Incentive and Aggregation Mechanisms for Crowdsourcing

Abstract

I give an overview over different projects from my PhD research. In my job market paper, I investigate empirically and theoretically how to make use of "meta-beliefs” to improve the information aggregation capacity of crowds. I propose a new aggregation scheme, “self-aggregation” (SELF) and show that in a model in which individuals update their beliefs in a Bayesian fashion, SELF is predicted to outperform alternatives. In an experimental test of the model, respondents solve a binary decision problem in a stylized urn experiment, in which responses and aggregation results can be directly compared to the Bayesian prescription. I find that the use of meta-beliefs improves the accuracy of the crowds’ decisions, even though individual answers differ systematically from the Bayesian prescription.

I embed this project in the general aim of my dissertation to facilitate the use of external crowds to improve decision-making. In particular, I discuss an incentive scheme to encourage truthful participation by the crowd members.