MANCOVA and MANOVA Discussion > MANOVA or MANCOVA: help!!
Hi Katrina,
Thanks for your question, and sorry for my delay in response. From what I've read in your post, I do think that a MANOVA is a reasonable choice. However, I don't think you want all six variables as DVs. Instead, I think you want the 3 personality variables as covariates in the model (predictors). Does that make sense? If it seems like I'm misunderstanding what you're trying to do, let me know.
Thanks, Jeremy
July 28, 2012 |
Jeremy Taylor
Jeremy Taylor
Dear Jeremy,
After conducting MANOVA analysis, followed up by one-way ANOVAs to investigate significant differences, I'm starting to realize that this did not in fact analyze what I wanted to analyse.
Here's what I have:
I'm investigating consumers from 3 different countries, thus 1 IV (3 categories), on their rating of 3 related DVs (measured on 5-point Likert scale, in this case considered continuous), and I would like to see whether their nationality (the IV) matters most or that there are differences depending on the scores of these people on 3 Personality variables (also measured on 5-point Likert scale and considered continues for analysis purposes). Should I consider these 3 variables as Co-variates? And should I thus run a MANCOVA?
What I did was run a MANOVA, where nationality was the IV, the scores on their rating of the 3 DVs, and the 3 personal measurements were together considered as 6 DVs. However, when I do this, I cannot say at all what the effect of one of the DV is on the other, right?
This is all very confusing.. especially since my SPSS Survival Manual does not cover MANCOVA.
Help!
A desperate master student,
Karina