1
$\begingroup$

I am using three different ANOVAs to determine significance across two treatment levels. Because we only have two species I chose to run the same set of analyses for each of the species instead of doing a MANOVA...therefore I have 6 ANOVAs.

I am not interested in conducting post-hoc analyses since we only have two levels, but do I need to correct for the fact that I ran 6 different ANOVAs (one for each of 3 dependent variables across two independent species)?

$\endgroup$

1 Answer 1

3
$\begingroup$

You need to account for false positives, Bonferroni is one option. The false discovery rate is also an option. Only way around this would be to narrow down the number of tests you are doing in advance. Are you really interested in all the possible comparisions?

Given your description, I also suspect your analysis strategy might not be optimal. Having only two levels of a factor is not a good reason to do one ANOVA per level instead of using a two-way ANOVA with an interaction term.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.