I designed an experiment with my advisor to run on Amazon Mechanical Turk that tests the relative effects of uncertain normative and empirical social information on the rates of dishonest reporting. This project fits within the frame of behavioral economics and ethics, where current research shows that various types of social norm information (including empirical and normative), that is presented as factual, affects ethical decision making differently.
However, most people do not receive definitive statistics about the behavior or thoughts of other people in their same situation as they consider whether to act unethically. People usually base their behavior on whatever social information they believe to be true, asking themselves, “what do I believe most other people would do or think in this situation?” I designed an experiment to run on Amazon Mechanical Turk that tests the relative effects of uncertain normative and empirical social information on the rates of dishonest reporting.
I found that normatively and empirically framed uncertain beliefs do not have significantly different effects on the decision to lie for one’s self or for a charity. However, the anti-social (lying) belief predicts higher rates of dishonest behavior. Through this research, I learned a great deal about current behavioral ethics research, how to professionally produce data, and how to think critically and dynamically about my results and their place within the wider context of behavioral ethics research.