Of all Americans 33% believe they can make a difference by working to solve community problems.

Margin of error = ±2.2%

Identity, discrimination, and civic engagement

By Deborah J. Schildkraut, Professor of Political Science, and Jayanthi Mistry, Professor of Child Study and Human Development

The perception that one has faced discrimination promotes civic engagement. Viewing one’s racial identity as important or viewing being American as an important part of one’s identity generally does not.

These findings apply for behavioral outcomes (donating, canvasing, etc.) as well as for what we call “attitudinal engagement” (efficacy). The exception is when respondents are prompted to think in explicitly collective terms, such as when asked whether people can make a difference when they work together. When people are prompted to think specifically about their relationship to a larger group and its potential power, their racial identity and American identity matter more than perceptions of discrimination in promoting civic engagement.

These are findings from the Tufts Equity in America survey. They build on prior research that finds that discrimination can be a mobilizing agent, although discrimination has also been shown to lead people to withdraw from civic life (Oskooii 2016; Oskooii 2018; Schildkraut 2005; Pantoja, Ramirez, and Segura 2001; Ramirez 2013). And while some scholars maintain that social identities, or the sense of self that one derives from group memberships, particularly identifying as American, can be important for instilling a sense of civic duty and obligation (political participation being one such duty in the US), empirical results on the impact of national identity on engagement have not shown there to be a strong relationship (Smith 2003; Schildkraut 2011; Theiss-Morse 2009).

Before describing the analysis, a few notes are in order. First, this survey represents a random sample of Americans (n = 1,257), which means that the sample is majority white, non-Hispanic. Second, the survey was in field in the early months of the COVID-19 pandemic and as protests against police brutality toward Black Americans occurred nationwide (May 20 to June 10, 2020). Many people were confined to their homes and grappling with a threatening and rapidly changing context in which discussion of discrimination was prominent in the news. Third, the Equity survey only asked about perceptions of discrimination against oneself, not about whether one’s group is mistreated. Finally, this analysis presents a preliminary examination of the results; this work has not undergone blind peer review.

Below, we provide descriptive statistics on some of the key measures used in this analysis. Then we examine their relationship with one another.

To assess the salience and centrality of different possible social identities, respondents were asked, “Please indicate how important these characteristics are to your identity.” The options included, among other things, one’s racial and ethnic group and being American. The results in figure 1 show that respondents who are white and of 2 or more races are quite unlikely to say that their racial or ethnic identity is important to them, while respondents who are Black, Hispanic, and some other race say their racial identity is very or extremely important. This is especially so for Black respondents, with 51% saying that being Black is “extremely” important to their identity.

When it comes to how important being American is, there is little difference by race and ethnicity, as shown in Figure 2. Together, the results in these two figures fit well with earlier work on racial and national identification among white and nonwhite respondents (Jardina 2019; Schildkraut 2019, 2011).

Our survey also asked people about their perceptions of discrimination. Some questions asked people about experiences in the past 12 months, or major perceived discrimination, as defined by Williams (2012); others asked about experiences in their typical, everyday life, or everyday discrimination (Williams 2012). One scale was made for each set of questions, with each scale running from 1 to 4. (See measures section below for details.)

Figure 3 shows that Black respondents report the highest level of discrimination while white respondents report the lowest. It is notable, however, that the gap between white and Black perceptions of discrimination is not large. This too fits well with existing scholarship that indicates that a non-trivial, and possibly growing, proportion of white Americans perceive discrimination (Schildkraut 2019, 2014; Jardina 2019; Norton and Sommers 2011). In the remaining analyses, we focus on white, Black, and Hispanic respondents.

In what follows, we look at the relationships between perceived discrimination, racial and American identity, and five measures of civic engagement:

  1. Collaboration: Have you ever worked together informally with someone or some group to solve a problem in the community where you live? (yes/no)
  2. Canvas: Have you ever worked as a canvasser – having gone door to door for a political or social group or candidate? (yes/no)
  3. Efficacy: How much difference do you believe you can personally make in working to solve the problems you see? (no difference/a little/some/a great deal)
  4. Collective efficacy: How much difference do you believe you and other members of your community can make if you work together? (no difference/a little/some/a great deal)
  5. Participation count: A count variable that runs from 0-9 that indicates how many activities participants say they have done in the past 12 months. (See Measures section below for details.)

