My dependent variable is a percent range (intervals of 4):

None 1-5% 6-10% 11-15% 16-20% 21-25% 26-30% 31-35%.....and so on until 100%

Is this an ordinal variable or an interval variable? What would be the best regression analysis to run for this type of variable?

Thank you

  • $\begingroup$ Ordinal, I think. $\endgroup$ – keval Mar 17 '18 at 16:45
  • $\begingroup$ can i ask why you think it's ordinal and not interval? $\endgroup$ – user3424836 Mar 17 '18 at 16:47
  • $\begingroup$ I would call it interval. Regarding your second question, you can't pick a statistical test without defining what you want to infer about the data. $\endgroup$ – Michael Chernick Mar 17 '18 at 17:08
  • $\begingroup$ Thank you for your response. I'd like to know the relationship between several categorical variables and the dependent variable. My apologies, I should have stated "what type of regression analysis" would be best for this type of variable. The previous poster thought ordinal. Can I ask why you think it is interval? $\endgroup$ – user3424836 Mar 17 '18 at 17:13
  • 1
    $\begingroup$ @user3424836 Each interval is a level, thus ordered factors. You could look into something like ordered logit regression / ordinal regression. $\endgroup$ – keval Mar 17 '18 at 18:28

Your Answer

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

Browse other questions tagged or ask your own question.