer several excerpts that reflect these sentiments. The quotes are included for illustrative purposes and are not a representative sample of open-ended responses. However, they offer insight into respondents’ experiences with and perspectives on immigrants in their locality.

One respondent stated that: “I think the diversity of our immigrants in our city can only be positive...The difference in our city’s citizens contributes to our choice of different foods, languages and getting to know those from cultures we may never have had an opportunity to meet otherwise.” For this respondent, encountering immigrants who bring different culinary, linguistic, and cultural traditions is a positive experience which comes to mind when asked about the local immigrant community. Another respondent expressed a similar sentiment, stating “[t]he cultural enrichment brought by immigrants is vital to a community” and that they “enjoy the diversity in the community, it enriched our communities, I especially enjoy the Hispanic cultures, they are a rich warm culture and bring much to our communities!” A third respondent said: “I work with legal immigrants every day, most from Asian and Middle East countries. This does not bother me at all. They are educated and bring valuable skills, pay taxes, and bring a different culture while assimilating into the US culture.”

Conclusion

This research addresses several questions about how Americans develop attitudes about immigration policy, a salient topic in contemporary American politics. To briefly summarize the key findings, the current immigrant population levels frame increases support for permissive immigration policies. In contrast, the recent changes frame does not affect policy attitudes. These effects vary by community- and individual-level characteristics, including the demographics of local immigrant populations, as well as individual race and partisanship. Text analysis offers preliminary evidence that the mechanisms driving changes in policy views include exposure to diversity and contact with immigrants.

The results of this experimentally grounded examination have several important implications for understanding the consequences of local community frames of immigration. Most broadly, this research conveys that local community frames of immigration can shape public opinion on immigration policy. This is notable in the context of research on immigration framing effects, which finds that a wide range of framing techniques, including frames that feature positive and negative rhetoric, frames that focus immigration policy, “equivalency frames” that use different terms to describe immigrants, “episodic” frames that focus on individual immigrants, and “thematic” frames that offer a fuller picture of the issue shape immigration policy views (Alamillo et al., 2019; Haynes et al., 2016; Merolla et al., 2013; Newton, 2008; Reich & Mendoza, 2008) Our research shows that a very different framing device – frames that focus on immigrant populations in local communities – can also shape Americans’ policy views. Framing immigration in relation to individual’s experiences with immigrants in their most proximate local environments increases support for permissive immigration policies. This finding is notable against the backdrop of increasingly negative frames of immigrants and immigration in the national news media (Abrajano et al., 2017; Farris & Silber Mohamed, 2018).

However, not all local community frames are equally powerful. Frames that focus on current immigrant population levels rather than recent changes in the immigrant population affect American’s policy views. This is an interesting finding, considering that several observational studies find individuals are more responsive to objective changes than current levels of immigrant populations (Hopkins, 2010; Newman, 2013; Newman & Velez, 2014). Nevertheless, our results align with the fact that most Americans can perceive standing levels of immigrant populations with greater accuracy than recent changes in these populations. This suggests that frames focused on current immigrant levels may be easier to interpret than those focused on change.

Finally, local community frames of immigration increase support for permissive policies among many subgroups of Americans, with varied perspectives on immigration. For example, the immigrant population levels treatment increased support for a pathway to citizenship among non-Whites, who generally express support for more permissive immigration policies than Whites (Masuoka & Junn, 2013). Perhaps more notably, the same treatment increased support for sanctuary cities among Republicans, a group that typically favors restrictive immigration policies (Neiman et al., 2006). This finding aligns with existing work, which finds that frames of immigration can alter the policy views across the partisan spectrum (Knoll et al., 2011). Just as negative frames of immigration can galvanize support for restrictive immigration polices among Democrats, neutral community frames of immigration can increase support for permissive policies among Republicans. Although we cannot definitively identify the mechanism driving this partisan effect, perhaps the immigrant levels treatment shifts thinking about immigration from an abstraction to a concrete personal experience related to immigrants in the community (Haynes et al., 2016).

