This editorial outlines the Review of Behavioral Economics (RBE) policy on the interpretation and reporting of statistical significance in empirical research. Building on the 2016 American Statistical Association statement and the longstanding critiques of excessive focus on statistical significance and large sample size, it argues that significance testing has too often become a ritual substitute for scientific reasoning. RBE requires that all empirical submissions discuss the size, direction, and substantive importance of estimated effects — not merely their significance levels. The journal recognises that behavioural and experimental studies often employ contextualised designs, small samples, and within-subject structures that yield rich observations from few participants. Such designs can generate scientifically meaningful insights even when traditional thresholds are not met. By emphasizing effect magnitude, transparency, and variance discussion over mechanical tests, RBE aims to strengthen both the scientific robustness and the real-world relevance of research in behavioural and experimental economics — and to encourage similar reform in the review policies of other journals.
Morris Altman (Tue,) studied this question.