The test was administered by teachers in their science classes and the test papers were scored by independent researchers.Test scores for the academically selective group and the comparison Groups A and B were compared statistically using a general linear model repeated measure analysis (ANOVA) to determine the interaction between the two independent variables of time (pre- and post-intervention) and participation in the thinking intervention for Comparison Group A, and time and attendance at an academically selective school for Comparison Group B. Cohen’s effect size (d) was used to quantify the differences between the groups as suggested by Allen and Bennett (2008), p. 120 using comparisons of gain scores. Using Cohen (1988) conventions as a guide, d of 0.20 can be considered small, a d of 0.50 is medium, and a d of 0.80 is large (Cohen,1988).Phase 2