When Interest Doesn’t Turn into Action: Discrimination,Group Identification,and Muslim Political Engagement in the Post-9/11 Era

Abstract

This article examines the effect of exposure to post-9/11 stigmatization on various types of Muslim political engagement, using a mixed-methods approach that combines propensity score matching analysis of data from the Muslims in the American Public Square (MAPS) survey administered immediately after 9/11 with experimental data of the U.S. Muslim population. I find that increased discrimination results in increased political interest but has a neutral or dampening effect on political participation. I use experimental data to argue that group identification acts as a partial mediator in this relationship, especially in accounting for the increase in political interest in response to discrimination.

Keywords

discrimination ethnicity identity immigration political participation

Throughout American history, various minority groups have gained, shed, and regained a stigmatized status. While some groups, such as African Americans, have faced consistent stigmatization, others, particularly immigrant groups, experience discrimination in cyclical waves, with one group becoming stigmatized as another becomes normalized. This stigmatization often results from the majority population feeling threatened. In the early to mid–nineteenth century, for example, Italian Americans faced a great deal of animosity from people who saw them as socially inferior, condemned their Catholicism, and were angered by perceived job loss on the part of the white majority (LaGumina 1999). Similarly, during World War II, and particularly after the attacks on Pearl Harbor in 1941, Japanese Americans (citizens and immigrants alike) were subject to severe discrimination in their communities and detained at internment camps throughout the country in mass numbers (Robinson 2009). These waves of stigmatization are not only the result of feeling threatened but have also often been triggered by a specific event, as in the case of Pearl Harbor and Japanese Americans in the mid–twentieth century or Irish Americans who began to immigrate to the United States in large numbers in the wake of the Irish potato famine in the mid–nineteenth century.

For almost two decades, Muslims, particularly those from countries in the Middle East, have occupied the role of a stigmatized outsider group in the United States.¹ Immediately after the attacks in 2001, anti-Muslim sentiment and discrimination surged in the United States. Muslims began to face hate crimes, racial slurs, verbal harassment, employment discrimination, and profiling by law enforcement officers at unprecedented rates (Project MAPS 2004; Senzai 2012). The FBI’s annual hate crime report, which details incidents involving violence, threats, vandalism, and arson, shows, for example, that immediately after 9/11, there was a 1,600 percent increase in anti-Muslim hate crimes and that the Department of Justice investigated more than 800 such crimes between 2001 and 2007, as shown in Figure 1.² Though this initial increase has dissipated, the annual number of hate crimes against Muslims in the United States has remained much higher than what it was in 2000, and research has shown a causal link between 9/11 and this surge (Disha, Cavendish, and King 2011).³ In recent years, anti-Muslim sentiment has been renewed.

Figure 1.

Number of Hate Crimes Motivated by Anti-Muslim Bias

After terrorist incidents in San Bernardino, Paris, and Orlando, anti-Muslim sentiment pervaded the media, and the number of suspected hate crimes skyrocketed, almost tripling in number (Levin 2016). Women wearing hijabs faced assaults, and news agencies reported numerous incidents of vandalism at mosques around the country (Stack 2015).

Politicians and presidential candidates have also taken public stances against Islam, as when President Trump proposed to ban all Muslims from the country early in his campaign, eventually ordering a travel ban that barred immigrants from a number of majority Muslim countries from entering the country (Diamond 2015; Haberman and Parker 2016). In the aftermath of the 2016 presidential election, news outlets reported surges in anti-Muslim hate crimes, with women being attacked for wearing hijabs in public and others facing verbal assaults and threats.⁴ The Southern Poverty Law Center reported on hundreds of incidents of “hateful harassment and intimidation” in the week right after the election, with those directed against Muslims being one of the most common.⁵

The Muslim American case provides a unique opportunity to examine how facing stigmatization or discrimination impacts political engagement. This study considers the political response to perceiving that one is the victim of discrimination or that one’s group is stigmatized using a mixed-methods approach, combining causal survey analysis that uses propensity-score matching techniques with a series of online survey experiments. I focus on the following questions:

Research Question 1: Does perceiving discrimination against one’s group (or oneself as a result of group membership) impact political engagement?

Research Question 2: Is this relationship mediated by group identification or the extent to which one identifies as a Muslim and feels positively toward other Muslims?

Background

It is not immediately clear whether we should expect perceived discrimination to motivate political engagement, prompting Muslims to unite and mobilize to action, or decrease engagement, alienating Muslims from the political process. Existing research shows mixed results (e.g., DeSipio 2002; Fleischmann, Phalet, and Klein 2011; Mattila and Papageorgiou 2017; Miller et al. 1981; Schildkraut 2004). On one hand, a great deal of research has supported the former hypothesis, highlighting that discrimination indirectly impacts political engagement by increasing group identification as Muslims become motivated to improve their group’s position in American society (Fleischmann et al. 2011; Miller et al. 1981). On the other hand, as rates of discrimination and stigmatization increase, Muslims may withdraw from the political system, feeling afraid to participate or feeling that their participation would be ineffectual. In the following section, I discuss the literature that supports each of these hypotheses, respectively.

The strand of research that supports the first view, that we should expect discrimination to indirectly increase political engagement, is based largely on work on social identity theory (SIT; Tajfel and Turner 1979). Fundamental to SIT is the belief that individuals want to maintain a positive self-concept, resulting from making positive comparisons between their ingroup and other outgroups. How are these comparisons made? Building off of Charles Cooley’s (1983) looking-glass self, we know that people internalize others’ views about their group as part of a dynamic process, and this internalization can lead to a diminished self-image, thereby threatening social identity (Dion, Dion, and Pak 1992). In the years immediately after 9/11, maintaining a positive self-concept was likely difficult for Muslims as rampant anti-Muslim sentiment resulted in the creation of a racialized and negative “Muslim” identity (Peek 2005). The word Muslim came to refer to a homogenized group in the American psyche, one that lumped together all people of Arab and Middle Eastern descent who were understood as being potentially threatening to the public (Bakalian and Bozorgmehr 2009). Because, as SIT highlights, individuals have an interest in maximizing their sense of self-worth, feeling that a group with which they identify has been stigmatized or devalued in this way may cause psychological harm.

In response to this threat, people employ various coping strategies. For example, some individuals may try to “pass,” that is, to change or hide their stigmatized attribute. This can be particularly challenging in the case of a visible stigma or one that, as is the case in the Muslim context, is based on an unalterable part of one’s identity. One’s Muslim identity may be based on strong religious beliefs, for example, making dissociation particularly costly.⁶ It is also possible that many U.S. Muslims were not able to dissociate because one’s Muslim identity is often ascribed externally. That is, others assign an individual to a given group based on physical traits or other characteristics, such that an individual who may not be a Muslim but may appear Middle Eastern may face the same discrimination as others.

Without the option to pass, individuals may instead try to compensate for the trait they view as a shortcoming, attempting to excel in areas that they might be expected to fall short in (Goffman 1986), measure themselves against others on different metrics (Tajfel and Turner 2004), or resist the stigma by deflecting and challenging it (Thoits 2011). Most relevant to this study, however, is the strategy of increasing one’s identity salience in response to discrimination, a process referred to as reactive ethnicity, whereby the stigmatized identity takes on an increased importance (Peek 2005; Schildkraut 2004, 2005). To protect against threat that stigmatization poses to social identity, especially when people feel like they are being unfairly stigmatized and cannot easily shed their stigmatized identity, they may respond by increasing their closeness with that stigmatized identity. Increasing group identity salience can then act as a defense mechanism, helping buffer individuals against the psychological harm of stigmatization (Branscombe, Schmitt, and Harvey 1999; Fleischmann et al. 2011; Redersdorff, Martinot, and Branscombe 2004; Schmitt and Branscombe 2002; Schmitt, Spears, and Branscombe 2003; Sellers et al. 2003). This protective effect may be due to stigmatized individuals connecting with others who have had similar experiences and can share “tricks of the trade” and offer moral support (Goffman 1986), or, as the rejection-identification model argues, the simple act of increasing one’s sense of belonging to and identification with a stigmatized group, even if unaccompanied by contact with other group members, can itself help protect against the otherwise adverse psychological effects of discrimination, avoiding feelings of low self-esteem and alienation (Branscombe et al. 1999; Schmitt et al. 2003). In these two ways, increased identity salience can help mitigate feelings of isolation and powerlessness that may be borne of stigmatization and act as an empowering force (Lamont, Welburn, and Fleming 2013).

