Ammar Hammasho, a migrant from Syria who lives in Cyprus, kisses one of his four children after they arrived with their mother at a refugee camp in Kokkinotrimithia in the eastern Mediterranean island on Sept. 10, 2017. (Petros Karadjias/AP)

Since President Trump first announced his candidacy, he has been opposed to immigration, including that of refugees — a stance he has put into action after taking office, announcing in January a ban on the entry of all refugees for 120 days and an indefinite ban on all Syrians. Two weeks ago, in the latest twist in the saga, the U.S. Court of Appeals for the 9th Circuit allowed partial enforcement of the ban.

Notably, when Trump talks about refugees, he repeatedly says their ranks are dominated by young men. The concern about dangerous young men has been with us for a while. In 2015, discussing the refugee crisis in Europe, Sen. Ted Cruz (R-Tex.) argued that “77 percent of those refugees were young men.” In the same year, Bill O’Reilly emphasized that he would support admitting refugees into the country if he could add this caveat: no single men.

What these politicians are voicing is the American concern that young Muslim men pose security threats.

But who are refugees, really? The State Department’s Refugee Processing Center finds that, in fact, arrivals of Syrian refugees are about equally divided between men and women. What’s more, from Jan. 1 to Nov. 1 of this year, more than half the applicants were children, with 43.4 percent younger than 14 and 11.9 percent between the ages of 14 and 20.

That composition matters for how Americans think about refugees. Our recent unpublished research suggests that, when encouraged to think about how many refugees are women and children, Americans are less likely to see them as security risks. These findings are consistent with research showing that the way the news media cover debates about Islam, terrorism or refugee policy profoundly shapes public opinion.

Here’s how we did our research

We wanted to see how much knowing about refugees’ gender and religion influences American support for accepting them into the country. Through Amazon.com’s Mechanical Turk service in late October, we conducted an experiment with 1,052 respondents to an opt-in Internet survey, who were randomly assigned slightly different versions of questions about the refugee population. The goal was to see how much framing the question in different ways influenced their attitudes.

In the baseline version, respondents were told:

In 2016 and 2017, the U.S. admitted 85,000 and 53,000 refugees respectively. In 2018, the U.S. plans to admit 45,000 refugees.

Respondents in the other groups received the same information — but with one of the clauses below added:

  • including many women and children.
  • including many young men.
  • including many Muslim women and children.
  • including many young Muslim men.

All respondents were then asked, “Should the number of refugees entering the U.S. decrease, stay about the same, or increase?”

This allowed us to see whether attitudes about refugees varied according to the information about gender, children and religion.

Americans look more favorably on accepting refugees who are women and children

Here’s what we found. As you can see in the figure below, when we emphasized women and children, people were 7.4 percent more likely to support increasing the number admitted than those who just got the paragraph without a descriptive phrase. That’s a statistically significant difference.

Emphasizing young men changed support only by a small, statistically insignificant amount, suggesting that young men may be what respondents had in mind when thinking about refugees. Similarly, when we mentioned Muslim refugees, support didn’t change — again, probably because Americans associate refugees with Islam. However, when we emphasized young Muslim men, support dropped for allowing in more refugees.


Americans are more likely to think young Muslim men are a national security risk

We also asked whether accepting refugees was a national security risk. Here the distinction is clearer. Respondents who read the version mentioning women and children, but not religion, were less likely to see accepting refugees as a national security risk.


That’s especially true for Republicans

We find especially strong effects among Republicans. For example, among Republicans who saw the version that mentioned women and children, 24 percent fewer said they wanted to decrease the number of refugees than those who read the unmodified paragraph. Even Republicans who saw the paragraph that mentioned Muslim women and children were 12 percent less likely to want to decrease the number of refugees allowed into the United States.

Similarly, 16 percent fewer Republicans identified refugees as a national security threat if what they read emphasized women and children — although that dropped to only a 2 percent decline for Muslim women and children.

Democrats were less influenced by an emphasis on women and children. Only one of the added phrases had statistical significance: Adding “Muslim women and children” led Democrats to be 8.5 percent more likely to support reducing the number of refugees into the country That stability is partly because Democrats are more likely to support refugees to begin with.

Timothy S. Rich is an associate professor of political science at Western Kentucky University. His primary research agenda focuses on the domestic and international politics of East Asia.

Maggie Sullivan is a Western Kentucky University undergraduate senior majoring in international affairs and gender and women’s studies who has worked resettling refugees.