I am confused about Holm-Bonferroni.

Suppose we want to compare age of female and male students using t tset. We also want to compare length of arm and elbow between male and female students.

I think we need to use Holm-Bonferroni correction in elbow and arm cases. But there is no need to Holm-Bonferroni correction for the p value of age.

Is this right? what means multiple comparison which is said about Holm-Bonferroni? please mention one example. Is Holm-Bonferroni correction needed in some logistic regression models too?

  • 1
    $\begingroup$ I have never hear of a Holm-Bonferroni test. It seems like you mean the Holm-Bonferroni multiple testing correction procedure, but this is not a hypothesis test. Regression models are also unrelated to tests, and to multiple testing correction. If you can narrow down your question to a single subject, perhaps it can be answered here, but as it stands it is far too broad to answer succinctly. $\endgroup$ – Frans Rodenburg Apr 22 '19 at 3:39
  • $\begingroup$ There are many questions on this site with the tag [multiple-comparisons] which may help you. Try stats.stackexchange.com/questions/1458/… for instance. $\endgroup$ – mdewey Apr 22 '19 at 10:09