News Release

Study: survey methodology should be calibrated to account for negative attitudes about immigrants and asylum-seekers

Reports and Proceedings

George Washington University

WASHINGTON (May 2, 2023)- As of 2021 it is estimated that world-wide the number of refugees and asylum-seekers is over 31 million. While academics and researchers have been attempting to capture global and domestic attitudes towards these vulnerable groups, many of these studies will report biased findings and influence public opinion and policy. That is why researchers surveying socially charged topics such as immigration must make sure their methodology doesn’t reinforce common anti-immigration attitudes.

A team led by George Washington University researchers has done just that.

Today, Philip J. Candilis, M.D. a professor of psychiatry at the George Washington University School of Medicine & Health Sciences, will present the findings of a new study that debunks previous research showing that mere exposure to refugees  increases hostility towards refugees and asylum policies. With the collaboration of Allen R. Dyer, M.D. Ph.D, Sean D. Cleary, Ph.D. and former GW medical student Sarah Qadir, the research was an international joint project of the George Washington University Global Mental Health Program and the Hellenic American Psychiatric Association, with data from a survey of nine non-governmental organizations working in Greece.

“Because these types of studies can be influential, we felt it was important to demonstrate that a previous study of anti-immigrant attitudes in the Greek Aegean islands was flawed,” Candilis said. “Without a carefully chosen target population and carefully structured questions, surveys can provide incomplete and biased responses – responses that then inflame politics and policy and even fuel anti-refugee sentiment. That is why we choose to study a group that has maximum exposure to refugees – humanitarian aid workers -- so our results drew very different conclusions.”

Using methods intended to mitigate bias through survey design, sample selection, and statistical modeling, Candilis said the team tested whether humanitarian aid-workers in Greece, a group that is heavily exposed to refugees, would remain tolerant, committed, and progressive in their views of the immigrant population. Their results showed that it is not primarily exposure to refugees, but factors arising from context, background, and education which held sway in determining attitudes. The study demonstrates that a combination of better techniques can result in more informative and less divisive survey results on hot-button issues, Candilis said.

The abstract, “Antiracism Ethics & Research in the Refugee Crisis: A Greece/US Collaboration,” was presented May 2 at the 6th Annual International Forum on Ethics of the Athens Institute for Education and Research held in Athens, Greece.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.