#ReadyResources: Publications for teaching p-values
Are you talking p-values with your health sciences colleagues and learners? One punching bag in the "reproducibility crisis" is the p-value and the use of the 0.05 threshold.
Both statisticians and non-statisticians are weighing in. ASA put out a statement about p-values, with multiple additional perspectives, in 2016. Most recently, this article came up in my Slack feed with data-science and clinical collaborators, The Proposal to Lower P Value Thresholds to .005, by John P. A. Ioannidis, MD, DSc (JAMA 2018).
If you find yourself being called on to talk p-values, here are some #ReadyResources from the Points of Significance column in Nature Methods.
Points of significance: Significance, P values and t-tests (2013) "To make informed judgments about the observations in a biological context, we must understand what the P value is telling us and how to interpret it. This month we will develop the concept of statistical significance and tests by introducing the one-sample t-test."
Points of significance: P values and the search for significance (2017) "Our goal this month will be to identify some circumstances that can give rise to such questionable practices—broadly termed 'P value hacking' and 'data dredging'. In addition, statistically significant results may not translate into biologically meaningful conclusions—with large sample sizes or small variability, even tiny effects can be statistically significant."
Points of Significance: Interpreting P values (2017) "A P value measures a sample's compatibility with a hypothesis, not the truth of the hypothesis."
Each of these three articles was written by Naomi Altman and Martin Krzywinski. Each 2-page tutorial is aimed at biologists, but no specific discipline is assumed. The statistical concepts behind p-values and significance are presented with text and figures. Selecting one of these tutorials to pair with an opinion piece or a clinical research publication, could make a nice journal club format for learners.
Do you have a go-to publication or method for talking about p-values? Is there an article or opinion piece that has resonated with your health science learners? Please share it in the comments below, or send it to the blog editor for possible inclusion in future posts.