A note on tests for relevant differences with extremely large sample sizes. Issue 1 (11th November 2018)

• Record Type: Journal Article
• Authors: Callegaro, Andrea; Ndour, Cheikh; Aris, Emmanuel; Legrand, Catherine
• Is Part Of: Biometrical journal. Volume 61:Issue 1(2019:Jan.)
• Page Start: 162
• Page End: 165
• Publication Date: 2018-11-11
• Subjects: two‐tailed hypothesis tests -- large sample size -- testing for relevant differences; Biometry -- Periodicals; Medical statistics -- Periodicals; 570.15195
• DOI: 10.1002/bimj.201800195
• Languages: English
• ISSNs: 0323-3847

Abstract

Abstract: A well‐known problem in classical two‐tailed hypothesis testing is that P ‐values go to zero when the sample size goes to infinity, irrespectively of the effect size. This pitfall can make the testing of data consisting of large sample sizes potentially unreliable. In this note, we propose to test for relevant differences to overcome this issue. We illustrate the proposed test on a real data set of about 40 million privately insured patients.