Religion and Social Behaviors: Using Religious Reminders to Promote Honesty and Using Prompt-based Interactions to Reduce Bias Against Religious Minorities

Abstract

Religion is intertwined with social behaviors. While most of the world’s population claims an affiliation with some religion, little research exists on how multiple dimensions of religion and religious reminders impact social behaviors. Moreover, some minorities are discriminated against because of their religious affiliation. This dissertation is motivated by two primary questions that revolve around religion and social behaviors: 1) Can dishonesty be reduced through different types of religious reminders, and how do their effects compare with non-religious reminders? 2) To what extent do certain prompt-based interventions reduce bias towards religious others?

In this thesis, we present the findings from two studies. In the first study, we compare the impact of religiously-framed extrinsic and intrinsic reminders versus non-religiously-framed ones in promoting honesty in Sudan, a predominantly Muslim low-income country in sub-Saharan Africa. In doing so, we develop a theoretical framework of how religious beliefs affect prosocial behaviors and identify a set of mechanisms underlying religious-based motivators. A randomized control trial shows that, compared to the control condition that has a cheating rate of 50.4%, the religious extrinsic and intrinsic reminders increase honesty by 16.1 and 15.1 percentage points, respectively. Moreover, the non-religious intrinsic reminder marginally increases honesty by 10.8 percentage points compared to the control. By contrast, the non-religious extrinsic reminder does not increase honesty compared to the control potentially due to common beliefs that the government is corrupt in Sudan or that it has limited monitoring ability. Overall, the findings suggest that in contexts where governance is weak, but faith is widespread, religion-based mechanisms may be effective in increasing prosocial behavior.

In the second study, we examine the impact of two compact prompt-based interventions — perspective-taking and value-consistency — on reducing anti-Muslim bias in the United States. Perspective-taking entails having people reflect on others’ experiences of being discriminated against while value-consistency entails having people reflect on important values they hold, such as kindness or tolerance, which are consistent with bias reduction. We conduct two field experiments – one online and one in-person. We find that value-consistency increases trust behaviors as measured by the investment amount in the trust game (aka the investment game) on receivers with Muslim or non-Muslim sounding names. Value-consistency participants invest 7% and 13.2% more in their partners in the in-person and online experiments, respectively. Furthermore, we observe that value-consistency increases beliefs about the trustworthiness of Muslim and non-Muslim receivers and that perspective taking increases beliefs about the trustworthiness of Muslim receivers in the in-person study compared to their control counterparts. Unlike previous work that shows the effectiveness of perspective-taking on reducing bias against transgender individuals, we do not observe an impact of perspective-taking on reducing anti-Muslim bias.

Overall, the findings from these two studies suggest that reminders based on intrinsic motivations targeting self-concept can be a promising avenue to promote prosocial behaviors, as there are indications that they work in two very different contexts – Sudan and the United States – and for two different purposes – increasing honesty and reducing prejudice. On the other hand, we find that reminders based on extrinsic motivations are conditional on beliefs about the enforcing agent. Finally, we suggest that religious reminders can be an effective means to promote honesty in some contexts where religion is dominant, and institutions are weak or corrupt.