Course -taking and achievement in mathematics and science: A comparison of three nations using TIMSS
This research focuses on the differential achievement of students taking advanced mathematics or physics or both versus students taking none of these courses. The target sample, extracted from the Third International Mathematics and Science Study, included secondary school students in their last year of school in the United States who were compared to those in Australia and Israel. Propensity scoring was used to control for background variables while estimating the achievement differences between students taking advanced courses and those who did not take such courses. Findings were similar in each of the three countries. Students taking advanced courses achieved higher on a basic mathematics and science test. The adjusted estimate was found to be smaller after controlling for background variables than the unadjusted estimate. The effect size could be deemed as “medium” according to Cohen's standards. In addition, subclassification through propensity score showed that as one moves from the lower quintiles to the higher quintiles, the achievement gap is larger. Students in the higher quintiles were those from a more advantaged background and could be stereotyped as the model student. School variables such as shortage of instructional materials or percentage of teachers with no university level certification in mathematics or science could not explain this phenomenon. Sensitivity analysis was also conducted, which showed that the achievement gap estimate was susceptible to hidden bias across the three countries. An omitted covariate with extreme values could possibly explain away the differential achievement after it was accounted for.
Leow, Christine S, "Course -taking and achievement in mathematics and science: A comparison of three nations using TIMSS" (2002). Dissertations available from ProQuest. AAI3073026.