Results in this study suggest that it is possible to use performance on a moderator or predictability scale to select a subgroup for whom correspondence between two predictors and the criterion is much greater than that of a subgroup whose scores on the moderator indicate a lesser degree of correspondence. (Author/ES)

Four commonly employed procedures were repeatedly applied to computer-simulated samples to provide comparative data pertaining to two questions: (a) which procedure can be expected to produce and equation that yields the most accurate predictions for the population, and (b) which procedure is most likely to identify the optimal set of independent…