Group Empathy as an Explanation for Variation in U.S. Intergroup Reactions to Humanitarian Emergencies Abroad and Refugee Crises
Download data and study materials from OSF
Cigdem V. Sirin
University of Texas at El Paso
Nicholas A. Valentino
University of Michigan, Ann Arbor
Jose D. Villalobos
University of Texas at El Paso
Sample size: 508
Field period: 05/28/2016-10/14/2016
Group Empathy Theory argues that similar life experiences can lead different groups to develop favorable policy attitudes toward one another even in the presence of economic competition and even perceived existential threat. In this study, we conduct a national survey experiment with a random sample of Anglos and stratified, random oversamples of African Americans and Latinos to test the premises of Group Empathy Theory in the context of foreign policy. We specifically examine intergroup attitudes toward the U.S.'s role in dealing with international humanitarian emergencies, foreign aid, the Syrian civil war, and the closed door-policy towards Muslim immigrants proposed by Donald Trump. In line with our theoretical expectations, African Americans and Latinos compared with Anglos express significantly higher group empathy in general and also specifically towards refugees, Arabs, and Muslims. African Americans and Latinos, again compared to Anglos, also attribute higher responsibility to the U.S. to protect other nations in need, are more supportive of foreign aid and assistance for Syrian civilians, and more strongly oppose Trump's Muslim ban. Differences in empathy for outgroups appear to mediate these racial/ethnic group differences in foreign policy attitudes concerning humanitarian crises abroad.
Our study explores whether empathic concern for foreign racial/ethnic groups suffering under an oppressive government explains opinion variation among domestic racial/ethnic groups in the United States. We expect that groups who have experienced various forms of deprivation, oppression, discrimination, and criminalization in the U.S. may support policy actions on behalf of other groups in need of help as a result of their empathic understanding of those circumstances. This should occur even when the foreign ethnic group is seen as threatening to American minorities, either economically, culturally, or existentially.
With respect to intergroup variations in empathy as a trait characteristic, we expect African Americans and Latinos have higher levels of general group empathy than Anglos (H1a). Based on their experiences as minorities, African Americans and Latinos should also have higher levels of group-specific empathy for outgroups, including refugees, Muslims, and Arabs (H1b). Given these baseline predictions about group differences in empathy, we expect significant group differences in pre-existing foreign policy attitudes. Specifically, we predict that compared to Anglos, African Americans and Latinos are likely to attribute higher responsibility to the U.S. for providing humanitarian assistance to people from other countries in need, all else equal, particularly during times of war and natural disasters (H2a). We also expect African Americans and Latinos to be more supportive of increasing the portion of the U.S. budget allocated to foreign aid (H2b). We further expect both minority groups to be supportive of a more proactive U.S. foreign policy in dealing with the Syrian civil war—including providing aid, asylum, and military assistance to help the victims of this violent strife (H2c). And when it comes to exclusionary foreign policy propositions under the disguise of homeland security as in the case of Trump's Muslim ban, we expect that African Americans and Latinos should display stronger opposition to such discriminatory policies (H2d).
Group Empathy Theory also predicts African Americans and Latinos, compared to Anglos, will react differently to news depicting civilian suffering under an oppressive regime. Specifically, African Americans and Latinos should be more likely than Anglos to support foreign policy actions including military intervention, humanitarian aid, and asylum (H3a). We also expect to observe higher potential for empathy-driven political behavior—requesting more information, making a donation, and signing a petition to help the victims—among African American and Latino respondents than Anglos (H3b).
We further predict that African Americans' and Latinos' support for foreign policy actions to help the foreign nation in distress will remain significantly higher than Anglos even for victim groups that may be seen as threatening to the U.S (H3c). In particular, we expect African Americans and Latinos to be more supportive of humanitarian action in an Arab country compared to a predominantly white European one, and more accepting of the refugees seeking asylum due to their higher group empathy even for outgroups who are generally stigmatized and labeled as a source of threat. We perform an experiment to test this hypothesis, randomly assigning the particular nation that is in need of aid in our news vignette.
Finally, our test provides a clear distinction between ingroup material interest hypotheses versus the group empathy mechanism we believe is at work. Under the group interest hypothesis, those who perceive higher costs and risks of certain policy actions should be least likely to support humanitarian interventions. In addition, if there are differences between racial/ethnic groups in terms of perceptions of individual or national costs and risks, the groups that perceive the highest costs/risks should be the ones that react most strongly against taking any foreign policy action to help outgroups in need. Group Empathy Theory predicts no simple expected utility or ingroup interest mechanism to be present and instead proposes that foreign policy attitudes related to humanitarian rights, needs, and emergencies abroad are significantly mediated by group empathy (H4).
We randomly assigned half of each of the three racial/ethnic respondent groups (whites, blacks, and Latinos) to one of two conditions. Participants first read a short hypothetical news report about the violent actions of an oppressive government in an island nation abroad, which was depicted either in (1) the Balkans or (2) the Arabian Peninsula, and also accompanied by a photograph of a young person (our racial/ethnic cue), who appeared to be either (1) European or (2) Arab, who was identified as one of many victims who died as a result of government oppression and police brutality.
