Hello,
As part of my statistical analysis for my thesis, I have done a number of one-way MANOVAs to look at whether there are between-group differences on a number of different variables.
Included in this are post-hoc Tukey HSD tests with significance set at p < .05. However, due to multiple comparisons taking place, do I need to correct my significance level to account for this - or does the Tukey test do this automatically? If so, how many times do I need to divide .05 by?
In general, Tukey's HSD corrects for multiple comparisons automatically (see, for instance in R: "giving the p-value after adjustment for the multiple comparisons" (https://www.rdocumentation.org/packages/stats/versions/3.6.2/topics/TukeyHSD)), but I would check the documentation for your preferred software to confirm
Hello Matthew Haywood. MANOVA = multivariate ANOVA, and multivariate means there are multiple dependent variables. I wonder, therefore, if you mean that you estimated several ANOVA models, one for each DV. It will help to clarify things if you:
1) Tell us what software you used.
2) Show us the command syntax.
3) Tell us how many groups there are.
Thanks.
- Badges
- Science topic