Dear all, I have a problem for multinomial logistic regression: the response variable is multinomial (score 1-5) and the two predictors are categorical; all that comes from panelists (it's a kind of preference study), which I treat as a block and include in the model (is it correct?). I would like to see the results in the ANOVA-like framework. Fitting the multinom() function from the nnet package works pretty well and using the Anova() from the package car I got the hypothesis testing for the predictors as well as their interaction. However, I would like to go further and (because I am do look at means when interpreting the results) I would like to make some kind of multiple comparisons, if it is at all possible.
Any suggestions if I am doing anything wrong and if not, what could I do to get these multiple comparisons? Thank you, Stats Wolf ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.