The purpose of this paper is to develop a more general method for sample recruitment in experiments that is purposive (not random) and that results in a sample that is compositionally similar to the generalization population. This work builds on Tipton et al. (2011) by offering solutions to a larger class of problems than the non-overlapping eligible and population case described in this paper. There are three parts to this approach: (1) Populations and samples defined; (2) Covariate selection assumptions; and (3) Sample selection method.