![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
Some research suggests that reminders of religious beliefs and concepts can decrease immoral behavior, such as cheating, via fear of supernatural punishment among other mechanisms. However, one of the most common natural religious primes, petitionary prayer, could in theory have the opposite effect, as it implies and asserts external attributions for behavior. We tested whether petitionary prayer, despite its association with religiosity, might nevertheless increase cheating and whether such effects would differ as a function of participants’ religious beliefs. American participants (N = 251) completed an online “Swahili translation” task that afforded cheating; half were asked to compose a prayer to improve their performance. Results showed that religiosity (measured as supernatural beliefs) was associated with a greater probability of cheating, as well as more extensive cheating among those that did cheat; prayer decreased the likelihood of cheating (but not its extent) among religious people only. Mediational analyses suggested that, counterintuitively, it was believers’ beliefs about God’s control, rather than about God’s capacity for punishment, that explained the effects.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 It is for just this reason that Pope Francis (Citation2017) recently proposed modifying “lead us not into temptation” to “don’t let me fall into temptation.” Pope Francis reasoned that “lead us not into temptation” suggests that God is responsible for temptation rather than humans or Satan.
2 Indeed, the eponymous “mechanical turk” was a clever device that cheated at chess.
3 Internal consistency compared positively to Rotter’s I-E scale (Levenson, Citation1973). The Kuder-Richardson formula 20 (KR-20) reliabilities ranged from mid .60’s to high .70’s. Split-half reliabilities (Spearman-Brown) suggested reliability of .60. Test-retest reliabilities ranged from mid .60’s to high .70’s.
4 A negative binomial model was used because it fit the data better than a Poisson model, as suggested by fit indices (i.e., AIC, BIC, log likelihood, Vuong tests, and rootograms).