We take an innovative approach to estimating student mathematics learning in the sixth grade of three African countries. The study reinforces the notion that beyond the quality of the teaching process in classrooms, national contextual factors are important in understanding the contribution that schooling makes to student performance. Our approach enhances more typical cross-sectional production function estimates in three ways: (1) to respond to critiques that production function estimates usually do not include classroom processes, we measure both teacher characteristics and teaching process variables and include them in the model; (2) to more clearly identify student learning with schooling processes, we estimate the gain in learning associated with a student's exposure to teaching characteristics and processes during the sixth-grade academic year in each country; and (3) to begin to address the issue of possible "national institutional factors" influencing student achievement, we use a comparative approach to approximate and initiate discussion of "country fixed effects."