If experiencing discrimination can increase one’s sense of closeness with the stigmatized group, then work on the relationship between group identity and political engagement is particularly relevant to understanding how discrimination could indirectly influence Muslim political participation. Existing research provides support for a positive relationship between the two (Gurin and Epps 1975; Saunders et al. 2012; Shingles 1981; Stokes 2003; Uhlaner, Cain, and Kiewiet 1989).

Group identity may motivate political action through linked fate processes similar to those put forth by Michael Dawson (1994) in his book, Behind the Mule. There, Dawson argues that race is an important determinant of political behavior among African Americans due to their perception that their futures are inextricably linked to other black people. This sense of collectivity fuels black participation across class lines. Similarly, Verba and Nie (1987) found that awareness of a group’s disadvantaged status can bring about political participation, showing that black consciousness accounted for the fact that black individuals were more politically active than white individuals of a similar socioeconomic status.

Related to the linked fate perspective, relative deprivation theory (Stouffer et al. 1949) can also help explain the relationship between group identification and political engagement. This theory posits that when individuals draw comparisons between their own or their group’s situation and others’ and they feel like they are unfairly suffering or have unjustified low status, they may experience feelings of social or political group relative deprivation, which may then motivate action (Klandermans, Toorn, and Stekelenburg 2008; Miller et al. 1981; Walker and Smith 2002; Wright and Tropp 2002).

Increased socialization and emotions such as anger may play a role in this relationship. Muslims who begin associating more closely with other Muslims as a result of increased group identification might be more likely to discuss their negative experiences and become increasingly aware of the disadvantaged status of Muslims as a whole. When this happens, an individual cannot chalk up his or her negative experience to bad luck or think of it as an anomaly, likely instead seeing it as part of a wider pattern (Wright and Tropp 2002). This increased awareness can lead to feelings of anger and resentment, which may compel Muslims to make an effort to insist on their right to belong in the United States and feel a sense of duty to educate the public about their faith (Peek 2005). Indeed, anger has been shown to be particularly effective in motivating political engagement, particularly by social movement researchers (Gurr 2011; Simon and Klandermans 2001; Valentino et al. 2011). Even for those who are not necessarily socializing more with other Muslims as a result of group identification, feeling closer to one’s Muslim identity can make discrimination more personally consequential: it is no longer a problem that is external to the individual but rather one that is a part of the lived experience of being a Muslim. These feelings of group deprivation can then motivate an interest in improving the group’s status (Wright and Tropp 2002).

The aforementioned research lays out a framework supporting the hypothesis that discrimination increases political engagement. What about the inverse? While discrimination may motivate political engagement, it may also have a dampening effect—leading people to disengage from the political arena. The process of stigmatization can result in one viewing his or her own Muslim identity as a “spoiled identity” and lead to demobilization (Fleischmann et al. 2011). In this situation, stigmatized individuals may opt for a different coping mechanism, one that deemphasizes their Muslim identity, instead of adopting a strong, politically mobilized sense of group identity.

Even if they do increase their group identity, increased group identification may bring about insularity, resulting in stigmatized group members withdrawing from the mainstream. This is a particularly strong risk when stigmatization takes place in a country with limited multiculturalist policies, like the United States. In countries where diversity is celebrated and multiculturalist policies are embraced, immigrants generally feel like they have legitimate political standing and are likely to engage politically (Bloemraad 2006). Indeed, research has repeatedly found a strong link between dual identification (as compared to solely identifying with the native country) and political mobilization, with members of an ethnic minority of immigrants who feel a strong sense of both their native and adopted identity being particularly likely to engage in mainstream politics (Simon and Grabow 2010; Simon and Ruhs 2008). This is particularly relevant to consider in the U.S. context given that the United States does not make the same effort to incorporate immigrant minorities, especially in a way that allows them to preserve a sense of their dual identity, that other countries, such as Canada, do.

And increased sense of group identity, even if it does not result in insularity, may simply be insufficient in motivating action. We know, for example, that a politicized collective identity is not only contingent on a strong sense of group identity but rather a strong sense of group identity in an inclusive social context (Simon and Klandermans 2001). In such a context, feelings of belongingness are the norm so that if stigmatization suddenly increases, group members already have a sense of political efficacy and awareness of their rights and duties as members of society, which decreases the chances that they will withdraw politically in the face of negative treatment (Bloemraad 2006; Mettler 2002). Even before 9/11, however, Muslims in the United States occupied the role of stigmatized outsider, just to a lesser extent. When stigmatization takes place in the context of low feelings of belonging, there is a high risk of political alienation in the face of stigmatization, regardless of group identity. Indeed, some research has shown a direct link between perceptions of discrimination against one’s group and alienation from the political process (Schildkraut 2004).

A negative relationship between discrimination and political participation may also partly be due to feelings of fear. Fear may discourage political participation through processes that are similar as those operating in the case of stereotype threat. That is, Muslims may come to act in accordance with the expectation set forth by widespread rhetoric that casts them as deviant members of society who are at odds with the political mainstream. In this way, stigmatization can become a self-fulfilling prophecy. Similarly, Muslims may withdraw from the political system because they feel afraid of asserting their rights or feel that their participation would be ineffectual. Some research has shown that fear plays an important role in influencing the participation of minority groups (Salamon and Evera 1973) or that a sense of powerlessness or low political efficacy may diminish engagement (e.g., Karp and Banducci 2008; Verba and Nie 1987).

Of course, experiencing anti-Muslim stigma may also have no effect on political engagement. This would most likely be the case in situations where the stigma management technique of resistance is used (Thoits 2011). In the Muslim context, this might mean rejecting stereotypes that cast Muslims as terrorists and refusing to see oneself in this ascribed way. This can be done in a way that challenges stereotypes and therefore motivates political action, in line with the literature discussed previously, or can deflect stereotypes without confrontation or attempts to change people’s minds (Thoits 2011).

Building on this existing body of work, this paper seeks to understand the relationship between stigmatization and Muslim political engagement in the post-9/11 era. On one hand, we might expect discrimination to have led to an increase in political engagement as the experience of heightened stigma bound Muslims together, increasing group identification and mobilizing them to action. Conversely, we might expect discrimination to have led to a decrease in political engagement if these experiences did not increase group identification and instead resulted in alienation from the mainstream.

