Opportunity to learn and mathematics achievement: The roles of content coverage and instructional time in TIMSS 2023

Source abstract

Opportunity to learn (OTL) is often introduced through curriculum policy, but students meet it in a more practical form: The mathematics topics their teacher covers and the time set aside for those topics. This paper examines how these two classroom indicators are related to Grade 8 mathematics achievement in TIMSS 2023. Public-use student, mathematics-teacher, and student-teacher linkage files were brought together for the 47 education systems that supplied all three files. The final analytic sample contained 112,827 complete student-mathematics-teacher links, representing 111,954 individual students in 46 systems and 8,259 schools. Mathematics achievement was represented by five TIMSS plausible values. Content coverage was calculated as the mean percentage of students taught number, algebra, geometry/measurement, and data/probability topics. Instructional time was the teacher-reported number of mathematics hours per week. The weighted least-squares models adjusted for home educational resources, student sex, and education-system fixed effects. Each system received the same total analytic weight. Estimates were first calculated for each of the five plausible values and were then combined with Rubin's rules; standard errors were clustered by school. Average coverage was 79.67%, while mean weekly mathematics time was 3.87 hours. A one-standard-deviation increase in composite content coverage (16.27 percentage points) was associated with a 4.06-point higher mathematics score (SE = 1.06, p < .001). Instructional time did not show a clear linear association at the mean, although the small negative quadratic term was statistically distinguishable from zero. In the domain model, the clearest positive association was for number coverage. Algebra was weaker, and geometry/measurement and data/probability were not statistically clear. Overall, the results suggest that a reasonably broad coverage of content may tell us more about OTL than simply adding lesson time. The data are cross-sectional and coverage is partly based on teacher reports, so the findings are adjusted associations rather than causal effects.