Not My Problem: How Social Identity Shapes Problem Perceptions and Policy Attitudes

Abstract

Social identity plays a central role in mass politics, shaping the perceptions citizens have of politically relevant phenomena. Does identity bias perceptions of social problems, leading citizens to show preferential concern for problems affecting their ingroup? If so, why? Most experimental research has not found evidence of such ingroup bias, but when it has, it has not distinguished empirically between ingroup favoritism or outgroup hostility, leaving open the question of whether identity biases people for their group or against outgroups. Also unclear is whether symbolic or self-interested motivations drive ingroup bias. Employing a variety of social identities and social problems, three survey experiments show citizens perceive problems affecting outgroup members as less serious and more strongly oppose government aid in those cases. Ingroup favoritism was not found because participants did not perceive ingroup victims as more similar than non-identified victims. Outgroup hostility was driven more by concerns stemming from self-interest than symbolic identity-based motivations.

Keywords

perceived problem seriousness ingroup bias social identity causal attribution policy attitudes self-interest

Political actors often aim to broaden the scope of concern about social problems (Schattschneider, 1960).¹ They want to raise the alarm about what is happening so that others will join them in support (Baumgartner and Jones, 2015). What makes citizens notice those alarms instead of ignoring them? As a strong force in political reasoning, social identity is one likely explanation (Huddy, 2013; Mason and Wronski, 2018). When citizens learn that people like them, people who share their social identity, are suffering, they may react with more sympathy (Conover, 1988; Simas et al., 2020). Given that social problems often afflict some groups more severely than others, bias against groups other than one’s own could hamper support for ameliorative government action.

If so, it is important to tease out why citizens seem to care more about ingroup problems. Two potential, though not mutually exclusive, processes may be at work. In one, as Social Identity Theory would predict, citizens may be driven by the desire to maintain a positive view of their group (Tajfel and Turner, 1986). To do so, they may attribute causes of problems that affect their group to forces outside their control, seeing them as victims rather than deviants. Being victims, they are thus more deserving of government aid (Iyengar, 1990; Weiner, 1995). In another process, self-interest may be the driver. Citizens may react more acutely to problems suffered by their ingroup because they fear they might soon face the same fate. A problem affecting similar people may make it loom larger, seem more plausible, or its harm more vivid. Shared group identity may trigger their self-interest more than it does their concern for the group as such (Fiorina, 2018; Weeden and Kurzban, 2017).

In addition to teasing out its mechanisms, it is important to ascertain what type of discrimination underlies ingroup bias in how citizens perceive social problems (e.g., Lelkes and Westwood, 2017). Is it ingroup favoritism (discrimination in favor one’s group), outgroup hostility (discrimination against outgroups), or both? If ingroup favoritism, then citizens prefer government do more to ensure the well-being of their group while leaving the well-being of outgroups untouched. Outgroup hostility, on the other hand, implies citizens are more interested in reducing outgroups’ well-being, perhaps because they see helping the outgroup as taking away resources that might go to their group (Dovidio and Gaertner, 2010). Each type of discrimination has different repercussions for democracy, as well as different implications for how to address it (Brewer, 2016).

I use three novel survey experiments to examine the presence, mechanisms, and type of ingroup bias in perceptions of social problems. Two of the experiments leverage salient politically charged identities (rural identity and partisan identity), while a third elicits what respondents considered their most personally important identity (from a preset list). The experiments exploit three different social problems (internet insecurity, air pollution, and cyber-attacks) on three different types of populations. Subjects exhibited significant ingroup bias—in the form of outgroup hostility—that appeared to be driven by self-interest as opposed to social identity. The absence of ingroup favoritism was caused by the surprising difficulty in using identity to make victims appear more similar. When perceiving social problems, it’s easier to make victims look different than similar.

This paper moves the literature forward in several ways. First, it tests two prominent mechanisms that may drive problem perception bias. It examines whether people care more about ingroup social problems because they are protecting the esteem of their groups, as social identity would expect, or because they are worried about their own well-being, as self-interest would predict. Second, it tests whether bias in the perception of social problems is people caring more about ingroup problems, less about outgroup problems, or both. Third, by placing emphasis on how people perceive problems that affect groups rather than specific individuals, it may help explain the mixed results of previous studies, as the results here are fairly consistent. Fourth, the finding that increasing perceived similarity is difficult to achieve in survey-experimental settings may help explain prior null results and chart a path forward in social identity research in political behavior.

Social Identity and Perceptions of Social Problems

Social problems are harms at scale. They can harm people’s physical, economic, or psychological well-being, and they often harm some people, and thus some groups, more than others. The question this paper asks is, do citizens perceive problems suffered by people with whom they share a social identity to be more serious? And if so, why?

Social identities seem to play an increasingly large role in mass politics in the United States, due in part to more and more people from the same social groups identifying with the same political parties (Levendusky, 2009; Mason, 2018). By bringing their social identities in line with their partisan identities, people have connected their identities with politics and the arena of conflict it evokes (Groenendyk and Krupnikov, 2021; Iyengar et al., 2019). One consequence of this politicization of identities may be a tendency to bring a mindset of conflict—a belief that what others get is necessarily taken from me—to perceptions of the social problems that animate politics. For example, partisans tend to see the same economy as improving or worsening depending on whether their party is in power (Ang et al., 2022; Bartels, 2002), a bias that has only grown in recent years (Brady et al., 2022). A similar pattern may extend to the perceptions citizens form of the many problems that afflict groups across society.

A great deal of work has tested this hypothesis in experiments that hold constant a problem, such as poverty (Feezell et al., 2021; Iyengar, 1990), homelessness (Gross and Wronski, 2021), unemployment (Iyengar and Kinder, 1987), opioid addiction, climate change (Hart and Nisbet, 2012), opioid addiction (de Benedictis-Kessner and Hankinson, 2024), religious extremism (Feezell et al., 2021), fear of deportation (Ostfeld and Mutz, 2014), and diabetes (Gollust and Lynch, 2011), while manipulating the identity of the individual suffering from it. Usually that identity is race, but sometimes it’s nationality, religion, culture, or rurality. In a recent study, de Benedictis-Kessner and Hankinson (2024) found that subjects were more supportive of a more lenient, patient-focused approach to opioid addiction when exposed to a victim who shared their race, though the effect did not extend to sharing their rural identity or gender. Some studies support this finding on race (Brader et al., 2008; Hannah and Cafferty, 2006; Harell et al., 2016) whereas others do not (Fang and Huber, 2020; Gollust and Lynch, 2011; Gross, 2008; Gross and Wronski, 2021). The studies that leveraged culture (Ostfeld and Mutz, 2014) and religion (Feezell et al., 2021) also found no effect, while the one that manipulated nationality did (Hart and Nisbet, 2012).

Given the prominence of social identity in politics and political science, this mixed record calls for further examination of how it may bias perceptions of social problems. Hart and Nisbet (2012) showed that, because Republicans perceived American farmers as more similar than French ones, their support for helping American farmers harmed by climate change was higher. Perceiving them as similar was part of the process of preferring to help the American victims. Though it may be safe to assume that white Americans perceive other whites as more similar than Blacks, none of the studies verify this fact, leaving open the possibility that, at least in some cases, the variable of interest (shared ingroup identity/perceived similarity) was not actually manipulated.