As with other surveys that ask about these kinds of political engagement, we find that engagement levels are rather low. Thirty-one percent of respondents say they have worked with others to solve a problem in their community; 20% say they have worked as a canvasser; 33% say that they personally can make some (27%) or a great difference (6%) in addressing the problems they see; and the mean score on the participation scale is a tiny 0.81. The one question that fared well pertained to collective efficacy: 71% of respondents felt that when people in the community work together, they can make some (45%) or a great deal of difference (26%).

To assess the degree to which racial identity, American identity, and perceptions of discrimination are associated with these measures of civic engagement we ran a series of regression models (logit for collaboration and canvas). We controlled for age, education, income, and the respondent’s racial classification (using the categories depicted in figures 1 and 2, with “white” as the reference category).

The results across all models show that discrimination is consistently associated with increased civic engagement on all but one measure (collective efficacy). As figure 4 shows, as the frequency of perceived discrimination in the past 12 months increases, the likelihood of having worked with others informally to solve a community problem increases substantially. While a white or Hispanic person who has never experienced discrimination the past 12 months has only a 25% chance of this type of collaboration, a white or Hispanic person who experienced all types of discrimination frequently has a 72% chance. Black respondents show an equally impressive increase in engagement (19% to 63%)

Similarly, figure 5 shows that while a Black person who has never experienced discrimination in the past 12 months has only a 14% chance of having worked as a canvasser, a Black person who experienced all types of discrimination frequently has a 39% chance.

For both collaboration and canvassing, the measure of daily discrimination had an impact of similar magnitude as the measure of discrimination in the past 12 months.

The results also show that reported discrimination in the past 12 months is significantly associated with increased perceptions of individual efficacy (graph not shown), while everyday discrimination is not. When it comes to collective efficacy, neither measure of discrimination is significant, but both racial and American identification are: greater identity importance for either identity is associated with greater collective efficacy, as shown in figures 6 and 7.

Finally, we turn to our nine-point scale of political participation. Here we find that discrimination in the past 12 months (major experience of discrimination) is associated with more participation, while the everyday discrimination measure is not. Figure 8 shows that as people’s discrimination score goes from its lowest to its highest value, they are expected to engage in one additional act of participation.

Overall, these results tell a fairly consistent story: Perceived discrimination is associated with greater levels of civic engagement. This result holds true for actual engagement (canvassing, donating, working in one’s community) and for attitudinal engagement as measured through efficacy. At the same time, one’s racial identity and American identity rarely matters for civic engagement. The normative implications of these findings are far from straightforward. All things being equal, we care both about promoting high levels of civic engagement and creating a society where people do not feel that they encounter discrimination regularly. Civic engagement is essential for creating awareness about injustice and momentum for meaningful policy change. It makes sense that perceptions of discrimination could foster motivation to become engaged. But these findings force us to confront the possibility that two normative goals (high participation and low discrimination) may be incompatible. Our results suggest that lack of engagement could be a result of complacency rather than apathy or disillusionment (though apathy and disillusionment could certainly be separate factors that shape engagement levels). As the Black Lives Matter protests across the country these past few months illustrate, perceived discrimination can be a powerful motivator to get involved. It also may be the case that perceptions of discrimination can help overcome racial or ethnic disparities in engagement (Fraga 2018; Holman 2016). In the aggregate, Black and Hispanic respondents perceive more discrimination than white respondents, which means that the proportion of Black and Hispanic people moved to act as a result of such perceptions could be significant.

Finally, we want to draw attention again to the fact that the only measure of civic engagement where discrimination did not matter was collective efficacy. That was also the only dependent variable where one’s group identity matters and the only dependent variable that asks people to think about the group. This finding suggests that if we want to see higher levels of civic engagement that are not predicated on discrimination, focusing messaging on the power of collective action could be a fruitful approach.