Methods

How does perceiving discrimination impact political engagement? I answer this question in two steps. First, I explore the direct relationship between discrimination and political engagement. I find that discrimination after 9/11 likely led to a null or marginally positive increase in political interest but a decrease in active political participation, meaning that increased discrimination results in increased political interest but has a neutral or dampening effect on political participation for Muslim Americans. I then consider whether this is because discrimination does not result in an increase in group identification, as social identity theory (SIT) would lead us to expect, or whether it does lead to group identification, but that group identification, in this case, is not effectively increasing active participation.⁷

To answer these questions, I use a combination of online survey experiments that I conducted and a matching-based analysis of a nationally representative survey of the Muslim population conducted immediately after 2001. These two sources of data have complementary strengths. Using random assignment to a treatment and control condition in the experimental setting allowed me to isolate the effect of exposure to anti-Muslim stigma. Accounting for reverse causality in this way is particularly important in the context of research that shows that perceptions of stigmatization and social identity may be circuitous, with individuals who have a higher identity salience being more likely to perceive mistreatment (Branscombe et al. 1999; Sellers and Shelton 2003). Also, because I designed and administered the experimental study, I was able to tailor it to this project, asking questions about a wide variety of types of political engagement. I supplemented the experimental data with results from a representative survey of the Muslim population, administered by Zogby International in 2001, shortly after 9/11. This allowed me to examine the broad relationship between stigmatization and Muslim political incorporation in the U.S. Muslim population, testing whether my experimental findings were replicated in a nonexperimental setting with a larger sample size.⁸

For the survey experiments, I recruited respondents from September through November 2014 through two methods: (1) by contacting on-the-ground organizations such as mosques, Muslim community centers, and ethnic organizations (e.g., the Boston Iranian Association) in 13 cities across the United States (representing all regions of the country and from a wide range of countries with high Muslim populations, such as Iran, Iraq, and Turkey); and (2) by contacting members of online groups on social media website with no on-the-ground presence (e.g., members of the Facebook group “Iranians of Boston”). In both instances, I either posted directly to the group’s page or asked group administrators to share the survey link with group members. I elected to recruit respondents using both strategies, as opposed to just one or the other, to diversify my sample and avoid limiting the sample to individuals who feel a strong enough sense of their ethnic or religious identity to be part of an on-the-ground organizations, where they would ostensibly regularly socialize and attend events with other group members. Of course, to some extent, participants may still more strongly identify with either their Muslim or national identity given that, at the very least, they are part of a social media group with other similarly identifying individuals. In the absence of access to a representative sample, this recruitment method was the best option.

I dropped individuals who indicated that they did not identify as Muslim, those who did not finish the entire experiment, and those who failed an attention check question from my analysis.⁹ My final sample consisted of 574 respondents (nearly 200 in each of three experiments) from 115 different organizations.¹⁰ As can be seen in Table 1, this recruitment method yielded an experimental sample that was demographically similar to the general U.S. Muslim population, though with higher education levels on average and a higher proportion of men and non–first generation immigrants.

Table 1.

Demographics of Experiment Respondents Compared to U.S. Muslim Population

	Demographics of respondents in experimental sample	Demographics of Muslim population based on 2011 Pew Survey
Gender	69% Men 30% Women 1% Other/don’t know	55% Men 45% Women
Marital status	81% Married 16% Never married 3% Divorced or widowed	55% Married 35% Never married 8% Divorced, widowed, or separated
Age	Average: 46	Average: 40–54
Education	5% High school diploma or GED 10% Associate’s degree 51% Bachelor’s degree 34% Graduate degree	14% No high school diploma or GED 59% High school diploma or GED 15% Bachelor’s degree 11% Graduate degree
Family income	Average: $50,000–$60,000	Average: $50,000–$100,000
Political party	21% Republican 45% Democrat 26% Independent 8% Other/don’t know	11% Republican 46% Democrat 24% Lean Democrat 19% Independent or no preference
Region of country	17% Northeast 32% Midwest 26% South 25% West	28% Northeast 20% Midwest 30% South 22% West
Immigrant status	26% Immigrant 72% Born in United States 2% Don’t know/not sure	63% Immigrant 37% Born in United States

Note: This table uses demographic data from the 2011 Pew survey for comparison because it is the nationally representative survey that was administered closest in time to the experiment.

For the survey analysis, I used data from the MAPS: Muslims in the American Public Square Survey, a cross-sectional survey administered by Zogby International immediately after the attacks in 2001. The 1,781 Muslim adult respondents were recruited by matching the zip codes of 300 randomly selected Islamic centers against local telephone exchanges and then contacting individuals with common Muslim surnames from these telephone exchanges. The survey was conducted from November 8 through November 19, 2001. Demographic information about this sample can be found in Table 2.

Table 2.

Demographics of MAPS: Muslims in the American Public Square Survey Respondents (N = 1,781)

Age	Average: 41 years
Gender	Male: 59% Female: 41%
Education	Less than high school degree: 6% High school graduate: 35% College graduate: 59%
Household Income	$50,000–$75,000
Region	East: 38% South: 21% Central: 28% West: 13%
Immigrant	No: 32% Yes: 68%

This sampling strategy may introduce bias for two reasons. First, it removes from the sample individuals who do not have common Muslim surnames—a particularly important point given that having a Muslim sounding name may increase the likelihood of discrimination. Sampling individuals based on proximity to Islamic centers also likely results in an overrepresentation of Muslims who live in areas with large Muslim populations and have more opportunities for political activity. Though these shortcomings of the survey are not inconsequential, I did find that the demographics of survey respondents were fairly similar to those from a nationally representative survey conducted by the Pew Research Center in 2007, using a probability sampling approach based on a random-digit dial method. I used the MAPS survey instead of the Pew 2007 survey or other, more recent surveys of the Muslim population conducted by Pew in 2011 and 2016 because these surveys had very limited information about political engagement.

Results

Discrimination and Political Engagement

I first examined the direct relationship between anti-Muslim stigma and political engagement using both the online experiment and survey analysis. In the survey experiment, I randomly assigned respondents to either a treatment or control condition. The treatment primed respondents’ awareness of anti-Muslim stigma. Respondents assigned to that condition were shown four images of Muslims being treated poorly or Islam being stigmatized. One image presented to those in the treatment condition, for example, showed an anti-Muslim rally with one protestor holding up a sign reading “All I need to know about Islam I learned on 9/11.” Another showed sprayed graffiti on a future site for an Islamic center that read “Not welcome.” Those in the control condition were shown four images of basic geometric shapes.¹¹ After viewing each image, the respondent was asked describe the image they had seen in 150 words or less, a question designed to reinforce the effects of the image, allowing for a stronger manipulation.

Though most traditional studies about political engagement focus exclusively on voting, this paper focuses on a wider variety of forms of engagement in trying to form a broader picture of how Muslims have (or have not) engaged in the post-9/11 era. In addition to voting, I examine other types of active participation (acts that are aimed at bringing about political change and grant citizens a direct say in the political system such as participating in protest and rallies and contacting political representatives) as well as political interest engagement (acts that are precursors to participation such as participating in political discussions and staying informed about politics and current events: see Conge 1988 and Milbrath 1977). Though political interest alone might not bring about political change, it can be a fundamental stepping-stone toward more active participation, as well as a marker of interest in current events and the political culture of a society.

In the experiment, I measured the effect of exposure to stigma on three types of active political engagement: (1) active engagement while taking the experiment—providing respondents with the option to donate money to political organizations, join political mailing lists, and sign petitions on a political issue^12,13; (2) predictions of future voting; and (3) predictions about other types of future active participation, such as political contributions, participating in rallies, and writing a politician about an issue. I also measured the effect of exposure to stigma on political interest, indicated by respondents’ predictions of the likelihood of discussing politics or vising a political website in the future. Respondents were shown all questions about political engagement in random order.¹⁴

To analyze the experimental data, I conducted a series of Mann-Whitney U Tests. Mann-Whitney tests evaluate whether two samples have equal sample means. It determines, for example, if there is a significant difference in political engagement between those assigned to the control group and those assigned to the treatment group (Mann and Whitney 1947; McKnight and Najab 2010). Unlike more commonly used ANOVA tests, the Mann-Whitney U does not assume interval level outcome variables or a normal distribution.

The results of the analysis, as seen in Table 3, show that being exposed to stigmatization decreases real-time behavioral engagement by 13.9 percent. That is, those assigned to the treatment condition had an average rate of signing petitions, joining mailing lists of political groups, and making donations to political organizations that was about 14 percent lower than those in the control condition while taking the experiment. There was not, however, a statistically significant effect of exposure to stigmatization on predictions of future political engagement (including voting).

Table 3.

Results of Survey Experiment: The Effect of Perceived Stigma (N = 180)

	Predictions of future voting	Predictions of other future active participation	Active participation during survey	Predictions about future political interest
Treatment condition mean	6.17	6.51	6.38	7.12
Control condition mean	6.20	7.01	7.77	7.10
Percentage change and significance (Mann-Whitney test)	−.03	5.00	−13.90*	.02

p < .10.