At least one other study, by Ostfeld and Mutz (2014), ensured that participants perceived one set of victims as more similar than the other. Surprisingly, however, it did not find similar-seeming victims were given more policy aid. This could be because the study focused on portrayals of specific individuals. Focusing on individuals rather than large groups may hinder people from considering the problem as large-scale and thus suitable for government policy. As a host of studies have found, support for policy is higher when problems are presented in “thematic” rather than “episodic” terms (Feezell et al., 2021; Iyengar, 1990).

Other studies present social problems thematically, but do so in terms of group, typically race-based, disparities. In these, ingroup biases surface more consistently. For instance, Rigby et al. (2009) found health disparities based on income and education were perceived as bigger problems than disparities based on race. When told COVID-19 disproportionately affected communities of color, Skinner-Dorkenoo et al. (2022) found whites reduced their support for safety precautions. These studies rely on problems described in relative terms, a “substantively important” difference from absolute terms (Gollust and Haselswerdt, 2023, 3). By casting problems as relative to another group, they evoke group conflict, thus changing the nature of the problem. They cannot test whether, had the victimized people been ingroup members, as opposed to outgroup members, would attitudes toward ameliorative policy been more positive? Since most existing evidence speaks to problems described as relative or portrayed through particular individuals, it is valuable to investigate perceptions of social problems described as large-scale phenomena in absolute terms.

In light of extant theory and previous research, I expect ingroup problems to be perceived and treated differently than outgroup problems. First, when a social problem harms people with whom citizens share a politically charged social identity, I expect them to perceive the problem as more serious.

Hypothesis 1 Ingroup problems are perceived as more serious than outgroup problems.

As a result, they should express stronger support for ameliorative public policy.

Hypothesis 2 Ingroup problems receive more support for policy aid than outgroup problems.

The Mechanisms of Ingroup Bias

Group Esteem

Why do people show preference for addressing problems that affect their group? It could be in part because they interpret their group members as less culpable for their problems. Social Identity Theory (Tajfel and Turner, 1986), an influential framework in psychology and increasingly so in political science (Huddy, 2001; Mason, 2023), posits that people strive to differentiate their groups from others and keep them in high esteem. This motivation is especially strong when there is competition over resources (whether material or political) (Dovidio and Gaertner, 2010). This may be why partisans tend to credit their party (but not opposing parties) for good societal outcomes and deflect blame for bad outcomes (Bisgaard, 2015; Zell et al., 2022). This tendency, known as intergroup attribution bias (Hewstone, 1990), is particularly likely when the causes of behavior are ambiguous or multi-faceted (Brewer, 2016).

Social problems are prime cases of ambiguous causation. They are large-scale phenomena involving complex interactions of personal choice, environment, and public policy, among others. Knowledge of their scale and impact on individuals is channeled through news reports, conversation, and personal experience. Such complexity leaves plenty of room for arriving at conclusions consistent with one’s ideological orientation or social identities. Indeed, the causal beliefs citizens hold regarding perennial problems such as poverty and unemployment are sharply divided along ideological and party lines, and they explain a lot of the variance in policy preferences (Gollust and Lynch, 2011; Iyengar, 1990). Causal beliefs are central to how people make sense of politics. Telling them that problems they think are caused by personal behavior are actually mostly shaped by environmental factors can even lead to them doubling-down on their prior positions (Gollust et al., 2009).

Moreover, problems that are thought to be caused by factors outside the victims’ control are perceived as more deserving of assistance (Weiner, 1995). Attributing the causes of social problems to external forces, in other words, grants the sufferers the status of being innocent victims worthy of help rather than perpetrators of their own misery (Conover, 1988; Schneider and Ingram, 1993). Citizens may, therefore, be motivated to attribute social problems affecting their ingroup to external forces, thus preserving their group’s good image. At the same token, because they wish to ensure the positive distinctiveness of their group, they may be more likely to attribute outgroups’ social problems to internal causes, such as inherent characteristics and personal choices. This pattern would be consistent with problem perception bias driven by social identity.

Hypothesis 3 Ingroup problems are attributed more to external forces than are outgroup problems.

Self-Interest

Alternatively, the social groups affected by a problem may signal to citizens how likely of a threat that problem is to them. A problem affecting ingroup members may seem more likely to strike them than one affecting outgroup members.

A vast literature has explored the extent to which citizens view politics through their personal interests as opposed to symbolically, in which “one’s political and personal lives exist largely isolated from one another” (Sears et al., 1980, 671). The lengthy debate has not resulted in a clear consensus, though recent work has begun to uncover a significant role, under certain conditions, for self-interested concern in the public’s political attitudes (Weeden and Kurzban, 2017).

This research has identified two conditions that encourage the role of self-interest. First, citizens must be aware of the stakes involved and the stakes should be fairly high (Chong et al., 2001). The higher the stakes, the more likely citizens are to incorporate what they see as being gained or lost in their decision-making. As Haselswerdt (2020) shows, citizens who were spared the negative consequences of a policy reform were less opposed to it than those who would feel the full brunt of it. Second, self-interest is more likely to impose itself when citizens understand the stakes and worry about being able to cope with the problem. Melcher (2023) demonstrates that people who thought they would soon lose their job did not necessarily have higher support for unemployment benefits. Instead, people who anticipated losing their job and were worried about getting a new one soon enough to make ends meet in the interim that showed higher support for aid to the unemployed. Losing a job is not harmful if you can quickly find a new one or you have a lot in savings.

Additionally, because imagining oneself in situations faced by similar others is cognitively easier, social problems harming similar others may appear to have clearer and higher consequences (Van Boven et al., 2013). The clarity of the consequences and the shorter social distance between oneself and ingroup members may cause the problem to appear likelier to happen to oneself than it would otherwise, thus raising personal concern.

This heightened personal concern suggests that people interpret what happens to ingroup members as a signal of what may happen to them. In this way, it bears some similarity to the linked fate heuristic (McClain et al., 2009). Conceptualized to explain the political behavior of Black Americans (Dawson, 1994), and extended to apply to other racial (Marsh and Ramírez, 2019) and non-racial (Donnelly, 2021) groups, it gauges how much people believe their life consequences are affected by those of the groups to which they belong. As more of a trait-like characteristic, linked fate is not the outcome measure here. Instead, I am testing whether perceptions of personal concern are why ingroup problems can appear more serious.

Hypothesis 4 Ingroup problems evoke more personal concern than outgroup problems.

Ingroup Favoritism vs. Outgroup Hostility

Whatever mechanisms drive ingroup bias, it is an open question whether that bias is better characterized as discrimination in favor of the ingroup (ingroup favoritism) or discrimination against the outgroup (outgroup hostility). Though early research in social psychology posited that the two were reciprocal—with greater ingroup favoritism meaning greater outgroup hostility and vice-versa—the two are empirically distinct (Brewer, 2016). In fact, the bulk of social psychological research suggests that ingroup favoritism underlies most ingroup bias, in part because ingroup love exists before outgroup hate is formed. First you become a member of a group, then you may discover there are other, potentially threatening groups out there. When the outgroup is perceived as a threat, “not only to the self but also to the integrity, interests, or identity of the ingroup as a whole”—as is often the case in politics—people tend to exhibit both forms of discrimination (Brewer, 2016, 92).

