In this “Conversation with Authors,” we spoke to Ala’ Alrababa’h (AA) and Alexandra Siegel (AS), two of the authors of a recent APSR open access article “Can Exposure to Celebrities Reduce Prejudice? The Effect of Mohamed Salah on Islamophobic Behaviors and Attitudes”.

APSR: Can you discuss the aims of the paper, and how the idea for this paper was born?

AA: Well, some of us are football fans and have been watching Salah since even before he joined Liverpool, back when he was in Roma (AS Roma FC). And then, when he joined Liverpool, there was this big hype around him, and we talked about writing this paper when he joined and started playing well. But what really did it for us is when these videos of Liverpool fans started coming out of them singing about Salah and his religiosity – like, “If he scores another few/I’ll be Muslim, too!” which really made us think something was there and decided to actually work on this project. 

AS: And just to add on to that, we had been interested in studying prejudice reduction in various contexts for a long time, and this seemed to be this real-world example of prejudice reduction in action, at least anecdotally. We are all big fans of systematic data-oriented empirical work, and so we thought: Can we take this to the data and find evidence for this? Or is this just a song that fans are singing, an interesting phenomenon but it doesn’t go beyond that? I think on the one hand there’s this tendency to dismiss research that’s about celebrities or leisure activity, but day to day, most people don’t actually come across political activity or intergroup contact in the ways that we’re often manufacturing in a laboratory setting or trying to study as social scientists. And here, you have people being incidentally exposed to a celebrity member of an outgroup, giving us a real-world opportunity to test this long-standing hypothesis about intergroup contact – in particular, parasocial contact, or that positive exposure to a celebrity or a fictional character from a member of a stigmatized out-group can actually generalize towards positive attitudes towards that group as a whole.

AA: I completely agree with this – and just to briefly to add a slightly different point on this: We were not sure if there was going to be an effect or not. As Alex mentioned, we study these topics, and we know changing attitudes and reducing prejudice is extremely difficult.

APSR: In this paper, you marshal a lot of different data – rates of hate crime from police precincts across the country, posts on Twitter, and a survey experiment conducted on Facebook. Can you tell us a little about why you used all three of these and any challenges you faced along the way?

AS: In this case, particularly because the initial observational findings were so striking – that we saw the reduction of hate crimes in Merseyside and a decrease in Islamophobic tweets by self-described Liverpool supporters in the period after Salah’s signing – we really wanted to make sure we had this right and were not missing something. We initially designed our survey experiment with some ideas in mind regarding the mechanisms, but we hadn’t fully immersed ourselves in some of the existing literature on the parasocial contact hypothesis. So, we really had to drill down and figure out how to interpret our three pieces of evidence as a whole, and while in some ways that was a challenge, I think it’s ultimately what makes the research most compelling. Any one of these pieces of evidence individually might not convince you that something systematic is going on here, but when you look at them together they start to tell this story and also serve as a really nice jumping off point for more explicitly testing some of the more specific mechanisms and ideas that we propose throughout the paper.

APSR: When you were writing this piece, was there something that really surprised, confounded, or excited you?

AA: Simply the fact that we found this effect was striking. We know from the literature that prejudice reduction is extremely difficult, so the fact that there was an effect in this direction was really remarkable and why we spent a lot of time working on additional analyses, robustness tests, and so on.

Having said that, there aren’t that many studies of the effect of celebrities in the real world. There are some experiments, but there isn’t that much in the literature, and in many ways, Salah is the ideal case – one, he’s been amazingly successful as a player during this study period. Also, his identity is very salient – his name is Mohamed, he goes down in prayer after goals, fasts during Ramadan, his daughter is called Makka, his wife wears a hijab – so his identity is very salient, and if there were going to be an effect somewhere, it would be in this particular case.

AS: Just to add on to this, there was an environment of intensely positive media coverage of Salah, beyond just fans’ personal exposure. And we think that this media coverage may also contribute to this ideal environment for prejudice reduction. 

APSR: You mention that Salah is sort of an ideal case, even though there have been many Muslim successful soccer players in prominent leagues in the past–including Karim Benzema (Real Madrid), Mesut Özil (Real Madrid & Arsenal), Paul Pogba (Manchester United and Juventus), etc.  What do you think makes Salah so unique, and might cause this particular effect?

