Hi all, first of all thank you for your help in advance.

I have the following problem:

I would like to perform a cluster analysis on my sample with a set of variables categorized in several domains. This is just fine but my problem is that two domains (one consisting of four and one of five variables) measure the same thing (just in a different way). As my aim is to profile the people in the sample with the cluster analysis I should only be considering one of the two domains measuring the same thing. Before excluding one preferring another is there a formal way to demonstrate that my population behave similarly to the questions of both domains? If I can prove that they answer similarly there would be no problem to pick one or the other!