The survey analysis shows somewhat similar findings. To maximize the robustness of my results, I used a propensity score–based matching approach, as opposed to a standard ordinary least squares (OLS) regression. Propensity score matching, initially developed by Rosenbaum and Rubin (1983), artificially creates a control group that is similar to the treatment group on a set of observed confounding covariates and then yields a series of weights that can then be applied in a regression analysis to estimate the relationship between two variables. To create these groups, individuals are matched based on propensity score values, single number summaries of the likelihood that a person would have been assigned to the treatment group conditional on the specified confounders. Here, I matched individuals within their five nearest neighbors based on Mahalanobis distance, with replacement, meaning that one individual in the control group can be matched multiple times to individuals in the treatment group, to create two groups that were similar on a set of observed confounders, shown in Table 4. This is a particularly useful approach with a smaller data set such as this one.¹⁵

Table 4.

Matching and Control Covariates Used in Survey Analysis

Used in propensity score analysis—all respondents were matched on the following variables:
Employment status	Region of country
Ethnicity	Marital status
Race	Gender
Political ideology	Immigrant
Political party	Number of children
Education	Age
Income
Used in regression analysis but not propensity score analysis (potential posttreatment confounders)
Civic engagement
Prayer frequency
Converted to Islam
Importance of domestic policies to voting decision
Importance of Islam

After matching, I assessed the balance between the created comparison groups by looking at the difference in means and ratio of standard deviations between them. I was able to achieve much greater balance on all variables than the benchmarks specified by Rubin (2001). I also confirmed that there was sufficient overlap by checking the graphic distribution of the estimated propensity score for the treatment and control group and checking for sufficient areas of common support, as recommended by Harder, Stuart, and Anthony (2010).

Similar to regression analysis, propensity score matching assumes ignorability—basically that we are able to observe all confounders. Though the MAPS survey that I use here did include information about a standard set of confounders such as education, income, and age, there are others that could potentially have affected treatment assignment. One of the main advantages of the propensity score approach, however, is that it allows for sensitivity analyses, which estimate how susceptible results are to any unobserved bias. Sensitivity analysis allowed me to estimate how much the results would change if there were unobserved confounders (Rosenbaum 2002). Basically, it considers what the effect would be of an unobserved confounding variable that increased the odds of exposure to treatment. In this case, that would mean a variable that introduced bias because it made it more likely for an individual to have perceived discrimination. It allows us to answer how much “hidden bias” there can be without the results changing. If our results are not sensitive, then they are more robust. The sensitivity analysis I conduct here uses Rosenbaum (2002) bounds to indicate whether the treatment effect would remain significant if an unobserved confounder increased the odds of exposure to treatment for the treated group, meaning that the model overestimated the treatment effect.¹⁶

The other significant advantage of the propensity score approach is that it limits results to the proportion of the population for whom we have reliable evidence. That is, instead of OLS regression methods, which generalize results to the entire population despite minimal evidence that the relationship between two variables remains consistent throughout the distribution, propensity score matching–based analysis estimates the relationship between perceptions of anti-Muslim stigma and political stigma among those who perceived stigmatization.

Because the sample size of the survey was relatively small, I used a multiple imputation (MI) method, as initially laid out by Rubin in 1978, to account for missing data. MI fills in each missing value with a range of possible values based on existing values from other variables. I generated 25 unique imputed data sets here and then pooled the results to create a single data set used in running the analysis described in the following. MI did not substantially change the mean and standard deviation of the variables used in the analysis.

To examine the relationship between discrimination and political engagement in the survey, I ran two separate propensity score matching models (one for each predictor) looking at the effect of (1) perception of general anti-Muslim stigma, measured by whether respondents thought the media fairly portrayed Muslims and Islam, and (2) personal experiences of anti-Muslim stigma—reporting that one personally or one’s community had experienced stigma in the form of “verbal abuse” after 9/11. I measured the effect of each of these on active political engagement, indicated by response to a question asking respondents how politically active they considered themselves to be and political interest, indicated by an index of questions asking, for example, how often respondents discuss politics with family and friends and follow politics.¹⁷ The only relevant question about voting was about voter registration, and this question did not have enough variation to render meaningful results. No other question captured political behavior meant to effect change.

I found a negative relationship between personal discrimination and active political participation (Table 5). That is, after matching on covariates, I found that an individual who perceived to having personally experienced anti-Muslim stigma was 6.9 percent less likely to report being politically active than if he or she had not perceived such abuse. Sensitivity analysis, also shown in Table 5, showed that these results are insensitive to any bias that would increase the odds of exposure up to 1.4 times (i.e., by 140 percent), with the effect becoming statistically insignificant at the .05 level only as Γ approaches 1.5. This helps affirm the analysis findings given that a confounder at this level of importance would have likely been included in the model.

Table 5.

Propensity Score Matching–Weighted Regression Results of Survey Analysis

	Active political participation		Political interest
Media stigma (mean = .73; SD = .44)	.017(.035)		.030*(.018)
Personal stigma (mean = .25; SD = .43)		.069*(.031)		.022**(.010)
R²	.10	.18	.26	.13
N	866	866	866	866
Sensitivity analysis	n/a (effect not significant)	p < .1 until Γ = 1.5	p < .05 until Γ = 1.5	p < .05 until Γ = 1.2

Note: Sensitivity analysis uses Rosenbaum (2002) bounds.

p < .10. **p < .05.

While I found a negative effect of discrimination on active political participation, I found it to have a positive effect on political interest. Those who perceived themselves to have been victims of verbal abuse or were otherwise aware of anti-Muslim stigma faced by their community were more interested in politics by 2 percent to 4 percent. I found these results to be robust using the same sensitivity analysis as previously described—with the effect of anti-Muslim stigma impervious to any unobserved confounder that would increase the odds of exposure to the treatment by up to 1.6 times (160 percent) and personal discrimination robust to a confounder that would increase exposure by up to 1.3 times.

What Is the Role of Group Identity?

The previous section highlighted the direct relationship between perceived discrimination and political engagement, finding that although discrimination increased one’s level of interest in politics, it decreased behavioral participation. If discrimination is not followed by political mobilization, as we might expect based on social identity theory (SIT), might it be because it is not actually increasing group identification? Or if it is, is group identification not effectively translating into political action? I answer each of these questions in turn.

Because the MAPS surveys did not include any questions indicating group identification explicitly, I test this proposed framework using two survey experiments, one in which I considered the effect of perceived discrimination on group identification and another where I looked at the effect of group identification on political engagement.¹⁸ Respondents were recruited using the same strategy in the experiment outline previously described, and respondents in the first experiment were randomly assigned to treatment and control conditions that were identical to those in the experiment previously described, primed with four images of anti-Muslim stigma or with various geometric shapes. I then measured identity salience using a battery of questions such as how strongly respondents identified as being Muslim and how close they felt to other Muslims, which I combined to create an additive index.¹⁹

Again, I analyzed the results using a series of Mann-Whitney U tests. I found that exposure to anti-Muslim stigma increases group identification by 6.33 percent (table 6). Those assigned to the high-stigma condition reported significantly higher positive feelings toward other Muslims and increased closeness with their Muslim identity than those assigned to the control condition. This means that exposure to anti-Muslim stigma does indeed increase group identification, so the only explanation for a decrease in active political participation is that group identification is not motivating such behavioral changes.

Table 6.

Effect of Anti-Muslim Stigma on Group Identification: Experiment Mann-Whitney Results

	Group identification
Stigmatization condition mean	12.69
Control condition mean	11.74
Percentage change and significance (Mann-Whitney test)	6.33**
N	180

p < .05.

In the final experiment, I examined this question, looking at the effect of group identification on political engagement. There is limited prior research that manipulates group identification. I did so here by randomly assigning respondents to a highly identified condition, where they were shown images of successful Muslims in the United States, Ramadan celebrations, and Mecca, with a short, three- to four-sentence description of each image or the control condition. I also included a manipulation check in this experiment to ensure that the manipulation effectively increased feelings of group identification.²⁰ I measured the effect of group identification on two types of political engagement: active political participation, or political engagement while taking the survey, indicated by willingness to make political donations, subscribe to mailing lists of political organizations, and sign political petitions; and predictions about future engagement and political interest, indicated by how likely respondents think they are to visit a political website in the next year and discuss politics with family and friends.

