Religion and prosocial behaviour: a field test

Abstract

Religious people are thought to be more prosocial than nonreligious people. Laboratory studies of this using ultimatum, dictator, public goods and trust games have produced mixed results, which could be due to lack of context. This article examines the relationship between religion and prosocial behaviour using data from a context-rich, naturally occurring field experiment that closely resembles the dictator game – tipping in restaurants. Customers were surveyed as they left a set of restaurants in Richmond, Virginia, in the summers of 2002 and 2003. Our findings reveal no evidence of religious prosociality.

Acknowledgements

Funding for both waves of the survey data collection was provided by the National Science Foundation (SBE-0241935). We thank Orn Bodvarsson, Mana Komai and David Mitchell for their helpful comments. The authors alone are responsible for any errors.

Notes

¹See Norenzayan and Shariff (2008) for a review of experimental and survey studies of religious prosociality in the psychology and sociology literature.

²Tipping qualifies as a natural field experiment according to Harrison and List's (2004) taxonomy of field experiments. Subjects engage in a task they would naturally undertake and are unaware that they are participating in an experiment.

³Ruffle (1998) and Parrett (2006) used laboratory studies to examine tipping using dictator games with pie sizes determined endogenously by the recipient's effort with mixed results. Both find a positive tip–service relationship, but Parrett (2006), which is framed to more closely resemble a restaurant tipping setting, finds that this relationship is weak. Meta-analyses of field studies examining the effect of service quality on tip size find the same positive, but tenuous relationship (Lynn and McCall, 2000). That diners tip based primarily on some percentage of the bill is supported by Azar (2007) who, in a recent literature review, reported that bill size is the most important variable in determining dollar tip.

⁴Survey available on request.

⁵Our measure of religiosity (i.e. regular attendance at religious services) is similar to that used in many previous laboratory studies. Only Tan (2006) and Tan and Vogel (2008) use more extensive measures.

⁶We only report results for the significant variables and the religion variable. Regressions included all of the independent variables listed in Table 1 plus bill size squared, table size squared and customer age squared. Complete results available on request.

⁷Interacting religion and each of service quality, dining frequency, table size and bill size did not alter our conclusion.