Using Online Experiments for Behaviourally Informed Consumer Policy

Abstract

In this paper, I discuss the merits of online experiments for behaviourally informed policy making. The online experiments we have in mind combine elements of laboratory and natural field experiments (RCTs), with a focus on external validity. These experiments are increasingly used to investigate consumer behaviour in different settings, including the testing of prospective consumer policies. After establishing that online experiments are well-suited to certain types of consumer policy research, I discuss how to design such experiments. I pay special attention to the key challenges that the policy researcher encounters when designing an experiment: recruiting and rewarding participants, structuring the experimental task and deciding whether to incorporate elements that diverge from the target field context (‘abstractions’) into the design. I also review the academic literature on experimental validity and reflect on its implications for online experiments by policy researchers. Online experiments have good prospects to become an important part of the policy research toolkit and it is likely that their use will grow in years to come. I hope this paper serves as a useful guide to the policy researcher looking to design an online experiment.