When Do Americans “See Something, Say Something”? Experimental Evidence on the Willingness to Report Terrorist Activity

Abstract

Public reporting programs such as the “See Something, Say Something” campaign are important counterterrorism measures. Yet public knowledge about terrorism is low, and Americans tend to associate terrorist activity with Muslims and Middle Easterners rather than with Whites. The consequence may be biases in public reporting that lead to discriminatory law enforcement. Using data from a national survey experiment (n = 700), we examine how suspect ethnicity, ethnic stereotypes, and national identity intersect to affect willingness to report terrorism-related behaviors to the police. Our results reveal that reporting is not based on suspect ethnicity alone (White vs. Muslim Middle Eastern). Rather, the relationship between willingness to report and suspect ethnicity is conditional on endorsement of Muslim stereotypes and national identity. Our analyses also identify key antecedents of terrorism reporting, including strength of national identity, perceived procedural justice, and anger about terrorism. We conclude by offering recommendations for increasing the effectiveness of public reporting programs.

Keywords:

• Muslim stereotypes
• social identity theory
• terrorism
• willingness to report terrorist activity
• “if you see something
• say something”
• American identity

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1 YouGov interviewed 1224 respondents who were then matched down to a sample of 1000 based on gender, age, race, and education. The frame was constructed by stratified sampling from the full 2018 American Community Survey (ACS) 1-year sample with selection within strata by weighted sampling with replacements (YouGov, 2020).

2 When the Muslim subsample is included in the total sample, the key findings are similar for the treatment effects and Muslim Stereotypes, but American Identification no longer has a significant main or interactive effect.

3 We do not believe phrasing the question this way lead to social desirability bias, given that survey was anonymous and self-administered (Tourangeau & Yan, 2007), that the approach is no more direct than other experimental methods manipulating suspect ethnicity and/or religion (e.g., Huff & Kertzer, 2018), and that Blair and colleagues (2020) found that social desirability is generally not a problem in direct questions about racial prejudice.

4 The only item with a factor loading under .650 was the reversed-coded one. In supplementary analyses, we dropped this item from the Muslim Stereotypes index and reran the models. The results were unchanged.

5 Respondents who said they were “not sure” about their political ideology or party identification were coded as moderates and as independents, respectively.

6 Similar to previous research (Burton et al., 2021), we imputed missing data on income (about 10% missing). This did not change the main findings.

7 When the sample is disaggregated at the mean of Muslim stereotypes index, suspect ethnicity has a significant negative effect among those below the mean (b = ‒.433, p < .001) and a significant positive effect among those at or above the mean (b = .218, p = .032). For American identification the findings are similar, but only the coefficient for those below the mean is significant (b = ‒.265, p = .016 and b = .044, p = .641, respectively).