This research may help to allay fears that exposure to immigrant populations uniformly causes backlash and reduces support for pro-immigrant policies. To the contrary, asking residents to think about the current levels of local immigrant populations increases support for several pro-immigrant outcomes. By using subjective measures of local immigrant population dynamics in an original survey experiment, this research adds nuance and several new insights to the literature about the impact of local immigrant populations on public opinion. Future work might focus on identifying the mechanisms driving changes in immigration policy views and testing whether these findings hold in externally valid contexts.

Supplemental Material

Supplemental Material - Perceived Local Population Dynamics and Immigration Policy Views

Supplemental Material for Perceived Local Population Dynamics and Immigration Policy Views by Stephanie Chan, Tanika Raychaudhuri, and Ali Valenzuela in American Politics Research

Supplemental Material

Supplemental Material - Perceived Local Population Dynamics and Immigration Policy Views

Supplemental Material for Perceived Local Population Dynamics and Immigration Policy Views by Stephanie Chan, Tanika Raychaudhuri, and Ali Valenzuela in American Politics Research

Footnotes

Acknowledgements

We gratefully acknowledge support from Princeton University's office of the Dean of the College. We are also grateful to Joseph DeMarco for assisting in formulating the experimental design and data collection. Additionally, we thank the anonymous reviewers who have provided feedback on the manuscript as well as participants of the Midwest Political Science Association and American Political Science Association conference sessions at which we presented earlier versions of this work.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Princeton University.

ORCID iDs

Stephanie Chan

Tanika Raychaudhuri

Ali A. Valenzuela

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Stephanie Chan is an Assistant Professor in the Department of Government & Law at Lafayette College.

Tanika Raychaudhuri is an Assistant Professor in the Department of Political Science at the University of Houston.

Ali Valenzuela is an Associate Professor in the Department of Government at American University.

References

Abrajano

Hajnal

Z. L.

(2015). White backlash: Immigration, race, and American politics. Princeton University Press.

Abrajano

M. A.

Hajnal

Hassell

H. J. G.

(2017). Media framing and partisan identity: The case of immigration coverage and white macropartisanship. Journal of Race, Ethnicity, and Politics, 2(1), 5–34. https://doi.org/10.1017/rep.2016.25

Alamillo

Haynes

Madrid

Jr. (2019). Framing and immigration through the trump era. Sociology Compass, 13(5), Article e12676. https://doi.org/10.1111/soc4.12676

Alba

Rumbaut

R. G.

Marotz

(2005). A distorted nation: Perceptions of racial/ethnic group sizes and attitudes toward immigrants and other minorities. Social Forces, 84(2), 901–919. https://doi.org/10.1353/sof.2006.0002

Allport

G. W.

([1954] 1988). The nature of prejudice. Addison-Wesley.

Baybeck

(2006). Sorting out the competing effects of racial context. The Journal of Politics, 68(2), 386–396. https://doi.org/10.1111/j.1468-2508.2006.00414.x

Baybeck

McClurg

S. D.

(2005). What do they know and how do they know it? An examination of citizen awareness of context. American Politics Research, 33(4), 492–520. https://doi.org/10.1177/1532673X04270934

Benoit

Watanabe

Wang

Nulty

Obeng

Müller

, et al (2018). quanteda: An R package for the quantitative analysis of textual data. Journal of Open Source Software, 3(30), 774. https://doi.org/10.21105/joss.00774

Campbell

A. L.

Wong

Citrin

(2006). “Racial threat”, partisan climate, and direct democracy: Contextual effects in three California initiatives. Political Behavior, 28(2), 129–150. https://doi.org/10.1007/s11109-006-9005-6

10.

Casellas

J. P.

Leal

D. L.

(2013). Partisanship or population? House and senate immigration votes in the 109th and 110th congresses. Politics, Groups, and Identities, 1(1), 48–65. https://doi.org/10.1080/21565503.2012.758588

11.