I found that high Muslim identity salience does not have a statistically significant effect on active political participation (table 7). That is, those assigned to the treatment condition were no more likely to participate in political activities while taking the experiment or predict future engagement than those assigned to the control condition. I did find, however, that increased identity salience did increase interest in politics. Those assigned to the treatment condition had a mean response to these questions/this question that was 4 percent higher than those assigned to the control condition.

Table 7.

Effect of Group Identification on Political Engagement: Experiment Mann-Whitney Results

	Predictions of future voting	Predictions of other future participation	Participation during survey	Predictions of future political interest
Group identification condition mean	6.20	15.08	1.80	5.41
Control condition mean	5.96	15.47	1.76	4.58
Percentage change and significance (Mann-Whitney test)	−1.86	3.14	1.33	3.95***
N	173	173	173	173

***

p < .01.

The results of these two experiments together indicate that exposing Muslims to stigmatization similar to what they might read about in a newspaper or encounter in daily life makes them more likely to feel close to other group members and identify strongly as a Muslim. This increased identity salience in turn impacts their interest in the political system but is not able to fully mobilize them to action, having no immediate effect on the likelihood that they will sign petitions, make donations to political organizations, and join mailing lists of political groups in the present, or on their projections about their future behavioral engagement.

Discussion

The central finding of this paper is that a sudden surge in discrimination may bring about an interest in the political interest but can depress the active political engagement of a stigmatized group, meaning that those who perceive themselves or their group to be victims of discrimination begin to withdraw politically. Specifically, I found that post-9/11 stigmatization has a statistically significant negative effect on rates of Muslim political participation, with Muslims being about 14 percent less likely to engage politically after being exposed to examples of widespread stigma. As the accompanying survey analysis demonstrated, this negative relationship seems to be reflected in the U.S. Muslim population, at least in the months immediately following 9/11. Importantly, the experience of Muslims immediately after 9/11 seems to be similar to experiences they have had in the past couple of years, which have seen an even larger rise in the rate of hate crimes and other reports of Muslim discrimination.

This decrease in political participation is not because facing stigma does not increase group identification. Indeed, it does. In the second part of the paper, I looked at whether discrimination has an effect on group identification, demonstrating that it increases it, whether or not group identification is then increasing political engagement. I find that an increase in group identity, while likely increasing interest in the political system, does not lead to more active participation.

The fact that stigmatization increases political interest and that group identification is likely mediating this relationship provides evidence of the potential usefulness of group-based identity as a stigma management technique. Existing research on this relationship predicts that because individuals want to think of themselves and their identity in as positive a way as possible, they are likely to buffer themselves against the adverse effects of stigma by increasing closeness with their ethnic group (Tajfel, and Turner 1979). The sense of solidarity that is borne of increased group identity may motivate increased political interest because politics is more immediately consequential to the lives of group members. Political interest is an important measure partially because it can act as an indicator of future behavioral engagement—interest indicates that the Muslim population is not politically apathetic, even if feeling ineffectual. Political interest has indeed been shown to be important in bringing about behavioral engagement (McLeod, Scheufele, and Moy 1999). Additionally, political interest as defined here signals greater political literacy, which has been shown to have a positive effect on participation, for example, accounting for the gender disparity in participation between men and women (Verba, Burns, and Schlozman 1997).²¹

Group identification, however, does not seem to be sufficient in protecting against the adverse effects of discrimination when it comes to active political participation. This alienation from the political system may have a number of explanations. It may be because, as research on the importance of multiculturist policies shows, context matters. As discussed at the beginning of this paper, this strand of research has highlighted that the absence of multicultural policies and a lack of emphasis on the importance of diversity may deter political engagement. That is, in this context, an increased sense of group identity is not a sufficient condition for ensuring increased political engagement when there is a rise in stigmatization. What is needed, rather, is strong ethnic identification in the context of societal encouragement, or at least acceptance, of multiculturalism. In addition, the limited effectiveness of group identity as a buffer may also have to do with the sudden nature of anti-Muslim stigma. Research on the buffering effect of group identity has almost entirely looked at the response among members of groups who have consistently faced stigma over long periods of time, such as African Americans. It is perhaps not surprising that these findings cannot be perfectly generalized to members of suddenly stigmatized groups such as Muslims in the United States. Members of such suddenly stigmatized groups may not, for example, have the psychological or physical resources to shield against political alienation.

Diminished engagement may also relate to feelings of fear. Increased awareness of anti-Muslim sentiment and hate crimes may have triggered fear among Muslims, increasing the “cost” of behavioral engagement and thereby diminishing its frequency. The potentially powerful effect of fear was put forth by Salamon and Van Evera (1973), where they found that fear, measured by feelings of vulnerability among black people living in the southern United States, accounted for 69 percent of the variation in political participation. Feelings of fear could stem from concerns about encountering hostility if one were to engage in a protest after 9/11, for example, especially in the immediate months after the attacks. Fear could also operate less directly, implying a fear of alienating friends or community members (Rosenberg 1954).

Alternatively, as suggested by research on social stigma and political efficacy, it is possible that widespread stigmatization brings about feelings of low-political efficacy or powerlessness. Compounding reports of discrimination and hate crimes may make one think that he or she cannot change the status of Muslims in the United States and the participation in the political system would be futile. As defined early on by Campbell (1954), a sense of political efficacy broadly refers to feeling that individual political action can impact the political process: that it is worthwhile to participate and change is possible. Political efficacy is often considered under the umbrella of “personality effects”—a series of components of one’s personality that may influence whether one participates, such as a sense of civic duty or political interest. A low sense of political efficacy or similarly, feelings of political alienation could have a negative effect on political engagement (Gallego and Oberski 2012; Rodgers 1974).

The MAPS survey data, collected in 2001, just months after the 9/11 attack, captures a period when Muslims became an incredibly stigmatized group almost overnight. While, to some extent various Middle Eastern groups had faced various flare-ups of nativist sentiment in the past, such as during the Iran hostage crisis in 1979, the widespread, extreme hostility Muslims as a whole began to experience in the United States during that time was largely unprecedented. Given the suddenness of the rise in discrimination and hate crimes experienced by Muslims after 9/11, coupled with the general prevalence of anti-Muslim sentiment, it is highly conceivable that being faced with discrimination diminished Muslims’ sense of political efficacy, making them believe that they could not influence the political system as part of such a small minority group and that they had relatively little power over what was happening to them. Prior to building up any sort of defense, stigmatization can be a significant shock, heavily impacting an individual ill-equipped to deal with it. Of course, it is possible that over time, as people build up defenses to such stigma and the Muslim group has organized and mobilized more effectively, such a negative response would diminish.

Voting presents a unique case in my results. I did not find perceptions of anti-Muslim stigma to have any effect on predictions about future voting. This result is important given the prevalence with which voting is used as a stand-in for general political participation in existing research, highlighting the importance of using a wider variety of political engagement measures. Of course, it is an important measure because it is a widely available form of political activity, but it is also one that has minimal cost and might capture a slightly different type of motivation than other types of political behavior. Accordingly, voting has been shown to often cluster more closely with other patriotic acts (e.g., agreeing that “I love my country” or displaying an American flag) than it does with political participation, a finding that is largely confirmed by this paper (Buffalo Survey 1968 as cited in Milbrath 1977). This might be because voting, especially in national elections, is mostly about affirming one’s loyalty and sense of civic duty instead of making demands on the political system (Milbrath 1977).

Conclusion

This paper contributes to our understanding of the relationship between perceived discrimination and political engagement among minority groups suddenly facing a spike in discrimination, as Muslims have in post-9/11 America. I used a mixed-methods approach, combining propensity score matching and experimental analysis to look at the effect of discrimination.

