Of critical importance to education policy is the monitoring of trends in education over time. Indeed, the United Nations Sustainable Development Goals identied Goal 4 as focusing on quality education for all. Developing optimally predictive models allows researchers to assess cross-country progress and forecasts toward that goal. One approach to optimizing prediction is Bayesian model averaging (BMA). Although BMA has been applied to prediction and forecasting in economics and weather research, it has not been applied to the field of education, where international large-scale educational data are readily available. We extend and apply BMA to cross-country gender differences in mathematics achievement from the IEA Trends in International Mathematics and Science Study (TIMSS). At the country level, these data are longitudinal and are uniquely suited for application of cross-country growth regression. Preliminary results indicate some benefit to applying BMA to these data in terms of optimal prediction and forecasting.
Presented in Session 20. Methods for Evaluating Population Programs