P-hacking

The phenomenon of p-hacking occurs when researchers engage in questionable practices that enable them to report findings as being statistically significant. This chapter offers four models of p-hacking behavior—unconditional, strategic, greedy, and restrained—and explores the implications of each model. The author then discusses the implications of recent reforms (routine replication, preregistration, and blinded data analysis) with respect to these models. It seems clear that the set of behaviors that constitute p-hacking are quite consequential for the validity of our research and the perceived legitimacy of our authority as empirical researchers. The models of p-hacking presented here are speculative, but hopefully help shine light on some of the shadier aspects of empirical social science.