Cho

W. K. T.

Baer

(2011). Environmental determinants of racial attitudes redux: The critical decisions related to operationalizing context. American Politics Research, 39(2), 414–436. https://doi.org/10.1177/1532673X10377167

12.

Christ

Schmid

Lolliot

Swart

Stolle

Tausch

Ramiah

A. A.

Wagner

Vertovec

Hewstone

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

13.

Coulton

C. J.

Korbin

Chan

(2001). Mapping residents’ perceptions of neighborhood boundaries: A methodological note. American Journal of Community Psychology, 29(2), 371–383. https://doi.org/10.1023/A:1010303419034

14.

Daniller

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

15.

Dyck

J. J.

Pearson-Merkowitz

(2014). To know you is not necessarily to love you: The partisan mediators of intergroup contact. Political Behavior, 36(3), 553–580. https://doi.org/10.1007/s11109-013-9240-6

16.

Enos

R. D.

(2014). Causal effect of intergroup contact on exclusionary attitudes. Proceedings of the National Academy of Sciences, 111(10), 3699–3704. https://doi.org/10.1073/pnas.1404262111

17.

Enos

R. D.

(2016). What the demolition of public housing teaches us about the impact of racial threat on political behavior. American Journal of Political Science, 60(1), 123–142. https://doi.org/10.1111/ajps.12156

18.

Enos

R. D.

(2017). The space between us: Social geography and politics. Cambridge University Press.

19.

Farris

E. M.

Silber Mohamed

(2018). Picturing immigration: How the media criminalizes immigrants. Politics, Groups, and Identities, 6(4), 814–824. https://doi.org/10.1080/21565503.2018.1484375

20.

Gaines

B. J.

Kuklinski

J. H.

Quirk

P. J.

(2007). The logic of the survey experiment reexamined. Political Analysis, 15(1), 1–20. https://doi.org/10.1093/pan/mpl008

21.

Giles

M. W.

Hertz

(1994). Racial threat and partisan identification. The American Political Science Review, 88(2), 317–326. https://doi.org/10.2307/2944706

22.

Glasford

D. E.

Calcagno

(2012). The conflict of harmony: Intergroup contact, commonality and political solidarity between minority groups. Journal of Experimental Social Psychology, 48(1), 323–328. https://doi.org/10.1016/j.jesp.2011.10.001

23.

Goldstein

Healy

M. J. R.

(1995). The graphical presentation of a collection of means. Journal of the Royal Statistical Society: Series A (Statistics in Society), 158(1), 175–177.

24.

S. E.

(2010). The consequences of multiracial contexts on public attitudes toward immigration. Political Research Quarterly, 63(1), 29–42. https://doi.org/10.1177/1065912908325255

25.

Hawley

(2011). Political threat and immigration: Party identification, demographic context, and immigration policy preference*. Social Science Quarterly, 92(2), 404–422. https://doi.org/10.1111/j.1540-6237.2011.00775.x

26.

Haynes

Merolla

Ramakrishnan

S. K.

(2016). Framing immigrants: News coverage, public opinion, and policy. Russell Sage Foundation.

27.

Hero

(1998). Faces of inequality: Social diversity in American politics. Oxford University Press.

28.

Hewstone

Swart

(2011). Fifty-odd years of inter-group contact: From hypothesis to integrated theory. British Journal of Social Psychology, 50(3), 374–386. https://doi.org/10.1111/j.2044-8309.2011.02047.x

29.

Hood

M. V.

Morris

I. L.

(1997). ¿Amigo o enemigo? Context, attitudes, and anglo public opinion toward immigration. Social Science Quarterly, 78(2), 309–323.

30.

Hopkins

D. J.

(2010). Politicized places: Explaining where and when immigrants provoke local opposition. The American Political Science Review, 104(1), 40–60. https://doi.org/10.1017/s0003055409990360

31.

Hopkins

D. J.

Sides

Citrin

(2019). The muted consequences of correct information about immigration. The Journal of Politics, 81(1), 315–320. https://doi.org/10.1086/699914

32.

