In 1981 a meta-analysis of research in science education was completed by a team of researchers under the direction of Ronald Anderson at the University of Colorado. Individual teams synthesized the extant research using meta-analysis procedures. One of the syntheses showed that the post-Sputnik or "new" science programs on the whole produced very positive gains in student achievement. Statisticians have questioned the integrity of meta-analysis results because standard procedures fail to take into account the error associated with the sample estimation of the effect size. It is argued that when a set of different effect sizes are pooled across studies, the means most properly should be computed using weighted factors associated with the precision of the effect size values. This research repeated the meta-analysis of research on the new science curricula using weighted procedures proposed by Hedges. The results provide more precise information on the actual effectiveness of the post-Sputnik science curricula meta-analyzed in the Anderson project and establish clearer guidelines regarding the application of meta-analysis techniques in synthesizing research results. (Author/JN)