QUASI-EXPERIMENTAL ESTIMATES OF THE EFFECTS OF A PRESCHOOL INTERVENTION - PSYCHOMETRIC AND ECONOMETRIC COMPARISONS

Estimates of program effects from a compensatory preschool interventio n were investigated using several contemporary techniques of bias redu ction. These included econometric simultaneous modeling, latent-variab le structural modeling, and ordinary least squares regression. These t echniques were applied to longitudinal data from Chicago's Child Paren t Center preschool program, a Head Start-type intervention. Analyses o f 806 Black children followed from kindergarten to Grade 6 indicated t hat the techniques, by and large, produced similar estimates of progra m impact on school achievement test scores. Effect sizes ranged from.4 3 to.67 at school entry to.24 to.30 at Grade 6 (7 years postprogram). These findings were robust across different model specifications and s imilar to those of randomized studies. The only differences among meth ods occurred at Grade 3. Findings suggest that different quasi-experim ental estimation methods that attempt to control for selection bias ca n generate similar estimates of the effects of an educational treatmen t.