Dundee Discussion Papers in Economics 153: Sampling bias in economics experiments: an analysis of the effects of attrition among subjects who sign up for experiments

Experiments in economics tend to use a restricted subject pool consisting of university students. This obviously raises concerns about the possibility of sampling bias. Although work has been undertaken in particular experiments to compare the results obtained from different subject pools, there has been little systematic investigation of the potential problems caused by the subject recruitment methods used. In this paper we examine one particular aspect of this – the fact that students who sign up for experiments do not always attend. Using two ‘classic’ experiments we tested for the existence of this form of sampling bias, and our results indicated its presence in one of the two experiments undertaken. This suggests that the issue of sampling bias is an important subject for further research in experimental economics, and an issue to be carefully considered in the design and implementation of experiments.