All references to white and Black respondents are for non-Hispanic white and Black respondents. Hispanic respondents can be of any other race. In the unweighted sample, 893 respondents are white, non-Hispanic; 117 are Black, non-Hispanic; 151 are Hispanic; 60 are “other”; and 46 are two are more races. Using weights provided by Ipsos, the resulting sample is 63% white, non-Hispanic; 12% Black, non-Hispanic; 16% Hispanic; 7% “other race”; and 1% two or more races. All results presented here employ the survey weights.

For discrimination in the past 12 months, respondents were asked: You have been unfairly stopped, searched, questioned, physically threatened or abused by the police; You were mistaken for someone else of your same race/ethnicity (who may not look like you at all); You have been unfairly prevented from having access to a service or been treated unfairly by a service provider. Response options were: never, not in past year but in the past, once or twice, or three or more times. For everyday discrimination, respondents were asked: Being treated with less courtesy or respect than other people; Feeling that people act as if they are afraid of you; Receiving poorer service than others in restaurants and stores. Response options were: never, rarely, sometimes, frequently. The correlation between the two scales is 0.47. Although our survey also asked people what they thought the reason was for their discrimination, we do not use those questions here, as discrimination based on race, class, gender, and immigration status is often intertwined.

The forms of participation included in the count were: Attended a political protest or rally; contacted a government official; volunteered to work for a presidential candidate; volunteered to work for a political campaign; volunteered to work for a political party; written a letter or email to a newspaper; commented about politics on social media; signed a petition; donated to a political party or candidate.

  • Fraga, Bernard L. 2018. The Turnout Gap: Race, Ethnicity, and Political Inequalityi in a Diversifying America. New York, NY: Cambridge University Press.
  • Holman, Mirya R. 2016. “The Differential Effect of Resources on Political Particiaption Across Gender and Racial Groups.” In Distinct Identities: Minority Women in U.S. Politics, edited by Nadia E. Brown and Sarah Allen Gershon, 13–28. New York, NY: Routledge.
  • Jardina, Ashley. 2019. White Identity Politics. New York, NY: Cambridge University Press.
  • Norton, Michael I, and Samuel R Sommers. 2011. “Whites See Racism as a Zero-Sum Game That They Are Now Losing.” Perspectives on Psychological Science 6 (3): 215–18.
  • Oskooii, Kassra A.R. 2018. “Perceived Discrimination and Political Behavior.” British Journal of Political Science 50 (3): 1–26. https://doi.org/10.1017/S0007123418000133.
  • Oskooii, Kassra AR. 2016. “How Discrimination Impacts Sociopolitical Behavior: A Multidimensional Perspective.” Political Psychology 37 (5): 613–40. https://doi.org/10.1111/pops.12279.
  • Pantoja, Adrian D, Ricardo Ramirez, and Gary M Segura. 2001. “Citizens by Choice, Voters by Necessity: Patterns in Political Mobilization by Naturalized Latinos.” Political Research Quarterly 54 (4): 729–50. http://prq.sagepub.com/content/54/4/729.short.
  • Ramirez, Ricardo. 2013. Mobilizing Opportunities: The Evolving Latino Electorate and the Future of American Politics. Charlottesville, VA: University of Virginia Press.
  • Schildkraut, Deborah J. 2019. “The Political Meaning of Whiteness for Liberals and Conservatives.”
  • 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.
  • ———. 2011. Americanism in the Twenty-First Century: Public Opinion in the Age of Immigration. New York: Cambridge University Press.
  • ———. 2014. “White Attitudes about Descriptive Representation in the United States.” In .
  • Smith, Rogers M. 2003. Stories of Peoplehood: The Politics and Morals of Political Membership. Cambridge: Cambridge University Press.
  • Theiss-Morse, Elizabeth. 2009. Who Counts as an American? The Boundaries of National Identity. New York: Cambridge University Press.
  • Williams, David R. 2012. “Measuring Discrimination Resource.” https://scholar.harvard.edu/files/davidrwilliams/files/measuring_discrimination_resource_feb_2012_0_0.pdf.