Research on partisan animosity is more supportive of ingroup favoritism. Lelkes and Westwood (2017) show that the more affectively polarized partisans are, the more they support actions that would benefit their party, rather than hurt the other (see also Westwood et al., 2019). Similarly, Barber and Davis (2022) find partisans grant the lives of co-partisans more moral value than those of out-partisans. In hypothetical “trolley problem” scenarios, people are far less willing to sacrifice a co-partisan for the sake of a group of out-partisans than they are an out-partisan in the same situation (Barber and Davis, 2022).

The effects of shared identity on perceptions of social problems may follow a similar pattern, but prior research on social problems has not been able to adjudicate between ingroup favoritism and outgroup hostility. This is because the experiments did not incorporate an “identity-free” scenario. Incorporating one provides a useful baseline from which to assess both ingroup and outgroup treatment, which Studies 2 and 3 do, below.²

Outgroup hostility may be tempered by general group empathy, as people higher in group empathy are more likely to support pro-outgroup policies (Sirin et al., 2021, ch. 8). Racial and ethnic minorities in particular report higher levels of group empathy, and all groups demonstrate that greater empathy with one’s ingroup is associated with more empathy for outgroups (Sirin et al., 2017, 439). On other hand, dispositional empathy can cut the other way, such that those higher in (general) empathic concern take more enjoyment in the injury of an out-partisan (Simas et al., 2020, 265).

Data and Methods

I analyze three survey experiments fielded in the United States, two conducted on nationally representative samples, the other on an online convenience sample. I will describe the experiments in turn and then report the results by hypothesis. For comparability, all dependent variables are scaled to range from 0 to 1.

Study 1: Rural Identity and Internet Insecurity

The first survey experiment was embedded in the 2021 Cooperative Election Study (N = 1,000), a nationally representative online survey conducted by YouGov. Participants were randomly assigned to read one of two characterizations of internet insecurity. In one, the problem was described as predominantly rural; in the other, as urban.

Rural identity is an ideal test case for ingroup bias in problem perception because it is a strongly held, competition-fueled identity (Cramer, 2016) that cuts across party (Munis, 2022). Ingroup bias is more likely to be expressed here, particularly for rural Americans, because what urban Americans get from government seems to come at the expense of their communities (Dovidio and Gaertner, 2010; Jacobs and Munis, 2023). Internet insecurity as a social problem offers an excellent case for study because it is not highly partisan, yet it is a national issue that affects people across the rural-urban divide, so it can plausibly be presented as either rural or urban (White House, 2021).

The experimental manipulation consisted of a question about a recent report. The question read:

“According to a report by the OECD, an international think tank, one in five American households lacks access to reliable high-speed broadband internet. The report found the vast majority of people dealing with such ‘internet insecurity’ live in [rural, sparsely populated areas/urban, densely populated areas], where broadband internet can be either unaffordable or not offered by internet service providers.”

Rural respondents were classified as those who said they lived in a rural area (21% of the sample), as opposed to a city, town, or suburb. Rurality is therefore based on self-identification rather than geographic location (Nemerever and Rogers, 2021), as recent work does (Jacobs and Munis, 2019).

Dependent Variables

To measure perceived problem seriousness, respondents were asked: “How harmful do you think not having reliable access to high-speed internet is?” Subjects were asked about “financial or economic harm (such as lost income or job opportunities).” This was measured on a 5-point labeled response scale anchored by “extremely” and “not at all harmful.”

Attributed causal responsibility was measured using an item that asked, after a short preamble, “Who do you think is responsible for people not having high-speed internet in their homes?” Respondents could answer on a 7-point scale from “consumers themselves” to “internet service providers (ISPs)” with a midpoint of “equally consumers and internet providers.” Answers on the “consumers themselves” half of the scale therefore reflect victim-blaming.³

Support for government policy was measured using a 7-point agree-disagree item that asked, “In general, do you support or oppose the U.S. government increasing regulations and/or spending more, even if it means higher taxes, to help provide people with reliable, affordable high-speed internet access in their homes?”

Personal concern was tapped by asking, “How concerned are you that you may soon not have access to reliable high-speed internet in your home?” Answers ranged from “not at all concerned” to “extremely concerned” on a 5-point scale.

Manipulation Check

To gauge whether respondents perceived the victims of internet insecurity as more similar in the Rural condition, they were asked, “How much do you agree with the following statement? “I identify with people who lack access to high-speed internet.”” They responded on a 7-point agree-disagree scale.

Study 2: Partisan Identity and Air Pollution

The second survey experiment was fielded in July 2022 on a sample of 2,200 Republican and Democratic identifiers on the survey platform Prolific. After giving informed consent, participants were asked a series of demographic questions and then read a two-paragraph news clipping. The clipping described a “harmful air pollutant” that had recently appeared in rural counties in Arizona. Details of the county were manipulated to signal the residents were either predominantly Democratic or Republican. As a 2020 toss-up state with both Republican- and Democratic-voting rural counties (e.g., Mohave and Coconino), Arizona was a plausible (and as the manipulation check shows, believable) choice for the setting.

Participants were randomly assigned to one of three conditions: one Republican, one Democratic, and one without mention of party. Participants whose partisanship matched that of the victims were coded as having receiving the Ingroup treatment; those with opposing partisanship were in the Outgroup condition. The remainder were in the No Party condition.⁴ The first paragraph read as follows:

“Reports of a harmful air pollutant have been rising in some rural communities in Arizona. Over the past several months, residents have made hundreds of complaints with local county offices. The office of Douglas Johns, a six-term [Republican/Democrat] and the top official in the county, which went decisively for [Donald Trump/Joe Biden] in 2020, has received many alarming calls. “Residents,” he said, “are calling in about an unusual odor and health problems. They’re getting painful headaches, nausea, and weakness that can debilitate them for hours. Sometimes they can’t go to work or their kids can’t go to school.” Longer-term effects of exposure to the pollutant are still unknown, but concerns are mounting about its increasing prevalence.”

Participants were asked, “Have you heard of this developing story?” They proceeded to the items for perceived problem seriousness, policy attitudes, causal attribution, perceived similarity, and personal concern.

Dependent Variables

Perceived problem seriousness was gauged using a 5-item scale with items such as, “How serious of a problem do you think the air pollutant is?” Responses ranged from zero (“not at all”) to 10 (“the most possible”).

Attributed causal responsibility was measured using an item that asked, “How responsible do you think residents of the community are for causing the harmful air pollutant?” Responses were from “not at all” to “entirely” on a 7-point scale.

Support for government policy was measured using the 7-point agree-disagree item that asked, “how much you support or oppose the government providing affected residents with effective air purifiers free of charge for their homes?”

To assess personal concern, respondents were asked, “How concerned are you about this type of harmful air pollution coming to your area?” Answers ranged from “not at all concerned” to “extremely concerned” on a 5-point scale.

Manipulation Check

Perceived similarity was comprised of a 4-item scale (Hart and Nisbet, 2012), using agreement with such items as, “The residents in the news clipping are like me.”

