The Empirical March: Making Science Better at Self-Correction

Abstract

Psychology has been criticized recently for a range of research quality issues. The current article organizes these problems around the actions of the individual researcher and the existing norms of the field. Proposed solutions align the incentives of all those involved in the research process. I recommend moving away from a focus on statistical significance to one of statistical power, renewing an emphasis on prediction and the pre-registration of hypotheses, changing the timing and method of peer-review, and increasing the rate at which replications are conducted and published. These strategies seek to unify incentives toward increased methodological and statistical rigor to more effectively and efficiently reduce bias and error.