An Investigation of State Rankings Based on SAT I Scores

This paper is concerned with the statistical issues of state rankings based on the state average Scholastic Assessment Test I (SAT I) scores and other relevant demographic and policy-related variables that are to be controlled. The purposes of the study were threefold. One purpose was to investigate the optimal transformation of a particular predictor variable, the proportion of high school graduates taking the SAT I, in a regression equation. The second purpose was to apply the all-possible regressions procedure to select the best subset regression equation where the average SAT I scores are used as the predicted variable. The third purpose was to examine a way to combine both the SAT I and ACT Assessment scores for the state education performance indicator that can be used in the state ranking or other purpose. Two appendixes contain detailed descriptions of statistical procedures used in the analyses. (Contains 14 tables, 21 figures, and 53 references.) (Author/SLD) Reproductions supplied by EDRS are the best that can be made from the original document