What Can Bring Christians and Muslims Together?

Results from a large-scale, real-world intervention study to increase harmony.

Posted Aug 23, 2020

 Photo by R M from Pexels
Salma Mousa tested intergroup contact theory in displaced peoples in Iraq.
Source: Photo by R M from Pexels

In 2014, the Islamic State of Iraq and Syria (ISIS) was killing members of religious minorities in Iraq, including Yezidis, Shia Muslims, and Christians. The thousands of Christians who were displaced had difficulty trusting Muslim neighbors, who some believed were complicit in attacks. After the war, Salma Mousa wanted to understand how trust could be rebuilt among Christians and Muslims in Iraq. So she turned to a great group unifier: sports.

Soccer is popular among both Christians and Muslims in Northern Iraq, and in many ways soccer teams are an ideal way to test theories developed in U.S. based research labs about intergroup relations. Psychological theory suggests that intergroup contact will be more effective at creating better relations between the groups when (1) participants are on an equal footing, (2) the activity is endorsed by community authorities, and (3) participants have a common goal. Soccer teams with a mixture of Christian and Muslim players check all the boxes.

So Mousa and her research team contacted captains of local Christian soccer teams in Northern Iraq and asked if they’d be willing to participate in a study of people displaced by war. Importantly, they agreed to be randomly assigned to one of two research conditions: continue playing on an all-Christian team, or have three Muslim players—who had also been displaced by ISIS—added to the team.

Mousa measured behaviors and attitudes before the study, after the soccer season, and six months later. She also made a key distinction between on-the-field variables—which would show whether contact with the Muslim players increased Christian acceptance of Muslim soccer players—and off-the-field measures—which would show whether contact with Muslim players increased Christian acceptance of Muslims more generally. This distinction is important, because one of the larger goals of intergroup contact research is to find policies that can create general harmony between groups, not just create good relationships between the people who got to know each other.

 Photo by Jack Sparrow from Pexels
Soccer is an ideal place to test intergroup contact theory.
Source: Photo by Jack Sparrow from Pexels

Before I get to the results, it’s worth pointing out the strengths of this research. First, this is research that matters. This was addressing a pressing social issue, and could do real good in the lab. As I’ve argued previously, psychology needs to be doing a lot more of this. Second, many of the outcomes measured are real behaviors that non-psychologists inherently care about, like voting for a non-Christian for a league-wide award or going to a social gathering with Muslims. This is a much stronger way of testing theory than many artificial lab tests where, for example, how close a participant pulled their chair to a participant is taken as the primary indicator of acceptance. Third, the study methodology and analysis was preregistered, so we can be sure that the researchers are not selectively reporting the results that best fit their pet theory.

Finally, as Betsy Levy Paluck and Chelsey Clark argue in an essay about this study, Mousa’s research contributes to the “basic science” of intergroup relations. As Paluck knows, research that is done outside the lab is often denigrated by lab researchers as being less important, because it’s “just” applying things that we already know to a real-world context. But theories of intergroup contact have not previously been definitively tested using (as Paluck and Clark put it) “the most robust research methodologies.” If the difference between Basic and Applied research is that in Basic research you are still finding new and puzzling things that can inform and change our underlying theories, and in Applied research you’re just confirming that existing findings and theories are true, then this study should be classified as Basic.

So what is the puzzle in this study’s results? First, the good news. Christians assigned to teams with Muslims were more likely to later train with Muslims six months after the season was over, to vote for a Muslim player for a sportsmanship award, and to be willing to sign up for a mixed Christian-Muslim soccer league for the next season. The biggest jump was in training with Muslim players after the season (49% more likely), and it does not just represent training with teammates. Muslims were recruited to train with 15% of the teams in the treatment group. For all the behaviors related to soccer, the intervention worked.

The puzzle is that these effects did not extend to off-field behaviors. Christians who played soccer with Muslims were not more likely to visit a Muslim owned restaurant (which all players were given a gift card to), were not more likely to attend a mixed Muslim-Christian social event they were invited to, nor more likely to donate to a charity that supported both Muslims and Christians (as opposed to a Christian-only charity). Mousa also analyzed personal attitudes, finding that playing with Muslims increased Christians’ abstract feelings of national unity, but did not change how comfortable they said they were with Muslim neighbors or how much they blamed Muslims for Christian suffering. Playing soccer together was able to bring people together for sports-related purposes, but it was not able to bring people together in the wider community.

 Photo by Jack Sparrow from Pexels
The theoretical puzzle of Mousa's results is how to expand results to off-field behavior.
Source: Photo by Jack Sparrow from Pexels

Mousa also looked at “spillover effects,” or positive influences on people who weren’t directly targeted by the study. Just being in a league where other teams had Muslim players increased positive intergroup behaviors in players, relative to players in another league where no teams added Muslim players. Fans of the mixed league also reported seeing group divisions as more arbitrary. This suggests some positive changes can result from just seeing successful integration.

Overall, this study should give us hope that getting people from different groups to work together can help create harmony. Further, it should give us hope that science done rigorously—with preregistered plans that are reported fully—can yield impressive results. But it also provides us with necessary realism. Getting members of different groups to work together isn’t a silver bullet; positive contact on the field didn’t translate to positive behavior more broadly. As Paluck and Clark point out, figuring out why is an important theoretical puzzle. Maybe the Muslim players were seen not as typical Muslims, but as exceptional Muslims—and so their positive characteristics weren’t expected in other Muslims. Follow-ups on this basic science question are needed. For now, we can be happy that psychology research has created at least some increased harmony in the world.


Mousa, S. (2020). Building social cohesion between Christians and Muslims through soccer in post-ISIS Iraq. Science, 369(6505), 866-870.

Paluck, E. L., & Clark, C. S. (2020). Can playing together help us live together?. Science, 369(6505), 769-770.