Reexamined previous statistical comparisons of two models for adolescent drug abuse using new statistical estimation methods in causal modeling not requiring assumptions about normally distributed variables. An asymptotically distribution free method shows that the models fit even better than assumed in the initial work. (Author/JAC)

Addresses common pitfalls in homogeneous scale construction in clinical and social psychology. Offers suggestions about item writing, answer scale formats, data analysis procedures, and overall scale development strategy. Emphasizes effective use of factor-analytic methods to select items for scales and to determine its proper location in…