post test question

JohnM

TS Contributor
#2
If your hypotheses, stated before running the experiment/study, call for specific comparisons, then it helps you "cut to the chase" and also in many cases will enhance the power of the statistical tests.

On the other hand, if you have only a general hypothesis, then an overall or "omnibus" test should be done, and if significant, followed up with conservative (i.e., Tukey or Bonferroni) post-hoc tests.
 
#3
JohnM said:
If your hypotheses, stated before running the experiment/study, call for specific comparisons, then it helps you "cut to the chase" and also in many cases will enhance the power of the statistical tests.

On the other hand, if you have only a general hypothesis, then an overall or "omnibus" test should be done, and if significant, followed up with conservative (i.e., Tukey or Bonferroni) post-hoc tests.

Thanks for your valued insight!