Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have fitted a Poisson model to my data in R which includes two factors as independent variables. Each factor has 5 levels. I have used the contrasts = list(FactorA='contr.sum', FactorB='contr.sum')command to change the constraints of my model. So, the parameter estimates for each factor sum to 1.

My problem is that I find it very difficult to understand what exactly the p-values of the estimates mean. For instance, if the p-value for one level of FactorA is significant, does this mean that it's generally significant for explaining the response variable or only in comparison to the other levels of the same factor?

Are the constraints that I used involved in this interpretation that I have to make for the significance of each level?

Also, I do not get any output for the last level of each factor. Is there any way to interpret those two levels?

share|improve this question
Sorry, parameter estimates sum to 0 (not 1). – Achilles Aug 23 '12 at 9:27
When presented with a categorical variable, the GLM does pairwise comparisons against some base reference level to all the other levels. So if you have 5 levels in your categorical variable, the glm would compare lvl 1 and 2, lvl 1 and 3.. and so on. As to what levels are significant, there's a post [here][1] that may help. [1]:… – user4673 Aug 23 '12 at 16:07

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.