I guess I just use the dummy variables for the bivariate tests. It just seems less than ideal because in a sense one dummy variable has less information than the original variable.
For example, if I have geographic region and poverty level, I could do pairwise correlations. region 1 with pov...
I'm not sure if this belongs in the Basic area or here.
I have a dataset with variables that I think it makes sense to make them into dummy variables (some are categories, some are ordinal). I've been asked to do bivariate tests and regression. I don't understand how I can do bivariate...