Toggle menu

Preventing statistical errors in scientific journals

Issue: 42(1) February 2016. Essays Pages 8 – 10

Michèle B. Nuijten
Tilburg University, The Netherlands; M.B.Nuijten@tilburguniversity.edu

Abstract

There is evidence of a high prevalence of statistical reporting errors in psychology and other scientific fields. These errors display a systematic preference for statistically significant results, distorting the scientific literature. There are several possible causes for this systematic error prevalence, with publication bias being the most prominent one. Journal editors could play an important role in preventing statistical errors in published literature. Concrete solutions entail encouraging sharing data and preregistration, and using the automated procedure “statcheck” to check manuscripts for errors.

Full text

Download as PDF

Share this article