On Sept. 11, 2001, the Islamist terrorist group al-Qaeda coordinated four attacks against targets in the United States, killing nearly 3,000 individuals and injuring tens of thousands more. After these attacks, Americans grew more suspicious of and outwardly hostile to Muslims. Research shows that these views increased in the years that followed. Our recently published paper in the Public Administration Review shows that — even 19 years later — public officials in the United States discriminate against Muslims.

How we did our research

We drew our conclusions from a type of experiment that researchers call an audit or correspondence study. The experiment was designed to measure whether U.S. public school principals would respond differently to families based on their religious beliefs. To do this, we sent emails to a sample of more than 45,000 public school principals divided evenly across the country. The emails were purportedly sent by a fictional family interested in sending their child to the principal’s school, and asked principals for a meeting. We randomly assigned the family a religious affiliation or lack thereof.

We did that by embedding a quote at the bottom of most emails, in the signature line, that read, “[ … ] teaches that life is precious and beautiful. We should live our lives to the fullest, to the end of our days.” We signaled the family’s religious views by including either “Christianity,” “Catholicism,” “Islam” or “Atheism” in the quote, with the quote attributed to the Rev. Billy Graham, Pope Benedict, the prophet Muhammad or Richard Dawkins, depending on the fictional family’s beliefs.

We also randomly varied how strongly we emphasized the fictional family’s religious beliefs. In some cases, the email indicated that the family wanted to find a school that was compatible with their beliefs; in others, the family noted that it expected accommodation for their beliefs.

For comparison purposes, some emails included no quote or reference to religion at all.

Overall, principals or their staff responded to our emails about 43 percent of the time, comparable to response rates from similar studies.