Since the 2016 presidential election, reports of anti-Muslim (and anti-Black, anti-Hispanic, and anti-gay) hate crimes have continued to increase. In the years since the election and the vitriol and ethnic animosity that accompanied it, racial and ethnic tension have remained high, with anti-Muslim sentiment reported to be as high or higher than it was after 9/11, coupled with a FBI report finding that there were more anti-Muslim hate crimes in 2015, leading up to the 2016 election, than immediately after 9/11.²²

This surge of anti-Muslim stigma could bring with it important consequences. In 2001, shortly after 9/11, those who personally experienced discrimination reported diminished political activity. Priming Muslims to think about anti-Muslim hate crimes in 2014 brought about a similar result. In 2001 this decline took place parallel to an increase in political interest or interest in the political system and a recognition of its importance. To the extent that these patterns may repeat themselves, Muslims and other minority group members may feel politically alienated in the coming months, withdrawing politically even while discussing politics with family and friends and watching political news more often, a response that may be unexpected by many.

Supplemental Material

Supplemental_Material – Supplemental material for When Interest Doesn’t Turn into Action: Discrimination, Group Identification, and Muslim Political Engagement in the Post-9/11 Era

Supplemental material, Supplemental_Material for When Interest Doesn’t Turn into Action: Discrimination, Group Identification, and Muslim Political Engagement in the Post-9/11 Era by Delaram Takyar in Social Psychology Quarterly

Footnotes

Acknowledgements

I would like to thank Bart Bonikowski and Jennifer Hochschild for reading and helping to revise early drafts of this work and Jim Sidanius, Gwenyth McClendon, and Dohoon Lee for their methodological and substantive feedback at various stages of this project. I would also like to thank the participants of the NYU Sociology Research and Writing Seminar; Jeff Manza, Iddo Tavory, and Patrick Sharkey for their feedback; in addition to the three anonymous reviewers at Social Psychology Quarterly.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded in part by a small grant from the Center for American Political Studies at Harvard University.

ORCID iD

Delaram Takyar

Supplemental Material

Supplemental material for this article is available online.

1

Of course, as point out, anti-Muslim sentiment was not a wholly new phenomenon after 9/11. Indeed, after other historical events such as the Iranian hostage crisis and the 1973 oil embargo, Muslims experienced similar waves of stigmatization. These waves, however, did not become permanent, and the extreme hostility Muslims as a community have experienced since 9/11 is largely unprecedented.

2

The full report that discusses this rise, including examples of various incidents, can be found at .

3

It should be noted that members of the group facing discrimination in the wake of 9/11 were not exclusively Muslim, nor were they exclusively Middle Eastern or Arab. To some extent, these phrases have been used interchangeably by mass media and within American society. Muslims were chosen as the subject of study partially because they have been the focus of the limited surveys that have been conducted of this population in the post-9/11 era. This is probably because of the relative ease of defining this group compared to the Middle Eastern population (as there is contention as to which countries should be considered part of the Middle East) or the Arab population (a far smaller group). Thus, choosing Muslims as the group of study is a methodological decision much more so than a theoretical one and should not detract from the fact that there are individuals who are not Muslim who have experienced, and continue to experience, very similar forms of discrimination. As is the case with most studies that restrict their focus to members of an ascribed group, the boundaries of the Muslim group in the United States are fluid, and while someone might not identify as being religiously Muslim, they might be grouped with individuals who are being “othered” through anti-Muslim discrimination. An in-depth discussion about group boundary-making is beyond the scope of this paper but is helpful in thinking about the relationship between discrimination and group identity. As pointed out by Lamont (2014), the process of stigmatization is fundamental in sorting individuals into groups. Such group boundaries are not innate in society but can be created and strengthened as groups strengthen ties of solidarity in the face of stigmatization (). In the post-9/11 era, the group facing stigmatization was somewhat newly constructed; discrimination did not impact members of just one, single preexisting group in U.S. society.

4

See for more details.

5

See for more details.

6

A Muslim woman who wears a hijab for religious reasons will be more easily identified as Muslim and therefore might be more likely to face discrimination. But if she chooses to not wear the hijab, then she may be violating her religious beliefs. Other physical characteristics, such as wearing a turban, skin tone, facial hair, and speaking English with an accent, are often associated (and just as often misassociated) with Muslims and have served as a basis for discrimination in the years since 9/11 ().

7

In studying Muslim political engagement, I am not exclusively interested in acts that are only meant to change the status of Muslims as a group that faces discrimination, though this may be one primary reason that people choose to engage in the wake of stigmatizations. While this type of participation has immediately obvious importance to the Muslim community, looking solely at that type of engagement ignores the possibility that some stigmatized individuals may respond to discrimination by doubling down on their rights as members of American society—deflecting and rejecting the stigma by further investing in society the same way that nonstigmatized individuals do. This approach of challenging stigma is discussed in more detail on page 5.

8

It is important to note that I was not able to conduct these experiments in the immediate aftermath of 9/11, when the survey was conducted, which may present a limitation in terms of their comparability.

9

This question asked all respondents to select the word green from a list of words, ensuring that respondents were paying attention and allowing me to remove any automated, computer respondents from the final analysis

10

Between 25 percent and 35 percent (varying with each experiment) of initial respondents did not identify as Muslim (likely partly because individuals were recruited from non–religion based organizations, such as the Iranian Association of Boston), between 1 percent and 3 percent failed the attention check in each experiment, and around 33 percent did not complete the survey.

11

Though these geometric shapes are not necessarily comparable to being shown images of stigma, this approach was chosen because shapes are unlikely to trigger any sort of uncontrolled priming effect, being more neutral than, for example, images of people in a social setting.

12

The types of organization and issues selected were intentionally chosen to be of mainstream interest. That is, instead of asking respondents if they wanted to donate to Muslim political organizations, for example, I asked if they wanted to donate to the Democratic or Republican National Committee or join the mailing list for similar organizations.

13

Individuals who indicated that they wanted to donate money, join a mailing list, or assign a petition were provided with a link to do so and then asked whether they had been successful. Based on these self-reports, between 88 percent and 95 percent of respondents who clicked on the link to be redirected to donate money, join a mailing list, or sign a petition actually did so. Of course, these rates may be overestimated given that I could not independently verify whether people had in fact completed the activity. But measuring behavioral engagement while taking the survey in this way still presents a significant advantage over the traditionally used approach of relying on future predictions or past reports as a sole indicator of participation because research has shown that people often misremember past activity and overestimate future political engagement (Connelly 1945; Tittle and Hill 1967; ).

14

Full wording of all questions can be found in .

15

Because there are no longitudinal surveys of the U.S. Muslim population, which would have allowed me to match the treatment and control groups using confounders measured prior to the treatment (discrimination), I eliminated any confounders that might have theoretically taken place posttreatment from the matching process. This means that any variable that could have changed after experiences of discrimination, such as degree of religiosity (which could conceivably have increased after a person began to perceive discrimination), was removed from the matching part of the analysis, though I included these variables in the next stage of analysis as controls as is done in standard ordinary least squares models.

16

Because this approach is solely concerned with the association between an unobserved confounder and the treatment, it assumes a near perfect association between any confounder and the outcome, making it a particularly conservative method to measure sensitivity.

17

All four questions used to create this index can be found in along with an indicator of how strongly they loaded on to the political interest factor in factor analysis.

18

This approach of directly manipulating a mediator variable and measuring its effect on the outcome of interest allowed me to avoid the “black box” problem of mediation research (Bullock, Green, and Ha 2010; Spencer, Zanna, and Fong 2005).

19

The full battery of questions can be found in .

20

This manipulation check consisted of three questions meant to measure feelings of group identification. These questions asked respondents how strongly they identify as a Muslim, how close they feel to other Muslims, and much they like other Muslims in the United States. Factor analysis revealed that identifying as a Muslim fell on a different dimension of group identification than feeling close to and liking other Muslims (α = .75). Mann-Whitney tests confirmed that those in the experimental condition had higher rates of both Muslim identification (p = .02) and group affinity (p = .09), meaning that the manipulation effectively increased feelings of group identification among the experimental group as intended. Full wording of questions and manipulation check results can be found in .

