If everyone in the world believed in god except me, there wouldn’t be much of a reason to form prejudices against me. At worst you’d probably just think I’m crazy. I wouldn’t be perceived as a threat. I’d just be an eccentric member of society who doesn’t grasp something blatantly obvious to the rest of its members. But what happens when the population of non-believers rivals that of the world’s major religions?
Common sense — and, indeed, previous research — would suggest as the out-group populations grow, in-group populations form stronger prejudices about the out-group. But, interestingly, the opposite appears true for prejudices about atheism. In this episode of Science Sunday, I’ll be looking at Will M. Gervais’ “Finding the Faithless: Perceived Atheist Prevalence Reduces Anti-Atheist Prejudice,” published in Personality and Social Psychology Bulletin 37(4) 543–556, 2011.
Gervais conducts four studies to come to this conclusion: 1) What is the relationship between anti-atheist prejudice and believers in countries with robust atheist populations? 2) What is the relationship between atheist prevalence and anti-atheist prejudice? 3) Is there a causal relationship between reminders of atheist prevalence and atheist distrust? And 4) what is the relationship between prevalence information and atheist distrust?
To come to his conclusions he runs an assortment of various correlation analyses and multivariate regressions. For the sake of saving time and not littering this post with p-values, r-squares, and whatnot, I’ll summarize his findings. For the statisticians out there who want to see the results of his analyses, you can check his work on the “Publications” section of his website. Please note, Gervais posts his work for personal academic analysis, not for commercial purposes or for widespread dissemination. Please respect his requests.
Gervais looks at anti-atheist prejudices across 54 countries and concludes a negative correlation: As atheist prevalence in a country increases, the perception of anti-atheist prejudice decreases.
He then asks the same question, but instead of applying it to states generally, he brings it down to the individual level. With a sample size of 104 undergraduates, he finds a statistically significant negative correlation between atheist prevalence and anti-atheist prejudices. As atheist prevalence increases, individual prejudice against atheists decreases.
Studies 3 and 4
Using samples of 112 and 80 undergraduates, respectably, Gervais finds a negative correlation between reminders of atheist prevalence and atheist distrust. He writes, “Information that atheists are actually quite common, both worldwide and in the immediate environment, reduced distrust of atheists.” He furthermore finds, “As hypothesized, implicit distrust was lower in the atheist prevalence condition than in the control condition.”
Gervais has a very hearty discussion section in his article, but for the sake of this post, I will (largely) ignore his discussion and treat this as part of a literature review.
Last week I looked at research by Andrew S. Franks and Kyle C. Scherr that suggests atheists are the most disadvantaged group at the polls. In other words, if a self-proclaimed atheist is running for office in the US — and all other things equal — he or she suffers a deficit in support that greatly reduces the chances of them being elected. This deficit is the result, the study suggests, of the perception of atheists as 1) distrustful, 2) disgusting, and 3) threatening.
Gervais’ research, however, might offer us some insight into how to mitigate and possibly reverse Franks’ and Scherr’s findings. If, for example, distrust against atheists is lowered by simply reminding people that atheists are common, then a simple non-partisan media campaign by a secular organization prior to an election might reduce some of the deficit the atheist candidate suffers.
Furthermore, we might be able to run similar studies to find relationships between atheist prevalence information and Franks’ and Scherr’s other variables. If the dissemination of atheist prevalence information reduces distrust against atheists, might it also reduce the perceptions of atheists as threatening and disgusting? I hypothesize that reminding the public that atheists are prevalent will reduce the perception of atheists as threatening and disgusting.
Finally, of particular note, Gervais tackles the “coming out of the closet” issue regarding atheists. Instead of suggesting atheists risk social estrangement by publicly stating their atheism in an effort to reduce anti-atheist prejudice, he offers a novel approach: Simply publicize atheist prevalence numbers. He writes,
[T]he present results indicate that merely providing information about collective prevalence could reduce prejudice, obviating the need for individual disclosure.
In other words, atheists need not “out” themselves and risk stigma. Removing this obstacle allows concealed atheists to have their cake and eat it too. They can share the benefits of living in a society that is less prejudiced against atheists while keeping their social identity free of stigma until such a time that the stigma has been sufficiently reduced. This is a rather remarkable suggestion, which is why next week I will be reviewing another study by Gervais.