Here at Cytel we have enjoyed following the debates on the p-value controversy currently taking place on the ASA website, and indeed debating the issue ourselves. While we plan to post our views on the topic in the weeks to come, we would like to invite your thoughts on the issue as well.
For those who may have missed the debate, the issue arose after the Journal of Basic and Applied Social Pyschology decided to restrict the publication of all papers where hypothesis testing is used for analysis. This includes p-values, Confidence Intervals, etc.
The basic contention of the journal is that p-values are misused, misinterpreted and misleading. However, there is also a general move, on the part of the journal, to replace the publication of inferential statistics with descriptive statistics. The editors do claim, however, that Bayesian findings will be considered on a case by case basis [1].
Clearly, the avoidance of inferential statistics is not going to occur in pharmaceutical development anytime soon. However, might there be reasons to encourage more Bayesian methods? Tell us what you think by taking the survey below.
Create your own user feedback survey
[1] Trafimow, D., Marks, M. (2015). Editorial, Basic and Applied Social Psychology, 37(1), 1-2.
