State Mean SAT Score as a Function of Participation Rate and Other Educational and Demographic Variables

Abstract

Abstract The present study examined the impact of participation rate and nine other educational and demographic variables on state mean scores for the Scholastic Aptitude Test (SAT). Participation rate, defined as the percentage of a state's high school seniors taking the SAT represented by the z-score mean of a truncated normal distribution, was the most important single predictor of state mean SAT score, accounting for 77% of the total variance in mean scores. The inclusion of significant educational and demographic variables raised the proportion of variance explained to 94%. The fallacy of comparing the mean standardized test scores of different locations or time periods without taking participation rate into account is discussed.