Eliciting dishonesty in online experiments: The observed vs. mind cheating game

In this paper, I compare two ways of eliciting cheating behavior in online experiments. I present data from two online experiments, one in which participants’ random drawing took place directly on their screens (Observed-Cheating Game) and another in which participants chose a color in their minds and then randomly drew a color from ten boxes with question marks presented on their screens (Mind-Cheating Game). The paper shows that observed online games are more likely to have non-significant treatment differences because the effect of observability is particularly strong. I show that using mind games to generate random draws in online settings solves the most prominent problems for eliciting lying by making the lies unidentifiable at the individual level. The MIND-CHEATING GAME used in the experiment is inspired by previous mind games and is specifically adapted to be easily implemented in online or field experiments.