I want to perform a Poisson regression to explain Abundance (Counts of individuals) through a number of continuous and categorical explanatory variables. Some of the categorical variables have more than two levels so I will be performing some dummy coding as well.

The question I have is can generalised linear models can cope with multiple types of variables?

  • $\begingroup$ Yes, I think any linear model can cope with multiple types of variables. $\endgroup$
    – Deep North
    Jul 31, 2015 at 7:28
  • $\begingroup$ Yes, just as with regression. $\endgroup$
    – Glen_b
    Jul 31, 2015 at 10:31

1 Answer 1


Yes, they can. The linear component of a GLM can contain whatever a standard linear model can, the difference of course being the distribution of the response and the need for a link function. So go ahead and fit your model but be mindful of the interpretation.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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