I am performing a MANOVA with a large multivariate dataset representing total body shape (dependent variable). My two independent variables are gender (sex) and cluster (from a previously computed cluster analysis).
body.shape=cbind(unix,uniy,pw1x,pw1y,pw2x,pw2y)
bshape.manova=manova(body.shape~sex*cluster,data=data)
summary(bshape.manova,test="Wilks")
The output works however, no matter what I do I cannot get the interaction term between the two independent variables (sex*cluster) which is really important. I have run 5 MANOVA's on other populations and it all worked great. This population specifically will not given the interaction between the independent variables and their effect on the dependent variable.