21

As noted previously, despite this finding in the survey analysis, I found no effect of discrimination on political interest in my experiment. This may be because discrimination does not increase group identity to high enough levels to have an immediate effect on political interest. This would mean that it is possible that over a period of weeks or months after an initial experience of discrimination (or in the face of an accumulation of such negative experiences), group identification continues to increase and becomes further strengthened, eventually helping to bring about an increased interest in politics, or that discrimination only immediately increases group identity if one has a very personal experience of discrimination, as opposed to being made aware of genera anti-Muslim sentiment.

22

See full report at: .

Bio

Delaram Takyar is a PhD candidate at New York University and JD candidate at Yale Law School. Her research focuses broadly on issues related to inequality in the United States, with a particular focus on class-based stratification and multimethods approaches.

References

Bakalian

Anny

Bozorgmehr

Medhi

. 2009. Backlash 9/11: Middle Eastern and Muslim Americans Respond. University of California Press.

Bloemraad

Irene

. 2006. Becoming a Citizen: Incorporating Immigrants and Refugees in the United States and Canada. Los Angeles: University of California Press.

Branscombe

Nyla R.

Schmitt

Michael T.

Harvey

Richard D.

1999. “Perceiving Pervasive Discrimination among African Americans: Implications for Group Identification and Well-Being.” Journal of Personality and Social Psychology 77(1):135–49.

Bullock

John G.

Green

Donald P.

Shang E.

2010. “Yes, but What’s the Mechanism? (Don’t Expect an Easy Answer).” Journal of Personality and Social Psychology 98(4):550–58.

Campbell

Angus

. 1954. The Voter Decides. New York: Row, Peterson.

Conge

Patrick J.

1988. “The Concept of Political Participation: Toward a Definition.” Comparative Politics 20(2):241–49.

Connelly

Gordon M.

1945. “The Questions the Polls Ask: Now Let’s Look at the Real Problem: Validity.” Public Opinion Quarterly 9(1):51–60.

Cooley

Charles Horton

. 1983. Human Nature and the Social Order. New Brunswick, NJ: Transaction Publishers.

Dawson

Michael C.

1994. Behind the Mule: Race and Class in African-American Politics. Princeton, NJ: Princeton University Press.

10.

DeSipio

Louis

. 2002. Immigrant Organizing, Civic Outcomes: Civic Engagement, Political Activity, National Attachment, and Identity in Latino Immigrant Communities. St. Mary’s City, MD: Center for the Study of Democracy.

11.

Diamond

Jeremy

. 2015. “Donald Trump: Ban All Muslim Travel to U.S.”http://www.cnn.com/2015/12/07/politics/donald-trump-muslim-ban-immigration/index.html.

12.

Dion

Kenneth L.

Dion

Karen K.

Pak

Anita W.

1992. “Personality-Based Hardiness as a Buffer for Discrimination-Related Stress in Members of Toronto’s Chinese Community.” Canadian Journal of Behavioural Science/Revue Canadienne Des Sciences Du Comportement 24(4):517–36.

13.

Disha

Ilir

Cavendish

James C.

King

Ryan D.

2011. “Historical Events and Spaces of Hate: Hate Crimes against Arabs and Muslims in Post-9/11 America.” Social Problems 58(1):21–46.

14.

Fleischmann

Fenella

Phalet

Karen

Klein

Olivier

. 2011. “Religious Identification and Politicization in the Face of Discrimination: Support for Political Islam and Political Action among the Turkish and Moroccan Second Generation in Europe.” British Journal of Social Psychology 50(4):628–48.

15.

Gallego

Aina

Oberski

Daniel

. 2012. “Personality and Political Participation: The Mediation Hypothesis.” Political Behavior 34(3):425–51.

16.

Goffman

Erving

. 1986. Stigma: Notes on the Management of Spoiled Identity. Reissue ed. New York: Touchstone.

17.

Gurin

Patricia

Epps

Edgar

. 1975. Black Consciousness, Identity, and Achievement: A Study of Students in Historically Black Colleges. New York: Wiley.

18.

Gurr

Ted Robert

. 2011. Why Men Rebel. London: Routledge.

19.

Haberman

Alexander Burns Maggie

Parker

Ashley

. 2016. “Donald Trump’s Confrontation with Muslim Soldier’s Parents Emerges as Unexpected Flash Point.” The New York Times, July31.

20.

Harder

Valerie S.

Stuart

Elizabeth A.

Anthony

James C.

2010. “Propensity Score Techniques and the Assessment of Measured Covariate Balance to Test Causal Associations in Psychological Research.” Psychological Methods 15(3):234–49.

21.

Karp

Jeffrey A.

Banducci

Susan A.

2008. “Political Efficacy and Participation in Twenty-Seven Democracies: How Electoral Systems Shape Political Behaviour.” British Journal of Political Science 38(2):311–34.

22.

Klandermans

Bert

van der Toorn

Jojanneke

van Stekelenburg

Jacquelien

. 2008. “Embeddedness and Identity: How Immigrants Turn Grievances into Action.” American Sociological Review 73(6):992–1012.

23.

LaGumina

Salvatore John

. 1999. Wop!: A Documentary History of Anti-Italian Discrimination in the United States. Montreal: Guernica Editions.

24.

Lamont

Michèle

. 2014. “Reflections Inspired by Ethnic Boundary Making: Institutions, Power, Networks by Andreas Wimmer.” Ethnic and Racial Studies 37(5):814–19.

25.

Lamont

Michele

Welburn

Jessica S.

Fleming

Crystal

. 2013. “Responses to Discrimination and Social Resilience Under Neo-Liberalism: The United States Compared.” Pp. 129–57 in Social Resilience in the Neoliberal Age, edited by Hall

P. A.

Lamont

Cambridge, UK: Cambridge University Press.

26.

Levin

Brian

. 2016. “Study: In Wake of Terror, Anti-Muslim Crimes Escalate.”http://www.huffingtonpost.com/brian-levin-jd/study-in-wake-of-terror-a_b_8838670.html.

27.

Mann

H. B.

Whitney

D. R.

1947. “On a Test of Whether One of Two Random Variables Is Stochastically Larger than the Other.” The Annals of Mathematical Statistics 18(1):50–60.

28.

Mattila

Mikko

Papageorgiou

Achillefs

. 2017. “Disability, Perceived Discrimination and Political Participation.” International Political Science Review 38(5):505–19.

29.

McKnight

Patrick E.

Najab

Julius

. 2010. “Mann-Whitney U Test.” Pp. 960-961in The Corsini Encyclopedia of Psychology, edited by Weiner

Irving B.

Craighead

W. Edward

. New York: Wiley.

30.

McLeod

Jack

Scheufele

Dietram

Moy

Patricia

. 1999. “Community, Communication, and Participation: The Role of Mass Media and Interpersonal Discussion in Local Political Participation.” Political Communication 16(3):315–36.

31.

Mettler

Suzanne

. 2002. “Bringing the State Back in to Civic Engagement: Policy Feedback Effects of the G.I. Bill for World War II Veterans.” The American Political Science Review 96(2):351–65.

32.

Milbrath

Lester W.

1977. Political Participation: How and Why Do People Get Involved in Politics?2nd ed. Edited by Goel

M. L.

Chicago: Rand McNally.

33.

Miller

Arthur H.

Gurin

Patricia

Gurin

Gerald

Malanchuk

Oksana

. 1981. “Group Consciousness and Political Participation.” American Journal of Political Science 25(3):494–511.

34.

Peek

Lori

. 2005. “Becoming Muslim: The Development of a Religious Identity.” Sociology of Religion 66(3):215–42.

35.

