Submitted:
30 July 2018
Posted:
05 August 2018
You are already at the latest version
Abstract
It is increasingly common for experiments in biology and medicine to involve large numbers of hypothesis tests. A natural graphical method for visualizing these tests is to construct a histogram from the p-values of these tests. In this article, we examine the shapes, both normal and abnormal, that these histograms can take on, as well as present simple inferential procedures that help to interpret the shapes in terms of diagnosing potential problems with the experiment. We examine potential causes of these problems in detail, and discuss potential remedies. Throughout, examples of abnormal-looking p-value histograms are provided and based on case studies involving real biological experiments.
Keywords:
p-value
; histogram
; inference
; diagnostic
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