To manipulate racial/ethnic cues, we chose photos that reflected the intended race/ethnicity while holding other traits as constant as possible. We accomplished this by selecting pictures from a pool of 40 photos based on ratings provided by an independent panel of eight judges naïve to our hypotheses. Judges rated how much each person appeared white, Arab, black, and Latino, as well as friendly, attractive, wealthy, law-abiding, educated, and trustworthy. We used pictures judged to be significantly different from one another on the race/ethnicity dimension—with perceptions of the race/ethnicity of the person in the picture matching the intended racial/ethnic cue for each experimental condition—but substantively similar and statistically indistinct across all other traits, to ensure the internal validity of our experiment.
After exposure to the experimental treatment, participants answered several questions about their reactions to the news report. The post-test questionnaire first included measures of empathy-driven policy preferences. We asked participants about how strongly they supported or opposed: (1) using U.S. military forces against the oppressive government of [Kornati/Farasan] in order to help the people experiencing violence; (2) the U.S. providing financial aid to the victims of this violent repression; and (3) granting asylum to refugees who attempt to escape from [Kornati/Farasan] to America.
We also included measures to capture participants' proclivity towards engaging in empathy-driven political behavior. Specifically, we measured the participants' willingness to (1) receive more information about the situation in [Kornati/Farasan]; (2) donate to a non-governmental organization that works to help the victims of the violent conflict in [Kornati/Farasan]; and (3) sign an electronic petition to be sent to Congress indicating their support for humanitarian aid to help the victims of the violent conflict in [Kornati/Farasan].
Summary of Results
With regards to empathy-driven policy attitudes, both African Americans and Latinos display higher support than Anglos for a U.S. military intervention against the oppressive regime depicted in the vignette in both European and Arab nation conditions. They are also more favorable towards the U.S. providing financial aid to the victims on average. That said, three racial/ethnic groups do not differ in their levels of support for granting asylum to the refugees of this hypothetical conflict regardless of the race/ethnicity of the nation depicted in the vignette. While we observe a significant main effect of race/ethnicity, we did not find a significant interaction between the race/ethnicity of the respondent and the race/ethnicity of the foreign nation in need. In other words, minority respondents in this experiment were as empathic towards the Arab nation as they were towards the European nation.
Race/ethnicity also appears to be related to differences in empathy-driven political behavior. Latinos showed significantly higher interest in becoming more informed about the conflict situation reported in the vignette than both Anglos and African Americans across both European and Arab conditions. Anglos were much more likely to donate their extra dollar in the European nation condition (.44) than the Arab condition (.37). We observed an opposite trend for African Americans—the marginal mean for African Americans was .26 in the European condition while it was .34 in the Arab one. Latinos' donation tendency parallels that of Anglos: the marginal mean for Latinos was .48 in the European condition and it was down to .41 for the Arab condition. However, we should note that in both conditions, Latinos displayed the highest propensity to donate their money to the victims of the conflict as compared to both African Americans and Anglos.
Both African Americans and particularly Latinos show a much stronger tendency in both conditions to sign a petition to be sent to the federal government than Anglos. This is especially remarkable given that minorities are often perceived to be more cautious about their dealings with the government agencies partly due to their experiences with being targeted or unfairly treated by law enforcement. However, they do not shy away from putting their names down for the record and signing a petition to help a foreign nation.
For more information and other relevant publications on Group Empathy Theory, please see:
Cigdem V. Sirin, Nicholas A. Valentino, and José D. Villalobos. "Group Empathy Theory: The Effect of Group Empathy on U.S. Intergroup Attitudes and Behavior in the Context of Immigration Threats." Journal of Politics 78, 3 (2016): 893-908.
Cigdem V. Sirin, Nicholas A. Valentino, and José D. Villalobos. "The Social and Political Consequences of Group Empathy." Political Psychology 38, 3 (2017): 427-448.
Cigdem V. Sirin, Nicholas A. Valentino, and José D. Villalobos. "Group Empathy in Response to Non-Verbal Racial/Ethnic Cues: A National Experiment on Immigration Policy Attitudes." American Behavioral Scientist 60, 14 (2016): 1676-1697.
Cigdem V. Sirin, Nicholas A. Valentino, and José D. Villalobos. Group Empathy Theory: How Empathy for Other Groups Affects Political Attitudes and Behavior. Book Manuscript (in Progress).
Cigdem V. Sirin, Nicholas A. Valentino, and José D. Villalobos. "Group Empathy Trumps Hate? Opinion Divergence on Refugees and the Muslim Ban." Presented at the Annual Meeting of the American Political Science Association, San Francisco, CA, August 2017. Also presented at the Annual Meeting of the International Organization of Social Sciences and Behavioral Research, Las Vegas, NV, October 2017.