Key

V. O.

(1949). Southern politics: In state and nation. Vintage Books.

33.

Knoll

B. R.

Redlawsk

D. P.

Sanborn

(2011). Framing labels and immigration policy attitudes in the Iowa caucuses: “Trying to out-tancredo tancredo. Political Behavior, 33(3), 433–454. https://doi.org/10.1007/s11109-010-9141-x

34.

Macdonald

(2021). Political trust and support for immigration in the American mass public. British Journal of Political Science, 51(4), 1402–1420. https://doi.org/10.1017/s0007123419000668

35.

Mangum

(2019). Revisiting economic threat and cultural concerns: Public opinion toward immigration and non-citizens by race. Social Science Research, 83, 102309. https://doi.org/10.1016/j.ssresearch.2019.05.002.

36.

Mason

(2016). A cross-cutting calm: How social sorting drives affective polarization. Public Opinion Quarterly, 80(S1), 351–377. https://doi.org/10.1093/poq/nfw001

37.

Masuoka

Junn

(2013). The politics of belonging: Race, public opinion, and immigration. University of Chicago Press.

38.

McCarty

Poole

K. T.

Rosenthal

(2016). Polarized America second edition: The dance of ideology and unequal riches. MIT Press.

39.

Merolla

Ramakrishnan

S. K.

Haynes

(2013). “Illegal,” “undocumented,” or “unauthorized”: Equivalency frames, issue frames, and public opinion on immigration. Perspectives on Politics, 11(3), 789–807. https://doi.org/10.1017/S1537592713002077

40.

Nadeau

Niemi

R. G.

Levine

(1993). Innumeracy about minority populations. The Public Opinion Quarterly, 57(3), 332–347. https://doi.org/10.1086/269379

41.

Neiman

Johnson

Bowler

(2006). Partisanship and views about immigration in Southern California: Just how partisan is the issue of immigration? International Migration, 44(2), 35–56. https://doi.org/10.1111/j.1468-2435.2006.00363.x

42.

Newman

B. J.

(2013). Acculturating contexts and anglo opposition to immigration in the United States. American Journal of Political Science, 57(2), 374–390. https://doi.org/10.1111/j.1540-5907.2012.00632.x

43.

Newman

B. J.

Velez

(2014). Group size versus change? Assessing Americans’ perception of local immigration. Political Research Quarterly, 67(2), 293–303. https://doi.org/10.1177/1065912913517303

44.

Newman

B. J.

Velez

Hartman

T. K.

Bankert

(2015). Are citizens “receiving the treatment”? Assessing a key link in contextual theories of public opinion and political behavior. Political Psychology, 36(1), 123–131. https://doi.org/10.1111/pops.12069

45.

Newton

(2008). Illegal, alien, or immigrant: The politics of immigration reform. In Illegal, alien, or immigrant. New York University Press. https://doi.org/10.18574/nyu/9780814759028.001.0001

46.

Nteta

T. M.

Rice

(2021). Driving a Wedge? Republicans, immigration, and the impact of substantive appeals on african American vote choice. Political Research Quarterly, 74(1), 228–242. https://doi.org/10.1177/1065912919900012

47.

Oliver

J. E.

Wong

(2003). Intergroup prejudice in multiethnic Settings. American Journal of Political Science, 47(4), 567–582. https://doi.org/10.1111/1540-5907.00040

48.

Oskooii

K. A. R.

Dreier

S. K.

Collingwood

(2018). Partisan attitudes toward sanctuary cities: The asymmetrical effects of political knowledge. Politics & Policy, 46(6), 951–984. https://doi.org/10.1111/polp.12278

49.

Oxford Learner's

Dictionaries.

(2022) Municipality. Retrieved February 19, 2022 from https://www.oxfordlearnersdictionaries.com/us/definition/american_english/municipality

50.

Pearson-Merkowitz

Filindra

Dyck

J. J.

(2016). When partisans and minorities interact: Interpersonal contact, partisanship, and public opinion preferences on immigration policy. Social Science Quarterly, 97(2), 311–324. https://doi.org/10.1111/ssqu.12175

51.

