Comparing Prisoner's Dilemma, Commons Dilemma, and Public Goods Provision Designs in Laboratory Experiments

The ability to cumulate and compare findings from laboratory experiments is impeded by the wide variety of designs employed in these studies and by the different incentive characteristics embedded in those designs. This article attempts to improve potential for comparing findings by proposing variables that specify incentive features of collective action designs. For many designs as seemingly disparate as those in prisoner's dilemma, commons dilemma, and public goods provision games, these specifications should enable different designs to be represented as values along common incentive parameters. With this procedure, the influence of incentives can be assessed and controlled across disparate experimental designs as well as the cumulative influence of other variables (for example, group size and identity, framing, and socioeconomic characteristics of subjects) that provoke collective action behavior. The article also notes the limitations of this procedure and other obstacles to the meaningful cumulation of findings on collective action. The procedure may not, for example, allow easy comparison of findings from experimental designs that have disparate nonlinear returns to cost contributions.