English

Choosing alpha post hoc: the danger of multiple standard significance thresholds

Applications 2025-03-11 v2 Methodology

Abstract

A fundamental assumption of classical hypothesis testing is that the significance threshold α\alpha is chosen independently from the data. The validity of confidence intervals likewise relies on choosing α\alpha beforehand. We point out that the independence of α\alpha is guaranteed in practice because, in most fields, there exists one standard α\alpha that everyone uses -- so that α\alpha is automatically independent of everything. However, there have been recent calls to decrease α\alpha from 0.050.05 to 0.0050.005. We note that this may lead to multiple accepted standard thresholds within one scientific field. For example, different journals may require different significance thresholds. As a consequence, some researchers may be tempted to conveniently choose their α\alpha based on their p-value. We use examples to illustrate that this severely invalidates hypothesis tests, and mention some potential solutions.

Keywords

Cite

@article{arxiv.2410.02306,
  title  = {Choosing alpha post hoc: the danger of multiple standard significance thresholds},
  author = {Jesse Hemerik and Nick W Koning},
  journal= {arXiv preprint arXiv:2410.02306},
  year   = {2025}
}

Comments

Accepted for publication in Statistical Science