Study 3: “Primary” Identity and Cyber-Attacks

Study 3 centers on cyber-attacks in which small-business owners of a certain social group are targeted and suffer financial damages. The purpose of the study was threefold. First, to maximize the degree of victims’ perceived similarity to respondents, the victims— the small-business owners—were assigned identities that matched respondents’ “primary” identity. Respondents’ “primary” identity was operationalized as the identity to which they allocated the most points from a set of socially relevant categories, including gender, race, religion, rurality, and partisanship (Spry, 2021). For example, if a Black respondent put more points in the race category than any other, the victims would be described as Black.

Second, Study 3 added variety to the range of social problems studied here. In contrast to the previous studies, Study 3 focuses on a problem whose cause was intentional rather than ambiguous. Intentional harm should evoke moral feelings and thus provide a stronger test of ingroup bias (Schein and Gray, 2018). Third, Study 3 revisits party identity, specifically out-party identity, to test the robustness of out-party hostility found in Study 2.

The survey experiment was embedded in the 2023 Cooperative Election Study (N = 1,000). After completing the identity point-allocation task earlier in the survey, respondents read a paragraph about cyber-attacks affecting small businesses. They were randomly assigned to one of three conditions that varied the identities of the small-business owners. In the Primary Identity condition, the victims were given the respondents’ “primary” identity; in the Out-party condition, victims were out-party members; in the third No Identity condition, no social identity was mentioned. The paragraph read:

“More and more small businesses are being hit by cyber-attacks. Recent news reports suggest that small businesses owned by [primary identity/out-party] entrepreneurs are being targeted by a group of internet hackers. In many cases, the cyber-attacks have interrupted operations, caused major loss of revenue, and compromised customers’ trust. The identities of the hackers are unknown. Trade associations representing the businesses have called for official investigations, increased spending on cyber-security, and financial support from the government.”

Perceived problem seriousness was measured using an item asking how serious the problem was on an 11-point scale.

Support for government policy was captured by asking a question regarding federal spending on aid for small businesses hit by cyber-attacks. Responses ranged on a 5-point scale from “decrease a lot” to “increase a lot.”

To measure attributed causal responsibility, respondents were asked, “Would you agree or disagree that small businesses are to blame for not preparing enough and leaving themselves vulnerable to cyber-attacks?” They could respond on a 7-point scale from “strongly disagree” to “strongly agree.”

Personal concern was measured the same as in Study 1.

Manipulation Check

To assess perceived similarity, respondents were asked how much they agreed or disagreed on a 7-point scale with the statement: “The small businesspeople in the news report are like me.”

Estimation Strategy

Since citizens are hypothesized to react to social problems differently depending on the relation between their identity and those of the victims of the problem, the quantities of interest are conditional average treatment effects (CATEs). These estimates capture the average effect of the treatments conditional on respondents’ sharing a social identity (an Ingroup condition) or having opposed social identities (an Outgroup condition) to the victims relative to a No Group condition. To increase statistical precision, I estimate linear regression models with pretreatment covariates (Gerber and Green, 2012). To illustrate, the Study 1 models take the form,

Y_{i} = α + X γ + T_{i} β_{N o n ‐ r u r a l} + (T_{i} \times R_{i}) β_{R u r a l} + ϵ_{i}

where Y is the dependent variable, X is a vector of pre-treatment covariates, T indicates assignment into the Rural condition, R indicates a rural respondent, α is an intercept, γ is a vector of coefficients, and ϵ is a disturbance term. The estimand of interest is β_Rural, which represents the CATE of the Rural condition for rural respondents as opposed to non-rural respondents. The vector of pretreatment covariates for all studies includes age, gender, education, political interest, and religiosity. Study 1 also includes pre-treatment support for infrastructure spending.

Results

Manipulation Checks

In Study 1, rural respondents perceived the rural victims of internet insecurity as 10.5 percentage points more similar than the urban ones (p = 0.031).⁵ Non-rural respondents perceived neither as more similar. In Study 2, partisan respondents perceived out-partisans as less similar than non-partisans (3.7% points, p = 0.023), but they did not perceive co-partisans as more similar (p = 0.70). Following the same pattern, respondents in Study 3 perceived out-party victims as less similar compared to the No Identity baseline (p = 0.012), but they did not perceive victims who shared their “primary” identity as more similar (p = 0.676). The surprising failures of both in-party and “primary” identities to make victims appear more similar underscore a potentially under-appreciated difficulty in using social identity to elicit ingroup feeling in political vignettes.

Hypothesis 1: Perceived Problem Seriousness

Rural and non-rural respondents perceived the economic harm of internet insecurity differently depending on who the victims were. Figure 1 shows lines representing the CATEs among rural and non-rural respondents. The green line’s slope, representing the effect for rural respondents, is 8.7 percentage points more positive (SE = 0.044, t = 1.98, p = 0.048) than that for non-rural respondents. This pattern is consistent with ingroup bias in perceived harm, though it is not clear if it is because of ingroup favoritism or outgroup hostility.

Figure 1.

Rural respondents perceive internet insecurity does more harm to rural victims than urban. Note: Points are estimated values with 95% confidence intervals, based on coefficient estimates in Appendix Table A1, Model 2. Dependent variable is perceived economic harm from not having reliable high-speed internet, scaled 0 to 1. Green line is for rural respondents, gray is for non-rural.

Study 2 sheds some light on this question. To do so, I analyze average treatment effects conditional on party ID. Figure 2 plots estimated values and 95% CIs of perceived problem seriousness by condition separately for Democrats and Republicans. The flat blue line shows Democrats reacted similarly to all three conditions.

Figure 2.

Republicans perceive air pollution affecting democrats as less serious. Note: Points are estimated values with 95% confidence intervals, based on coefficient estimates in Appendix Table A2, Model 2. The dependent variable is perceived seriousness of the air pollution problem, scaled 0 to 1. Red line is for Republicans, blue is for Democrats.

The red line stays flat when going from the In-party to No Party conditions, indicating no bias in favor of co-partisans over non-partisans. Starting from the No Party condition, however, the line slants downward, as Republicans reported significantly lower perceived problem seriousness when the problem affected Democrats. Ingroup bias exerted itself as outgroup hostility rather than ingroup favoritism. The Republicans’ Out-party mean dropped 3.4 percentage points from the No Party mean (SE = 0.017, t = −1.98, p = 0.048). The size of the drop in perceived seriousness is not trivial. It is more than half as large (59%) as the average partisan difference in overall perceived problem seriousness (β = −0.058, p < 0.05).

Study 3 corroborates Study 2’s findings. As presented in Figure 3, respondents reported 4.6 percentage points less perceived problem seriousness in the Out-party condition as compared to the No Identity condition (SE = 0.019, t = −2.48, p = 0.01), an effect equal to 0.20 standard deviations in the dependent variable.

Figure 3.

Partisans perceive cyber-attacks affecting out-partisans as less serious. Note: Points are estimated values with 95% confidence intervals, based on coefficient estimates in Appendix Table A3, Model 2. Dependent variable is perceived seriousness of the cyber-attack problem, scaled 0 to 1.

Hypothesis 2: Policy Support

If subjects showed ingroup bias in perceived problem seriousness, we would expect them to exhibit similar bias in their support for government action. Indeed, we see the effects on policy attitudes mirror those on perceived problem seriousness. Looking to Figure 4, presenting the predicted means of support for providing victims with free home air filters, we see another positive slope for rural respondents and a flat one for non-rural ones. The effect is 9.6 percentage points larger for the rural participants (SE = 0.042, t = 2.30, p = 0.022), equivalent to 0.29 standard deviations.

