A Monte Carlo study was performed to test under what conditions and to what degree various alternatives would be superior to multiple linear regression (MLR). The criteria used was the mean squared error of prediction of individual scores and the mean squared error of estimation of the regression weights. Populations were simulated containing either a two or four factor latent factor structure and high, medium, low, or wide range of communalities. Sampling was stimulated with 16, 64 or 256 'subjects'. Forty types of weights were examined. In all conditions an alternative weighting method was superior to MLR. The structural methods, principal components or factor analytic regression, were generally the best for most conditions. Validity weights were generally superior to both MLR and unit weights. The shrunken weights consistently dominated MLR. As much as a 64- fold reduction in errors was found. (Author/RL)