I have two datasets of demgraphic data which I wish to compare. Ideally I would like to run a do file as I actually have study sample and main population data for 4 different areas. However, as each variable has different numbers of categories, eg sex(2) or occupational classification(9), my reading seems to suggest different tests.
What would you use to best assess the similarity between each study sample and its host population whilst adequarely covering the dichotomous and the categorical data?