Multiplicity
Definition: Potential inflation of Type I error rates (incorrectly rejecting the null hypothesis) because of multiple statistical testing, for example, multiple outcomes, multiple follow-up time points, or multiple subgroup analyses. To overcome issues with multiplicity, researchers will often apply controlling procedures (e.g., Bonferroni, Holm-Bonferroni; Tukey) that correct the alpha value to control for inflated Type I errors. However, by controlling for Type I errors, one can increase the possibility of Type II errors (i.e., incorrectly accepting the null hypothesis).
Related terms: Alpha, False Discovery Rate, Multiple comparisons problem, Multiple testing, Null Hypothesis Significance Testing (NHST)
References:
- Sato, T. (1996). Type I and Type II error in multiple comparisons. The Journal of Psychology, 130(3), 293–302. https://doi.org/10.1080/00223980.1996.9915010
- Schulz, K. F., & Grimes, D. A. (2005). Multiplicity in randomised trials I: endpoints and treatments. The Lancet, 365(9470), 1591–1595. https://doi.org/10.1016/S0140-6736(05)66461-6
Originally drafted by: Aidan Cashin
Reviewed by: Jamie P. Cockcroft, Mahmoud Elsherif, Meng Liu, Charlotte R. Pennington