Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 1 down vote favorite

When I run a 2x4x5x6 (Sex*AgeGroup*HighestDegree*Ethnicity) MANOVA (with 20 dependent variables) in SPSS, it presents tests of between subjects effects for each of the IVs and their interactions after the main MANOVA stats as a kind of post hoc. I thought these tests were 2*4*5*6 factorial ANOVAs for each of the DVs examined in the MANOVA. But when I run those factorial ANOVAs separately, I get completely different results! (The F values in the MANOVA post hoc are generally higher than the F values resulting from the separate factorial ANOVAs.) Why is that? I know I shouldn't trust the main MANOVA stats (Pillai's trace, etc.) because the cell sample sizes are way too small, but I'm at a loss as to why the post hocs from the MANOVA are different from factorial ANOVAs for each of the DVs.

share|improve this question
I'm not sure if I fully understand your situation, & I don't have SPSS, but in general, MANOVA can be thought of as an ANOVA performed on a specific linear combination of your DVs which is basically never equivalent to exactly one of your DVs w/ the others dropped. – gung Jul 25 '12 at 16:38
Thanks! I understand that - I didn't expect the MANOVA to be equivalent to (any of) the ANOVAs for each of the DVs. What I expected was that each of the post hoc tests presented alongside the MANOVA, that analyzed the effects of each of the IVs (and IV interactions) on each of the DVs separately, would be equivalent to one of the ANOVAs for each of the DVs separately. But that expectation turned out to be wrong, which surprised me. – Ian Jul 25 '12 at 16:55

Know someone who can answer? Share a link to this question via email, Google+, Twitter, or Facebook.

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.