Figure 4.

Rural respondents support Gov’t helping rural victims more than urban. Note: Points are estimated values with 95% confidence intervals, based on coefficient estimates in Appendix Table A4, Model 2. Dependent variable is support for increasing government spending and/or regulations to help provide more reliable internet, scaled 0 to 1. Green line is for rural respondents, gray is for non-rural.

Figure 5 shows again that in-party victims do not receive preferential policy support; instead, outgroup victims receive less policy support among both Republicans and Democrats. The pooled effect is −3.7 percentage points (SE = 0.013, t = −2.93, p = 0.003), a magnitude about half the size of the estimated difference between men and women of −8.7 percentage points and about 0.15 standard deviations in the outcome.

Figure 5.

Partisans reduce support for Gov’t helping out-partisans, don’t increase support for co-partisans. Note: Points are estimated values with 95% confidence intervals, based on coefficients estimates in Appenix Table A5, Model 2. Dependent variable is support for free air purifiers to victims, scaled 0 to 1. Red line is for Republicans, blue is for Democrats.

The pattern for Study 3 mirrors that of Study 2 (see Figure 6). Support for policy aid is significantly lower for out-partisans (β = −0.035, SE = 0.018, t = −1.93, p = 0.054) than a general population. Again, support for helping the ingroup is not significantly higher and is in fact close to being significantly lower (p = 0.115).

Figure 6.

Respondents drop support for Gov’t helping out-partisans, don’t increase support for “primary” ingroup. Note: Points are estimated values with 95% confidence intervals, based on coefficient estimates in Appendix Table A6, Model 2. Dependent variable is support for increasing federal spending on cyber-security for small businesses, scaled 0 to 1.

Hypothesis 3: Causal Attribution

If citizens’ problem perceptions are driven by motivations to protect group esteem, we would expect respondents to direct blame away from their ingroup members. I find no evidence of this. In Study 1, rural respondents did not attribute significantly more responsibility to ISPs in the Rural condition (β_Rural = 0.062, SE = 0.043, t = 1.42, p = 0.155). Partisans in Study 2 did not attribute significantly less responsibility to the co-partisan victims (β_Pooled = 0.009, SE = 0.017, t = 0.535, p = 0.593). The null effect was very similar for both Republicans and Democrats (β_Republican = −0.005, SE = 0.040, t = −0.127, p = 0.899). Study 3 respondents did not place significantly less blame on victims in the Primary Identity condition (β = −0.019, SE = 0.018, t = −1.070, p = 0.285) nor did they place significantly more blame in the Out-party condition (β = 0.015, SE = 0.018, t = 0.833, p = 0.405), as compared to the No Identity baseline.

Hypothesis 4: Personal Concern

Instead of group esteem, people may be motivated by concern for their own welfare. I find evidence consistent with this hypothesis. Beginning with Study 1, Figure 7 shows that both rural and non-rural respondents felt similarly concerned in the Urban condition, but then diverged in the Rural condition. Rural respondents became more personally concerned, a full 10.9 percentage points so (SE = 0.051, t = 2.13, p = 0.033), equal to 0.32 standard deviations. Non-rural respondents became marginally less concerned (β_Non−rural = −0.037, SE = 0.024, t = −1.58, p = 0.115).

Figure 7.

Rural respondents feel more personally concerned when problem is rural rather than urban. Note: Points are estimated values with 95% confidence intervals, based on coefficient estimates in Appendix Table A10, Model 2. Dependent variable is personal concern about the problem soon affecting the respondent, scaled 0 to 1. Green line is for rural respondents, gray is non-rural.

Starting on the lefthand side of Figure 8, we see that the lines representing Republicans and Democrats stays relatively flat between the In-Party and No Party conditions, suggesting no ingroup favoritism. When shifting over to the Out-party condition, however, personal concern among Republicans and Democrats diverges. Consistent with a self-interest explanation for their drops in perceived problem seriousness and support for government aid, Republicans’ personal concern falls 6.1 percentage points (SE = 0.037, t = −1.65, p = 0.098), equal to 0.20 standard deviations in personal concern.

Figure 8.

Republicans feel less personally concerned when problem affects democrats. Note: Points are estimated values with 95% confidence intervals, based on coefficients estimates in Appendix Table A11, Model 2. Dependent variable is personal concern about the problem soon affecting the respondent, scaled 0 to 1. Red line is for Republicans, blue is for Democrats.

Given that the cyber-attacks were targeted at small-business owners and not ordinary people, it may not be surprising that respondents’ were not less concerned they too would soon be hit by a cyber-attack (β = 0.017, SE = 0.022, t = 0.787, p = 0.431). This null effect shows that personal concern is one, but not the only, potential pathway for antipathy toward outgroup’s social problems.

Discussion and Conclusion

By leveraging politically charged social identities across three distinct social problems, focusing on perceptions of groups rather than individuals, and measuring two potential mechanisms, I present unique evidence of the extent to which, and how, social identity shapes citizens’ perceptions of social problems. When subjects perceived the people suffering as dissimilar to themselves, they judged their harms to be less serious. They then registered weaker support for government intervention to help. Thus, depending on who was suffering, subjects showed bias in what problems they cared about and how much they wanted the government to do. This held for rural respondents thinking about internet insecurity, and partisans learning about out-partisans breathing polluted air or having their businesses hacked by cyber-criminals.

The effect sizes for ingroup bias in perceived problem seriousness and policy attitudes were in the range of 0.15–0.32 standard deviations in the dependent variable. While these sizes are modest, they are indicative of a process that may produce larger effects in the real world. One reason is because of the tameness of the experimental stimuli, which only switched group labels. Real-world depictions of such problems, particularly those on partisan media sources, could more forcefully highlight the “otherness” of the outgroup or similarity of the ingroup (Lelkes et al., 2017). Discussion with likeminded others may also exacerbate the effects of such depictions on citizens’ attitudes (Druckman et al., 2018; Hobolt et al., 2024).

In contrast to the experimental literature on ingroup bias, this paper finds consistently significant effects. The reason for this consistency may be that respondents were shown problems that affected groups rather than specific individuals (Feezell et al., 2021). Study 1, for example, shows rural respondents perceived internet insecurity as more serious when cast as a rural problem. de Benedictis-Kessner and Hankinson (2024) show, in contrast, that rural respondents did not report greater willingness to help a rural victim of opioid addiction. The disparity may be because shared identity matters less when the victims are presented as individuals, as portrayals of specific people may evoke sympathy regardless of their social identity (Slovic, 2007). In emotionally cooler, thematic (group-based) presentations of problems, shared identity may have a larger effect.

The bias exhibited in Studies 2 and 3 is more consistent with outgroup hostility, not ingroup favoritism. Compared to those in the “identity-free” conditions, respondents in the outgroup scenarios were significantly less sympathetic and supportive of helping victims. In contrast, respondents who read about ingroup members were not more sympathetic and supportive of helping them.

