The behavioral validity of the strategy method in public good experiments

We compare the strategy method and the direct response method in public good experiments in a within-subject design. This comparison is interesting because the strategy method is frequently used to investigate preference heterogeneity. We find that people identified by the strategy method as conditional cooperators also behave as conditional cooperators under the direct response method. Free-rider types contribute systematically less than all others. Overall, our results support the behavioral validity of the strategy method in public good experiments.

► We investigate the consistency of behavior between a form of strategy method and the direct response method.
► We find rather consistent behavior, in particular in the late periods of the repeated game and in the one shot game.
► Inconsistent behavior occurs mainly in the first part of the experiment and can be attributed to strategic behavior.