1
$\begingroup$

My aim is to see if there is a pattern between demographic features of my customers and the features of the product they most value. The six features of the product make the six dependent variables. Since I am going to ask them to rank the features in order of their preference, my dependent variables are going to be ordinal. I have multiple independent variables (demographic features). One is interval variable, others are nominal (categorical) variables. What statistical analysis should I use?

$\endgroup$
1
$\begingroup$

A good starting place would be ordinal logistic regression. You would then have to check the assumptions of the model. There are a number of alternatives if the assumptions aren't met.

| cite | improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ I'm not sure that is correct. He says that "Since I am going to ask them to rank the features in order of their preference, my dependent variables are going to be ordinal." That does not im my understanding describe an ordinal variable, what he describes are that the response variable records some ordering of features. That is something different! as the different responders defines different orderings. $\endgroup$ – kjetil b halvorsen Jan 29 '15 at 14:16
  • $\begingroup$ Thank you, Peter. I was thinking of ordinal logistic regression too. However, I don't understand what you mean by "check the assumptions of the model". Can you please elaborate? Kjetil, I believe my dependent variable is ordinal, because my customers are going to rank order the features. For example, one customer can say content - 1, flexibility - 2 while another can say flexibility - 1, content - 2. What I want to see if this ordering of features has anything to do with the customer's demographic features. I hope I clarified it now. $\endgroup$ – Saranya Jan 30 '15 at 7:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.