AA: That’s a great question, and unfortunately I think we can only speculate. There’s nothing in the paper to test what it is about Salah in particular that makes his identity very salient. I do think, again, compared to many other Muslim players, his identity is quite salient. I think what Alex mentioned about the positive media coverage plays a role – both coverage of how he was playing on the field and what he was saying and doing off the field. You mention Mesut Özil, there were many controversies about things he has said and done, and the media coverage on him was not always positive. And this is also the case with many other Muslim players.

AS: We can’t adjudicate between the different aspects here, but something for future research is that – in addition to his obvious success on the pitch and the positive media coverage – he’s not taking public political stances related to his identity or otherwise. And this is not to say that normatively we think that celebrities shouldn’t take controversial or political stances, but at least anecdotally we see often when they do this is followed by a barrage of negative media coverage. So one interesting question is how long can any popular figure sustain the positive coverage if we think the positive coverage is a key prerequisite for the parasocial contact to generalize positive attitudes towards an outgroup.

APSR: What is sort of interesting is that as someone who is a fairly avid follower of the Premier League, I don’t know that I’ve ever heard Mo Salah speak other than in post-match interviews! He just is an excellent footballer, and typically I have a fairly non-political experience with him.

AA: One thing Alex mentioned is that to the degree that there’s existing evidence from other studies that that’s thought to be an important precondition as well is likeability – it can’t just be how well you know a celebrity of fictional character, they have to be likeable. And how we quantify this is, of course, very, very challenging, but as you mention, a lot of people have this kind of positive connection to Mo Salah, and it feels like he’s very friendly and likeable.

APSR: One thing you mention in the piece that’s so interesting is the way that Liverpool fans have made Salah’s Muslim identity part of the way they express their support for him. Do you think that there’s some way that people’s identity as Liverpool fans might be affecting the outcomes you observe – is there something about being a Liverpool fan in particular?

AA: Again, this a great question, but we really cannot answer with this data right now. I mean, you’re a soccer fan – you know how much people identify with a team. Many people grow up with these teams; it’s sometimes even inherited from your parents, and it’s a really huge part of your identity. And that never changes, whether the team is doing good or doing bad. Liverpool  hadn’t won the league for a long time until Salah joined – they did win a Champion’s League in 2005, but they had never won the Premier League until Salah came, and he played a huge role in that.

There’s this great paper that’s forthcoming at APSR (Foos and Bischof, 2021) that looks at the effect of the Sun and attitudes of people in Liverpool towards Brexit, so that might have made it more likely to see an effect there. But one thing we mentioned in the paper is that prejudice against Muslims has been increasing around the world, not just in the UK, so we think this finding is particularly impactful – and it may be because Liverpool fans are particularly attached to their team. And if you’re so attached, you will strongly identify with a very successful player.

APSR: This was actually my next question! When I was reading this paper, I was thinking about what the opposite effect might be – when minority players lose or their performance suffers in a high salience situation. For instance, over the summer, England lost to Italy in the European Cup finals, with the last three penalty shots missed by three Black players (Marcus Rashford, Jadon Sancho, and Bukayo Saka). Do you worry that in the same way that their successes can reduce racial prejudice, minority players failure in high leverage situations might increase negative attitudes towards them?

AA: The example you mention is a really good one, in part because we saw some awful reactions to England’s loss, but we also saw some people defending them. Our paper allowed us to estimate the effect of Salah, but there is not as much that we can do about external validity. We did take a look at the effect of Sadio Mané (Editor’s note: Sadio Mané is a Senegalese Muslim player who also plays for Liverpool), and we did find some effect of his signing – not on hate crimes, but in decreasing anti-Muslim tweets among Liverpool supporters. But, as our conversation right now is showing, this paper really left us with way more questions than when we started the project, which is why we are working on a follow up project examining many more players, and the effect of both overperforming and underperforming. It’s been a really fun experience, working with Alex, Salma, and Will, and while we answer some questions, it also really led to more questions, which is why we are doing this follow-up project.

AS: And I think that the policy implications really come into play when we start to answer those questions. How durable these effects are, under which conditions do you see positive effects or backlash – especially if the media coverage turns. That is ultimately what we’re curious about understanding in the long run.

– Ala’ Alrababa’h, ETH Zurich

– Alexandra Siegel, University of Colorado Boulder

Alrababa’h, Marble, Mousa and Siegel’s APSR paper ‘Can Exposure to Celebrities Reduce Prejudice? The Effect of Mohamed Salah on Islamophobic Behaviors and Attitudes‘ is published as Open Access.