I would appreciate your advice.

I have a number of outcome variables that follow Normal, Gamma or Poisson distribution. The predictor is categorical with two levels, as follows: Category 1: N=321, 9.3% of the sample Category 2: N=3145, 90.7% of the sample

I understand that unequal cell size is a problem in ANOVA/regression/GLM. How can I run GLM analyses with above-mentioned distributions and unbalanced predictor?

  • 1
    $\begingroup$ @Naatalie: I understand your question to pertain to generalized linear models only. If I misunderstood you, please feel free to change my edit. $\endgroup$ – Michael R. Chernick Jan 10 '17 at 15:56
  • $\begingroup$ Michael, most of the time I use generalized LM since the data is not normally distributed. So the question pertains to the GLM with a variety of distribution, depending on the outcome variable. $\endgroup$ – Natalie Jan 10 '17 at 16:04
  • $\begingroup$ That is exactly how I interpreted it. My changes were very minor and in the title I changed "generalized general modeling" to "generalized linear modeling" because the term "general" seems redundant and the restriction to linear models was missing. $\endgroup$ – Michael R. Chernick Jan 10 '17 at 16:10

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.