Controversy exists regarding proper methods for the selection of variables in confounder control in epidemiological studies. Various approaches have been proposed for selecting a subset of confounders among many possible subsets. This paper describes the use of two practical tools, Stata...