While Studies 2 and 3 provide novel evidence of outgroup hostility in how citizens perceive problems and form policy attitudes, they do not disprove the existence of ingroup favoritism. This is because participants perceived neither co-partisans nor victims who shared a “primary” identity as more similar than identity-neutral victims. Thus, the effect found here is one of dissimilarity rather than similarity. Ingroup victims’ failure to elicit higher perceived similarity may have been because participants assumed the victims in the “identity-free” conditions were ingroup members as well. This seems unlikely, however, for two reasons. First, with regard to Study 2, if respondents had assumed the partisanship of the victims, they likely would have assumed they were Republican because they lived in a rural area. Democratic respondents would then likely have perceived the non-party-identified victims as less similar than the Democratic victims. But they did not. Moreover, Republicans and Democrats reacted the same—that is, not at all—to the non-party-identified victims as compared to their co-partisan victims (see Appendix Figure B2). Instead, this result suggests using co-partisanship as a marker of similarity may be more difficult than as a marker of dissimilarity. American partisans tend to assume the worst about opposing partisans (Ahler and Sood, 2018), while probably not making equally extreme glowing assumptions about their fellow partisans.

In Study 3, the failure to make the small businesspeople who shared the respondents’ “primary” identity appear more similar is even more surprising. Even respondents who selected one “dominant” identity, allocating at least half of their identity points to one social category, did not perceive their ingroup victims as more similar (p = 0.95). Ingroup favoritism may exist, but evoking the perceived similarity to observe it was not as easy as expected.

While previous studies have uncovered similar bias in political attitude formation, particularly in the context of race, none has simultaneously tested the contributions of social identity and self-interest. A social identity-driven approach would anticipate a process by which citizens strive to maintain the esteem of themselves and their group (Tajfel and Turner, 1986). When perceiving a social problem involving their group, they would pin the blame outside their group (Bisgaard, 2015; Zell et al., 2022). However, respondents here did not do this, in part because they did not perceive the ingroup as more similar. Instead, seeing the outgroup made them less fearful that the problem would soon be at the their doorstep. It seems outgroup identity de-activated a sense of shared fate. Indeed, the results of Study 2 suggest people may be motivated by unshared fate: when respondents saw that the victims were out-partisans, they felt much less personally concerned. This perception had downstream effects. The problem seemed confined to hurting “them,” not “us,” so it seemed less serious and less worthy of government aid. This finding casts the citizen as someone who asks herself, “how will this affect me?” rather than “how will this make my group look?”

Study 2 suggests the effect of partisan ingroup bias may be stronger for Republicans than for Democrats, but Democrats exhibited it too. Republicans in Study 2 clearly expressed stronger and more consistent ingroup bias—across perceived problem seriousness, policy support, and personal concern—than Democrats did. However, bias was not exclusive to Republicans, as Democrats’ support for policy aid dropped when it came to helping Republican victims. Furthermore, the rural identifiers from Study 1 (a nationally representative sample), are nearly equal parts Democratic and Republican: 36% identify as Democrats and 40% as Republican, suggesting bias by rural identifiers of both parties (Munis, 2022). In Study 3, there were no statistically significant differences between partisans. Thus, while the tendency to meet out-partisans’ problems with disregard seems stronger for Republicans, it is not absent among Democrats. It may be stronger for Republicans because they tend to dislike Democrats more than Democrats dislike them (Iyengar et al., 2012) and because conservatives tend to prioritize harm reduction less than liberals do (Graham et al., 2009). This suggests that moral values of equality and fairness may moderate the extent to which social identity biases problem perceptions.

Self-interest took the stage in Studies 1 and 2; group esteem was not cast a role. Participants did not seek to make their group look better or make the outgroup look worse by attributing responsibility for the problem differently. This is surprising, because maintaining group esteem is a core concern for social identity theory (Tajfel and Turner, 1986). It’s also surprising because causal attributions are central to whether victim populations are considered deserving of help (Conover, 1988; Petersen, 2012). Instead, the reason people granted less seriousness and fewer government resources to outgroups’ problems was because they felt less personally threatened. That process gels with research showing that individuals prioritize concerns of the personal self above those of the social self (for a review, see Sedikides et al., 2013).

The insights provided by this paper are not without their limitations. First, because no group perceived any victims as more similar than the neutral group, tests of ingroup favoritism could not be conducted. Future work should pursue this question. Second, neither the social problems nor the social identities explored here are exhaustive or representative of their variety. While efforts were made to employ social problems with ambiguous and unambiguous causes with physical and economic effects, future studies could leverage other social problems that vary along these dimensions as well as others. Similarly, exploiting rural and partisan identity constitutes only a fraction of the possible identities citizens possess. Efforts were made toward expanding the variety of identities by eliciting respondents’ “primary” identity, but the set of possible identities was limited to gender, race, religion, party, and rurality. Future research should explore other identities, with an eye toward measuring how competitive or otherwise they feel them to be.

Third, all the data were collected in the United States, a nation with relatively strong values for impersonal fairness and relatively weak ingroup biases (Henrich, 2020). Had data been collected from a variety of countries, especially non-Western ones, I would have expected to see stronger effects of ingroup bias. This pattern is suggested by the difference between Democrats and Republicans, as conservatives hold more traditional values of group loyalty and respect for the authority (Waytz et al., 2019), which are more common in non-Western populations. This paper thus shows only a sliver of the variation across nations.

Supplemental Material

Supplemental Material - Not My Problem: How Social Identity Shapes Problem Perceptions and Policy Attitudes

Supplemental Material for Not My Problem: How Social Identity Shapes Problem Perceptions and Policy Attitudes by Philip Moniz in Political Research Quarterly.

Footnotes

Acknowledgments

I would like to thank Amber Boydstun, Jacob Brown, Christopher Bryan, Nicholas Dias, Jamie Druckman, Stephen Jessee, Robert Luskin, Daron Shaw, Bill Swann, and attendees of the 2022 Toronto Political Behavior Workshop for helpful feedback.

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 disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received financial support from the Texas Politics Project at the University of Texas at Austin for the research of this article.

ORCID iD

Philip Moniz

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Ahler

Douglas J.

Sood

Gaurav

. 2018. “The Parties in Our Heads: Misperceptions about Party Composition and Their Consequences.” The Journal of Politics 80 (3): 964–81.

Ang

Zoe

Reeves

Andrew

Rogowski

Jon C.

, et al. 2022. “Partisanship, Economic Assessments, and Presidential Accountability.” American Journal of Political Science 66 (2): 468–84.

Barber

Michael

Davis

Ryan

. 2022. “Partisanship and the Trolley Problem: Partisan Willingness to Sacrifice Members of the Other Party.” Research & Politics 9 (4): 205316802211371.

Bartels

Larry M.

2002. “Beyond the Running Tally: Partisan Bias in Political Perceptions.” Political Behavior 24 (2): 117–50.

Baumgartner

Frank R.

Jones

Bryan D.

. 2015. The Politics of Information: Problem Definition and the Course of Public Policy in America. University of Chicago Press.

Bisgaard

Martin

. 2015. “Bias Will Find a Way: Economic Perceptions, Attributions of Blame, and Partisan-Motivated Reasoning during Crisis.” The Journal of Politics 77 (3): 849–60.

Brader

Ted

Valentino

Nicholas A.

Suhay

Elizabeth

. 2008. “What Triggers Public Opposition to Immigration? Anxiety, Group Cues, and Immigration Threat.” American Journal of Political Science 52 (4): 959–78.

Brady

David W.

Ferejohn

John A.

