I run a 2 way MANOVA to examine the interaction between two IVs and 6 DVs. My data met most of the assumptions, except for homogeneity of the covariance (Box's m test). It showed warning that "Box's test of equality of covariance Matrices is not computed because there are fewer than two nonsingular cell covariance matrices".
If I remove one of my DV, the box's m test appears with a p value of more than 0.05.
I am confused if my MANOVA analysis is valid if the homogeneity assumption is not met. I don't want to remove any variable as 5 of the DV are sub-construct or dimensions of main DV.
P. S. The sample sizes are not equal and the N is= 880
http://icit.zuj.edu.jo/icit2023/Index.html
Adnan Mohammad Awad Dafang Zhao Eman I. El-Awady
Suggest before you do anything else look at the attached paper and see why this is probably not what you want to do.
David Booth
Hello Insha Rasool
1- This happens to me all the time. Always. you know why?
2- I enter one of the dependent variables, which is the mean (or sum) of the other dependent variables, in the MANOVA model along with the others..
You don't do that, my dear friend.
- Badges
- Science method