A Meta-Reanalysis of Dream-ESP Studies: Comment on Storm et al. (2017)

Abstract

Storm, Sherwood, Roe, Tressoldi, Rock and Risio (2017) performed a meta-analysis of laboratory studies of ESP in dreams, finding a moderate overall effect (r = .20).  Certain statistical decisions, such as primarily reporting unweighted effects, may have inflated these observations.  In the current article, I perform an inverse-variance weighted meta-analysis using the database provided by Storm et al., finding a much smaller overall effect (r = .07).  I also find evidence that a significant relationship exists between effect size and sample size, suggesting that the prior results may have been primarily driven by large effects found in small-n studies.  I suggest that future researchers of ESP in dreams should produce more large-n studies, which could alleviate many notable criticisms of the current psi literature.