Parker

Brett

. 2022. “Cognitive Political Economy: A Growing Partisan Divide in Economic Perceptions.” American Politics Research 50 (1): 3–16.

Brewer

Marilynn B

. 2016. “Intergroup Discrimination: Ingroup Love or Outgroup Hate?” In The Cambridge Handbook of the Psychology of Prejudice, edited by Sibley

Chris G.

Barlow

Fiona Kate

. 1st ed., 90–110. Cambridge University Press.

10.

Chong

Dennis

Citrin

Jack

Conley

Patricia

. 2001. “When Self-Interest Matters.” Political Psychology 22 (3): 541–70.

11.

Conover

Pamela Johnston

. 1988. “The Role of Social Groups in Political Thinking.” British Journal of Political Science 18 (1): 51–76.

12.

Cramer

Katherine

. 2016. The Politics of Resentment: Rural Consciousness in Wisconsin and the Rise of Scott Walker. University of Chicago Press.

13.

Dawson

Michael

. 1994. Behind the Mule. Princeton University Press.

14.

de Benedictis-Kessner

Justin

Hankinson

Michael

. 2024. “How the Identity of Substance Users Shapes Public Opinion on Opioid Policy.” Political Behavior 46 (1): 609–29.

15.

Donnelly

Michael J

. 2021. “Material Interests, Identity and Linked Fate in Three Countries.” British Journal of Political Science 51 (3): 1119–37.

16.

Dovidio

John F.

Gaertner

Samuel L.

. 2010. “Intergroup Bias.” In Handbook of Social Psychology, edited by Fiske

Susan T.

Gilbert

Daniel T.

Lindzey

Gardner

. 5th ed, Vol. 2, 1084–1121. Wiley.

17.

Druckman

James N.

Levendusky

Matthew S.

Audrey

McLain.

. 2018. “No Need to Watch: How the Effects of Partisan Media Can Spread via Interpersonal Discussions.” American Journal of Political Science 62 (1): 99–112.

18.

Fang

Albert H.

Huber

Gregory A.

. 2020. “Perceptions of Deservingness and the Politicization of Social Insurance: Evidence from Disability Insurance in the United States.” American Politics Research 48 (5): 543–59.

19.

Feezell

Jessica T.

Glazier

Rebecca A.

Boydstun

Amber E.

. 2021. “Framing, Identity, and Responsibility: Do Episodic vs. Thematic Framing Effects Vary by Target Population?” Politics, Groups, and Identities 9 (2): 347–68.

20.

Fiorina

Morris P.

2018. “Identities for Realists.” Critical Review 30 (1-2): 49–56.

21.

Gerber

Alan S.

Green

Donald P.

. 2012. Field Experiments: Design, Analysis, and Interpretation. W. W. Norton.

22.

Gollust

Sarah E.

Haselswerdt

Jake

. 2023. “Who Does COVID-19 Hurt Most? Perceptions of Unequal Impact and Political Implications.” Social Science & Medicine 323: 115825.

23.

Gollust

Sarah E.

Lynch

Julia

. 2011. “Who Deserves Health Care? the Effects of Causal Attributions and Group Cues on Public Attitudes about Responsibility for Health Care Costs.” Journal of Health Politics, Policy and Law 36 (6): 1061–95.

24.

Gollust

Sarah E.

Lantz

Paula M.

Ubel

Peter A.

. 2009. “The Polarizing Effect of News Media Messages about the Social Determinants of Health.” American Journal of Public Health 99 (12): 2160–67.

25.

Graham

Jesse

Haidt

Jonathan

Nosek

Brian A.

. 2009. “Liberals and Conservatives Rely on Different Sets of Moral Foundations.” Journal of Personality and Social Psychology 96 (5): 1029–46.

26.

Groenendyk

Eric

Krupnikov

Yanna

. 2021. “What Motivates Reasoning? A Theory of Goal-dependent Political Evaluation.” American Journal of Political Science 65 (1): 180–96.

27.

Gross

Kimberly

. 2008. “Framing Persuasive Appeals: Episodic and Thematic Framing, Emotional Response, and Policy Opinion.” Political Psychology 29 (2): 169–92.

28.

Gross

Kimberly

Wronski

Julie

. 2021. “Helping the Homeless: The Role of Empathy, Race and Deservingness in Motivating Policy Support and Charitable Giving.” Political Behavior 43 (2): 585–613.

29.

Hannah

Gordon

Cafferty

Thomas P.

. 2006. “Attribute and Responsibility Framing Effects in Television News Coverage of Poverty.” Journal of Applied Social Psychology 36 (12): 2993–3014.

30.

Harell

Allison

Soroka

Stuart

Iyengar

Shanto

. 2016. “Race, Prejudice and Attitudes toward Redistribution: A Comparative Experimental Approach.” European Journal of Political Research 55 (4): 723–44.

31.

Hart

P. Sol

Nisbet

Erik C.

. 2012. “Boomerang Effects in Science Communication: How Motivated Reasoning and Identity Cues Amplify Opinion Polarization about Climate Mitigation Policies.” Communication Research 39 (6): 701–23.

32.

Haselswerdt

Jake

. 2020. “Carving Out: Isolating the True Effect of Self-Interest on Policy Attitudes.” American Political Science Review 114 (4): 1103–16.

33.

Henrich

Joseph

. 2020. The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous. Farrar, Straus and Giroux.

34.

Hewstone

Miles

. 1990. “The ‘Ultimate Attribution Error’? A Review of the Literature on Intergroup Causal Attribution.” European Journal of Social Psychology 20 (4): 311–35.

35.

Hobolt

Sara B.

Lawall

Katharina

Tilley

James

. 2024. “The Polarizing Effect of Partisan Echo Chambers.” American Political Science Review 118 (3): 1464–79.

36.

Huddy

Leonie

. 2001. “From Social to Political Identity: A Critical Examination of Social Identity Theory.” Political Psychology 22 (1): 127–56.

37.

Huddy

Leonie

. 2013. “From Group Identity to Political Cohesion and Commitment.” In The Oxford Handbook of Political Psychology, edited by Sears

David O.

Huddy

Leonie

Levy

Jack S.

. 2nd ed., 737–73. Oxford University Press.

38.

Iyengar

Shanto

. 1990. “Framing Responsibility for Political Issues: The Case of Poverty.” Political Behavior 12 (1): 19–40.

39.

Iyengar

Shanto

Kinder

Donald R.

. 1987. News that Matters. Chicago University Press.

40.

Iyengar

Shanto

Westwood

Sean J.

. 2015. “Fear and Loathing across Party Lines: New Evidence on Group Polarization.” American Journal of Political Science 59 (3): 690–707.

41.

Iyengar

Shanto

Sood

Gaurav

Lelkes

Yphtach

. 2012. “Affect, Not Ideology: A Social Identity Perspective on Polarization.” Public Opinion Quarterly 76 (3): 405–31.

42.

Iyengar

Shanto

Lelkes

Yphtach

Levendusky

Matthew

, et al. 2019. “The Origins and Consequences of Affective Polarization in the United States.” Annual Review of Political Science 22 (1): 129–46.

43.

Jacobs

Nicholas F.

Munis

B. Kal

. 2019. “Place-Based Imagery and Voter Evaluations: Experimental Evidence on the Politics of Place.” Political Research Quarterly 72 (2): 263–77.

