Hallin, N., Västfjäll, D., & Andersson, G. (2020). Bias against atheists and religious persons in Sweden. Journal for the Cognitive Science of Religion 5(2), 205–223. https://doi.org/10.1558/jcsr.41002. The article erroneously states on page 215: “Swedish participants made significantly fewer conjunction errors than Finnish participants when the target was atheist.” The correct text should read: “Swedish participants made significantly more conjunction errors than Finnish participants when the target was atheist."

Background Religion is an important ingroup characteristic for many people. For different reasons, people with different religious affiliations might prefer members of their religious outgroup. Previous studies have investigated perceptions of and behaviour toward religious ingroup and outgroup members in various contexts. The four studies presented here investigated whether competence and likeability ratings differ depending on the target’s and participant’s religious affiliations in a recruitment context. Two studies were conducted in Sweden, while the other two were conducted in the USA. Methods Participants in 4 studies rated a Christian, Muslim or atheist job applicant and a control applicant on 4 competence and 3 likeability items on 7-point Likert scales. The difference in ratings between the target applicant and control applicant was used to measure perceived competence and likeability of the target applicant. In the two latter studies, one in Sweden and one in the USA, participants also chose to hire either the target or the control applicant. Results Overall, participants in three studies rated control applicants as more likeable than target applicants. In the two US studies, targets were also rated as less competent than control applicants. Christian participants in the two US studies rated the Christian applicant as more likeable than both other targets. In the second US study, atheist participants rated Christians as less likeable than both other targets. In one of the Swedish studies, atheist participants rated the atheist applicant as more likeable than both other targets. The only significant difference in competence ratings between targets was made by Christian Swedes, who rated Muslim applicants as less competent than Christian applicants. The only significant difference in hiring decisions was that Swedish atheist participants hired Christians less often than they hired control applicants. Conclusion Together, the results suggest that job applicants are sometimes viewed as more likeable if they belong to a religious ingroup rather than a religious outgroup, but that this only rarely translate to significant differences in competence ratings or hiring decisions.

Religion is viewed as necessary for moral behaviour in many parts of the world. Today, however, many societies are becoming more secular and the view of atheists as immoral might therefore change. This study investigated anti-atheist and anti-religious bias in Sweden, by replicating Gervais et al. (2017). The study used an online version of the conjunction fallacy test, which investigates whether a description, in this case of extremely immoral behaviour, is viewed as representative of a category or group. In this study we chose atheists and religious believers as targets. The sample consisted of 268 Swedish participants. The results showed no significant difference between conditions. However, the anti-atheist bias was lower than in the original American sample from the replicated study, while the anti-religious bias was higher. Further research can investigate differences between cultures, using the conjunction fallacy test.

The aim of this study is to investigate religious autobiographical memories by having self-reported atheist, Christian, and religiously uncommitted Swedes perform the Autobiographical Memory Test (AMT) with added religious word blocks. The specific aims are to investigate (1) possible group differences in memory specificity, (2) whether positive or negative cue words evoked a larger number of specific memories, and (3) whether participants produced a larger number of specific memories in response to religious or nonreligious cue words. Sisty participants were included, with twenty in each group (atheists, uncommitted, and Christians). No group differences in memory specificity were found. However, positive and nonreligious cue words were associated with a larger number of specific memories. The possibility of using AMT to study cultural differences is discussed.