Bonferroni Adjustments

Abstract

The unpaired t-test can be used to test the hypothesis that the means of two parallel-group are not different (Chap. 7). When the experimental design involves multiple groups, and, thus, multiple tests, it will increase the chance of finding differences. This is, simply, due to the play of chance, rather than a real effect. Multiple testing without any adjustment for this increased chance is called data dredging, and is the source of multiple type I errors (chances of finding a difference where there is none). The Bonferroni adjusted t-test (and many other methods) are appropriate for adjusting the increased risk of type I errors. This chapter will assess how it works. In the current chapter only continuous outcome data are adjusted for multiple testing. However, binary data can equally be assessed using the Bonferroni equation.