44.

Jacobs

Nicholas

Munis

B. Kal

. 2023. “Place-Based Resentment in Contemporary U.S. Elections: The Individual Sources of America’s Urban-Rural Divide.” Political Research Quarterly 76 (3): 1102–18.

45.

Lelkes

Yphtach

Westwood

Sean J.

. 2017. “The Limits of Partisan Prejudice.” The Journal of Politics 79 (2): 485–501.

46.

Lelkes

Yphtach

Sood

Gaurav

Iyengar

Shanto

. 2017. “The Hostile Audience: The Effect of Access to Broadband Internet on Partisan Affect.” American Journal of Political Science 61 (1): 5–20.

47.

Levendusky

Matthew S.

2009. The Partisan Sort: How Liberals Became Democrats and Conservatives Became Republicans. Chicago University Press.

48.

Marsh

Wayde Z. C.

Ramírez

Ricardo

. 2019. “Unlinking Fate? Discrimination, Group- Consciousness, and Political Participation Among Latinos and Whites.” Politics, Groups, and Identities 7 (3): 625–41.

49.

Mason

Lilliana

. 2018. Uncivil Agreement: How Politics Became Our Identity. Chicago University Press.

50.

Mason

Lilliana

. 2023. “Political Identities.” In The Oxford Handbook of Political Psychology, edited by Sears

David O.

Huddy

Leonie

Levy

Jack S.

. 3rd ed., 886–917. Oxford University Press.

51.

Mason

Lilliana

Wronski

Julie

. 2018. “One Tribe to Bind Them All: How Our Social Group Attachments Strengthen Partisanship.” Political Psychology 39 (S1): 257–77.

52.

McClain

Paula D.

Johnson Carew

Jessica D.

Walton

Eugene

, et al. 2009. “Group Membership, Group Identity, and Group Consciousness: Measures of Racial Identity in American Politics?” Annual Review of Political Science 12 (1): 471–85.

53.

Melcher

Cody R.

2023. “Economic Self-Interest and Americans’ Redistributive, Class, and Racial Attitudes: The Case of Economic Insecurity.” Political Behavior 45 (1): 175–96.

54.

Munis

B. Kal

. 2022. “Us over Here versus Them over There…Literally: Measuring Place Resentment in American Politics.” Political Behavior 44 (3): 1057–78.

55.

Nemerever

Zoe

Rogers

Melissa

. 2021. “Measuring the Rural Continuum in Political Science.” Political Analysis 29 (3): 267–86.

56.

Ostfeld

Mara

Mutz

Diana

. 2014. “Revisiting the Effects of Case Reports in the News.” Political Communication 31 (1): 53–72.

57.

Petersen

Michael Bang

. 2012. “Social Welfare as Small-Scale Help: Evolutionary Psychology and the Deservingness Heuristic.” American Journal of Political Science 56 (1): 1–16.

58.

Rigby

Elizabeth

Soss

Joe

Booske

Bridget C.

, et al. 2009. “Public Responses to Health Disparities: How Group Cues Influence Support for Government Intervention.” Social Science Quarterly 90 (5): 1321–40.

59.

Schattschneider

E. E.

1960. The Semisovereign People: A Realist’s View of Democracy in America. Wadsworth.

60.

Schein

Chelsea

Gray

Kurt

. 2018. “The Theory of Dyadic Morality: Reinventing Moral Judgment by Redefining Harm.” Personality and Social Psychology Review 22 (1): 32–70.

61.

Schneider

Anne

Ingram

Helen

. 1993. “Social Construction of Target Populations: Implications for Politics and Policy.” American Political Science Review 87 (2): 334–47.

62.

Sears

David O.

Lau

Richard R.

Tyler

Tom R.

, et al. 1980. “Self-Interest vs. Symbolic Politics in Policy Attitudes and Presidential Voting.” American Political Science Review 74 (3): 670–84.

63.

Sedikides

Constantine

Gaertner

Lowell

Luke

Michelle A.

, et al. 2013. “A Three-Tier Hierarchy of Self-Potency: Individual Self, Relational Self, Collective Self.” Advances in Experimental Social Psychology 48: 235–95.

64.

Simas

Elizabeth N.

Scott

Clifford

Kirkland

Justin H.

. 2020. “How Empathic Concern Fuels Political Polarization.” American Political Science Review 114 (1): 258–69.

65.

Sirin

Cigdem V.

Valentino

Nicholas A.

Villalobos

Jos´e D.

. 2017. “The Social Causes and Political Consequences of Group Empathy.” Political Psychology 38 (3): 427–48.

66.

Sirin

Cigdem V.

Valentino

Nicholas A.

Villalobos

Jos´e D.

. 2021. Seeing Us in Them: Social Divisions and the Politics of Group Empathy. Cambridge University Press.

67.

Skinner-Dorkenoo

Allison L.

Sarmal

Apoorva

Rogbeer

Kasheena G.

, et al. 2022. “Highlighting COVID-19 Racial Disparities Can Reduce Support for Safety Precautions Among White U.S. Residents.” Social Science & Medicine 301: 114951.

68.

Slovic

Paul

. 2007. ““If I Look at the Mass I Will Never Act”: Psychic Numbing and Genocide.” Judgment and Decision Making 2 (2): 17–95.

69.

Spry

Amber D

. 2021. “Identity Experiments.” In Advances in Experimental Political Science, edited by Druckman

James N.

Green

Donald P.

, 431–46. Cambridge University Press.

70.

Tajfel

Henri

Turner

John C.

. 1986. “The Social Identity Theory of Intergroup Behavior.” In Psychology of Intergroup Relations, edited by Worchel

Stephen

Austin

William G.

, 7–24. Nelson-Hall.

71.

Van Boven

Leaf

Loewenstein

George

Dunning

David

, et al. 2013. “Changing Places: A Dual Judgment Model of Empathy Gaps in Emotional Perspective Taking.” Advances in Experimental Social Psychology 48: 117–71.

72.

Waytz

Adam

Iyer

Ravi

Young

Liane

, et al. 2019. “Ideological Differences in the Expanse of the Moral Circle.” Nature Communications 10 (1): 4389.

73.

Weeden

Jason

Kurzban

Robert

. 2017. “Self-Interest Is Often a Major Determinant of Issue Attitudes.” Political Psychology 38 (1): 67–90.

74.

Weiner

Bernard

. 1995. Judgments of Responsibility: Foundations for a Theory of Social Conduct. Guilford Press.

75.

Westwood

Sean J.

Peterson

Erik

Lelkes

Yphtach

. 2019. “Are There Still Limits on Partisan Prejudice?” Public Opinion Quarterly 83 (3): 584–97.

76.

White House . 2021. The Infrastructure Investment and Jobs Act Will Deliver for Massachusetts. U.S. Department of Transportation. https://bidenwhitehouse.archives.gov/wp-content/uploads/2021/08/MASSACHUSETTS_Infrastructure-Investment-and-Jobs-Act-State-Fact-Sheet.pdf.

77.

Zell

Ethan

Stockus

Christopher A.

Bernstein

Michael J.

. 2022. “It’s Their Fault: Partisan Attribution Bias and its Association with Voting Intentions.” Group Processes & Intergroup Relations 25 (4): 1139–56.

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.37 MB