Give me your Christian, your female, your English-speaking with a good education? While not the words on the Statue of Liberty, these seem to be the kinds of refugees that the American public prefers –according to a new study by researchers at the University of California San Diego, the University of Wisconsin-Madison and New York University Abu Dhabi.
The study shows that religion is the most powerful source of discrimination. When you hold constant national origin, religion matters more than gender, age, fluency in English or professional skill. “On a scale of 1 to 7, among otherwise completely identical refugees from Syria, Muslims are rated a full half point lower,” said UC San Diego political scientist and co-author Claire Adida.
In addition to showing that anti-Muslim bias prevails across the board in the U.S., the study also shows that it differs across subgroups: The bias is weakest among those who self-identify as non-Christian, non-white and Democrat, compared to self-identified Republicans, Christians and whites.
Published in the journal PLOS ONE, the study details a survey experiment conducted on the eve of the 2016 presidential election with a nationally representative sample of the U.S. public. Amercan attitudes now, the researchers suspect, are likely to have become only more pronounced and polarized.
The experiment was conducted, through YouGov, with 1,800 U.S. citizen adults. It was focused on refugees from Syria in part because it enabled the researchers to hold constant national origin and assess the influence of other salient characteristics.
To arrive at their findings, the researchers used a statistical technique common in market research called “conjoint analysis.” They presented all survey respondents with pairs of refugee profiles that differed by age, religion, English fluency, gender and skill-level. They then asked respondents to rate both refugees on a scale of 1 to 7, from “the U.S. should absolutely not admit” to “the US should definitely admit” that refugee. Each respondent scored three pairs of refugees, for a study total of 5,400 pairs (or 10,800 refugee profiles). The key to identifying discrimination lay in the random assignment of refugee characteristics, allowing the researchers to isolate the effects of a single trait on ratings.
One surprise of the study, Adida said, is that there doesn’t seem to be a special penalty for being a Muslim man. “There is a penalty for being Muslim and a penalty for being male but not a separate special penalty for Muslim men,” Adida said. “The same ‘male penalty’ applies to Christian refugees.” This finding leads the researchers to conclude that perceived security concerns were probably not the main drivers of the respondents’ choices.
Co-authors on the study, besides Adida of UC San Diego, are: Adeline Lo, a doctoral alumna of UC San Diego, who worked on the research while a postdoctoral scholar at Princeton University and is now assistant professor of political science at UW-Madison, and Melina R. Platas, a doctoral alumna of Stanford, now an assistant professor of political science at NYU Abu Dhabi.
In a different earlier paper, published in the Proceedings of the National Academy of Sciences in 2018, the researches show that a perspective-taking exercise – asking Americans to imagine themselves as refugees – promotes inclusionary behavior toward Syrian refugees.
Next steps: Adida, Lo and Platas are now beginning a research project to determine how well the American public understands the differences between categories of migrants (immigrant, refugee and asylum seeker) and what they know about U.S. policy on refugees. In fiscal year 2020, the United States plans to admit its lowest number of refugees since Congress created the nation’s refugee resettlement program in 1980.
The present study is funded by a National Science Foundation RAPID Collaborative grant (SES-1503802) and a UC San Diego Academic Senate grant to Adida (RP56G).