Estimating the replicability of statistically significant moderation effects in personality research using z-curve analysis

Abstract

We use z-curve analysis to estimate the replicability of 187 moderation effects in which the predictor, outcome, and moderator variables were all measured continuously and that were published in four leading personality journals—Journal of Personality, Journal of Research in Personality, European Journal of Personality, and Personality and Individual Differences. We found an estimated replication rate of 43.80%; that studies published after 2015 performed slightly better than studies published before 2015; and that studies published in the Journal of Research in Personality and the European Journal of Personality performed worse than studies published in the other two journals. We discuss the implications of our findings for studying moderation effects in personality and offer recommendations for improving the replicability of published studies.