Pettigrew

T. F.

(1997). Generalized intergroup contact effects on prejudice. Personality and Social Psychology Bulletin, 23(2), 173–185. https://doi.org/10.1177/0146167297232006

52.

Pew Research Center . (2020) Election 2020: Voters are highly engaged, but nearly half expect to have difficulties voting. Pew Research Center. Pew Research Center. https://www.pewresearch.org/politics/2020/08/13/election-2020-voters-are-highly-engaged-but-nearly-half-expect-to-have-difficulties-voting/

53.

Reich

Mendoza

A. A.

(2008). ‘Educating kids’ versus ‘coddling criminals’: Framing the debate over in-state tuition for undocumented students in Kansas. State Politics & Policy Quarterly, 8(2), 177–197. https://doi.org/10.1177/153244000800800204

54.

Reny

T. T.

Collingwood

Valenzuela

A. A.

(2019). Vote switching in the 2016 election: How racial and immigration attitudes, not economics, explain shifts in white voting. Public Opinion Quarterly, 83(1), 91–113. https://doi.org/10.1093/poq/nfz011

55.

Rocha

R. R.

Longoria

Wrinkle

R. D.

Knoll

B. R.

Polinard

J. L.

Wenzel

(2011). Ethnic context and immigration policy preferences among latinos and anglos*. Social Science Quarterly, 92(1), 1–19. https://doi.org/10.1111/j.1540-6237.2011.00754.x

56.

Schmid

Al Ramiah

Hewstone

(2014). Neighborhood ethnic diversity and trust: The role of intergroup contact and perceived threat. Psychological Science, 25(3), 665–674. https://doi.org/10.1177/0956797613508956

57.

Stein

R. M.

Post

S. S.

Rinden

A. L.

(2000). Reconciling context and contact effects on racial attitudes. Political Research Quarterly, 53(2), 285–303. https://doi.org/10.1177/106591290005300204

58.

Swire-Thompson

Ecker

U. K. H.

Lewandowsky

Berinsky

A. J.

(2020). They might be a liar but they’re my liar: Source evaluation and the prevalence of misinformation. Political Psychology, 41(1), 21–34. https://doi.org/10.1111/pops.12586

59.

Tausch

Hewstone

Kenworthy

J. B.

Psaltis

Schmid

Popan

J. R.

Cairns

Hughes

(2010). Secondary transfer effects of intergroup contact: Alternative accounts and underlying processes. Journal of Personality and Social Psychology, 99(2), 282–302. https://doi.org/10.1037/a0018553

60.

Taylor

M. C.

(1998). How white attitudes vary with the racial composition of local populations: Numbers count. American Sociological Review, 63(4), 512–535. https://doi.org/10.2307/2657265

61.

Tichenor

D. J.

(2009). Dividing lines: The politics of immigration control in America. Princeton University Press.

62.

Transue

J. E.

Lee

D. J.

Aldrich

J. H.

(2009). Treatment spillover effects across survey experiments. Political Analysis, 17(2), 143–161. https://doi.org/10.1093/pan/mpn012

63.

Velez

Y. R.

Wong

(2017). Assessing contextual measurement Strategies. The Journal of Politics, 79(3), 1084–1089. https://doi.org/10.1086/691281

64.

Wong

(2007). “Little” and “big” pictures in our heads: Race, local context, and innumeracy about racial groups in the United States. Public Opinion Quarterly, 71(3), 392–412. https://doi.org/10.1093/poq/nfm023

65.

Wong

Bowers

Williams

Simmons

K. D.

(2012). Bringing the person back in: Boundaries, perceptions, and the measurement of racial context. The Journal of Politics, 74(4), 1153–1170. https://doi.org/10.1017/S0022381612000552

Supplementary Material

Please find the following supplemental material available below.

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

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

3.39 MB

0.00 MB

1.00 MB