The issues and some proposed solutions regarding Follow Through (FT) site variability are examined with a review of developments in FT evaluation. The role of adjusted site means with differences within sponsors and between sponsors and background characteristics is discussed to determine whether adjusted means are the preferred measures of model effectiveness. In a Big City Group, attrition bias in data for non-FT and FT site analysis is considered. Improvements in measurement are shown in the sampling of content and behavior, including the use of computer systems with broad content samples. These procedures can eliminate reliance on multiple choice questions and the use of classroom process data with student reports on opportunity to learn (OTL) data expanding the dimensions of variability. A comtemporary model which crosses class type with school sites illustrates the multilevel regression analysis. Student scores are the dependent variable; and class type, sex, OTL class mean and individual math ability are the independent variables. The significant role of OTL to the stepwise fitting of the model is shown. (CM)