Project MAPS. 2004. “Muslims in the American Public Square Shifting Political Winds & Fallout from 9/11, Afghanistan, and Iraq.”https://dss.princeton.edu/catalog/resource1199.

36.

Redersdorff

Sandrine

Martinot

Delphine

Branscombe

Nyla

. 2004. “The Impact of Thinking about Group-Based Disadvantages or Advantages on Women’s Well-Being: An Experimental Test of the Rejection-Identification Model.” Cahiers de Psychologie Cognitive 22(2):203–22.

37.

Robinson

Greg

. 2009. By Order of the President: FDR and the Internment of Japanese Americans. Cambridge, MA: Harvard University Press.

38.

Rodgers

Harrell R.

1974. “Toward Explanation of the Political Efficacy and Political Cynicism of Black Adolescents: An Exploratory Study.” American Journal of Political Science 18(2):257–82.

39.

Rosenbaum

Paul R.

2002. Observational Studies, Springer Series in Statistics. New York: Springer.

40.

Rosenbaum

Paul R.

Rubin

Donald B.

1983. “The Central Role of the Propensity Score in Observational Studies for Causal Effects.” Biometrika 70(1):41–55.

41.

Rosenberg

Morris

. 1954. “Some Determinants of Political Apathy.” Public Opinion Quarterly 18(4):349–66.

42.

Rubin

Donald

. 1978. “Multiple Imputations in Sample Surveys: A Phenomenological [Bayesian] Approach to Nonresponse.” Pp. 20–28 in Proceedings of the Survey Research Methods Section. Boston: AMSA.

43.

Rubin

Donald B.

2001. “Using Propensity Scores to Help Design Observational Studies: Application to the Tobacco Litigation.” Health Services and Outcomes Research Methodology 2(3–4):169–88.

44.

Salamon

Lester M.

Van Evera

Stephen

. 1973. “Fear, Apathy, and Discrimination: A Test of Three Explanations of Political Participation.” The American Political Science Review 67(4):1288–306.

45.

Saunders

Clare

Grasso

Maria

Olcese

Cristiana

Rainsford

Emily

Rootes

Christopher

. 2012. “Explaining Differential Protest Participation: Novices, Returners, Repeaters, and Stalwarts.” Mobilization: An International Quarterly 17(3):263–80.

46.

Schildkraut

Deborah J.

2004. “Identity, Perceptions of Discrimination and Political Engagement: The Causes and Consequences of Reactive Identity Among Latinos.” Pp. 1–31 in Conference Papers—American Political Science Association. Washington, DC: American Political Science Association.

47.

Schildkraut

Deborah J.

2005. “The Rise and Fall of Political Engagement among Latinos: The Role of Identity and Perceptions of Discrimination.” Political Behavior 27(3):285–312.

48.

Schmitt

Michael T.

Branscombe

Nyla R.

2002. “The Meaning and Consequences of Perceived Discrimination in Disadvantaged and Privileged Social Groups.” European Review of Social Psychology 12(1):167–99.

49.

Schmitt

Michael T.

Spears

Russell

Branscombe

Nyla R.

2003. “Constructing a Minority Group Identity out of Shared Rejection: The Case of International Students.” European Journal of Social Psychology 33(1):1–12.

50.

Sellers

Robert M.

Caldwell

Cleopatra H.

Schmeelk-Cone

Karen H.

Zimmerman

Marc A.

2003. “Racial Identity, Racial Discrimination, Perceived Stress, and Psychological Distress among African American Young Adults.” Journal of Health and Social Behavior 44(3):302–17.

51.

Sellers

Robert M.

Shelton

J. Nicole

. 2003. “The Role of Racial Identity in Perceived Racial Discrimination.” Journal of Personality and Social Psychology 84(5):1079–92.

52.

Senzai. 2012. “ISPU Report: Engaging American Muslims: Political Trends and Attitudes.”http://www.ispu.org/GetReports/35/2457/Publications.aspx.

53.

Shingles

Richard D.

1981. “Black Consciousness and Political Participation: The Missing Link.” The American Political Science Review 75(1):76–91.

54.

Simon

Bernd

Grabow

Olga

. 2010. “The Politicization of Migrants: Further Evidence That Politicized Collective Identity Is a Dual Identity.” Political Psychology 31(5):717–38.

55.

Simon

Bernd

Klandermans

Bert

. 2001. “Politicized Collective Identity: A Social Psychological Analysis.” American Psychologist 56(4):319–31.

56.

Simon

Bernd

Ruhs

Daniela

. 2008. “Identity and Politicization among Turkish Migrants in Germany: The Role of Dual Identification.” Journal of Personality and Social Psychology 95(6):1354–66.

57.

Spencer

Steven J.

Zanna

Mark P.

Fong

Geoffrey T.

2005. “Establishing a Causal Chain: Why Experiments Are Often More Effective Than Mediational Analyses in Examining Psychological Processes.” Journal of Personality and Social Psychology 89(6):845–51.

58.

Stack

Liam

. 2015. “American Muslims under Attack.” The New York Times, December22.

59.

Stokes

Atiya Kai

. 2003. “Latino Group Consciousness and Political Participation.” American Politics Research 31(4):361–78.

60.

Stouffer

Samuel A.

Suchman

Edward A.

DeVinney

Leland C.

Star

Shirley A.

Williams

Robin M.

1949. The American Soldier: Adjustment During Army Life. Princeton, NJ: Princeton University Press.

61.

Tajfel

Henri

Turner

John C.

1979. “An Integrative Theory of Intergroup Conflict.” Pp. 33–47 in The Social Psychology of Intergroup Relations, edited by Austin

W. G.

Worchel

Pacific Grove, CA: Brooks/Cole.

62.

Tajfel

Henri

Turner

John C.

2004. “The Social Identity Theory of Intergroup Behavior.” Pp. 276–93 in Political Psychology: Key Readings in social Psychology, edited by Jost

J. T.

Sidanius

New York: Psychology Press.

63.

Thoits

Peggy A.

2011. “Resisting the Stigma of Mental Illness.” Social Psychology Quarterly 74(1):6–28.

64.

Tittle

Charles R.

Hill

Richard J.

1967. “The Accuracy of Self-Reported Data and Prediction of Political Activity.” The Public Opinion Quarterly 31(1):103–106.

65.

Traugott

Michael W.

Katosh

John P.

1979. “Response Validity in Surveys of Voting Behavior.” Public Opinion Quarterly 43(3):359–77.

66.

Uhlaner

Carole J.

Cain

Bruce E.

Roderick Kiewiet

1989. “Political Participation of Ethnic Minorities in the 1980s.” Political Behavior 11(3):195–231.

67.

Valentino

Nicholas A.

Brader

Ted

Groenendyk

Eric W.

Gregorowicz

Krysha

Hutchings

Vincent L.

2011. “Election Night’s Alright for Fighting: The Role of Emotions in Political Participation.” The Journal of Politics 73(1):156–70.

68.

Verba

Sidney

Burns

Nancy

Schlozman

Kay Lehman

. 1997. “Knowing and Caring about Politics: Gender and Political Engagement.” The Journal of Politics 59(04):1051–72.

69.

Verba

Sidney

Nie

Norman H.

1987. Participation in America: Political Democracy and Social Equality. Chicago: University of Chicago Press.

70.

Vickers

Jill

Isaac

Annette

. 2012. The Politics of Race: Canada, the United States, and Australia. Toronto: University of Toronto Press.

71.

Walker

Iain

Smith

Heather J.

2002. Relative Deprivation: Specification, Development, and Integration. Cambridge, UK: Cambridge University Press.

72.

Wimmer

Andreas

. 2013. Ethnic Boundary Making: Institutions, Power, Networks. New York: Oxford University Press.

73.

Wright

Stephen C.

Tropp

Linda R.

2002. “Collective Action in Response to Disadvantage: Intergroup Perceptions, Social Identification, and Social Change.” Pp. 200–36 in Relative deprivation: Specification, development, and integration, edited by Walker

Smith

H. J.

New York: Cambridge University Press.

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

0.11 MB