Solved: Re: How do I interpret type 3 tests of mixed effects? (with photo)

Angmar · Posted 03-22-2018 06:19 PM

How do I interpret the following?

I know that because of the significant interaction, I shouldn't interpret lower order terms.

What is the significant interaction telling me? And if it weren't significant, what would the lower order terms tell me? That there's a difference between groups?

New to SAS and kind of new to stats - I appreciate the help!

sld · Posted 03-26-2018 04:34 PM

The interaction indicates only that differences among the 3 DRUGs are not the same at all 7 TIMEHRs. It does not tell you anything about how or which means are different; you have to use post-hoc testing to sort that out. With a 3 x 7 factorial, the post-hoc testing is complex and (in my opinion) should be guided by the context of your study.

I would start by plotting the estimates of the interaction means. But then you'll need to learn a lot more about interpretation of ANOVA both with and without interaction. The topic is too complicated to address adequately in this Community forum. Good luck!

View solution in original post

sld · Posted 03-26-2018 04:34 PM

The interaction indicates only that differences among the 3 DRUGs are not the same at all 7 TIMEHRs. It does not tell you anything about how or which means are different; you have to use post-hoc testing to sort that out. With a 3 x 7 factorial, the post-hoc testing is complex and (in my opinion) should be guided by the context of your study.

I would start by plotting the estimates of the interaction means. But then you'll need to learn a lot more about interpretation of ANOVA both with and without interaction. The topic is too complicated to address adequately in this Community forum. Good luck!