The article cites a study which examines the effectiveness of economic education programs in the U.S. In this study, conventional models and statistical techniques were used to estimate student understanding and learning of economics in a pre or post test framework. The most important aspects of the study include the use of a large, randomly selected group of schools; the introduction of some variables that have not been used in published research on economic education; and the use of a more comprehensive set of variables than has been possible in many earlier studies. Particular attention was paid to the recent discussion on the Developmental Economic Education Program (DEEP) and to the relationship between DEEP and other variables used in this study. Although many of the student, teacher, and school system variables in this study had the expected signs and statistically significant coefficients, the DEEP variables were usually negatively and often significantly related to the student pretest scores. In equations using posttest scores as dependent variables, coefficients on the binary DEEP variable were positive in two out of four cases but were significant only once.
