Offers researchers a guide for selecting the best statistical model to use, as well as discussing such topics as contextual analysis with absolute/relative effects, and the choice between regression coefficients as fixed parameters or as random variables.