What is it about?

To better explore our research data, a double approach to data analysis may give us the best of two worlds: A significance test allows us to be wary of errors; Bayes factors allows us to explore the weight of alternative explanations.

Featured Image

Why is it important?

A recent proposal by Benjamin et al. (2017) calls for the use of a more stringent level of significance (p < 0.005, instead of p < 0.05) for claiming proof of evidence for novel results Unfortunately, statistical significance and substantive evidence imply different philosophical approaches, and the statistical tools derived under each approach are not necessarily compatible. A large number of authors have criticized Benjamin et al.'s proposal because of the mix-up, which could have been easily avoided by recommending a double-approach to data analysis (e.g., via using a data analysis software such as JASP).


I hope this article serves to minimize the "statistical wars" that have been raging for ages between objective (frequentist) and subjective (Bayesian) approaches to data analysis. My view is that both perspectives have an important role to play in fields such as psychology and the life sciences, and such role needs to be recognized and acknowledged for better science.

Jose Perezgonzalez
Massey University

Read the Original

This page is a summary of: Retract p < 0.005 and propose using JASP, instead, F1000Research, December 2017, Faculty of 1000, Ltd., DOI: 10.12688/f1000research.13389.1.
You can read the full text:

Open access logo


The following have contributed to this page