Trust in surveys and games – a matter of money and location?

This paper explores methods to study trust. Answers to survey questions and choices in a trust game are obtained from subjects approached by mail executing their tasks at home as well as classroom subjects. No differences between the results obtained by these methods were observed. Furthermore, one group played the trust game with purely hypothetical payments, and another received random lottery payments. This changed trust behavior dramatically, whereas trustworthiness was less affected. Subjects without any financial incentives exhibited less trust and their trust choices were significantly correlated with survey trust answers. There was no such correlation for the corresponding choices with real payments.