David Kaplan, University of Wisconsin – Madison

2024 Presidential Address

In 2015, the United Nations adopted the Sustainable Development Goals. With regard to education, the UN identified equitable, high-quality education, including the achievement of literacy and numeracy for all youth and adults as a key goal. To assess country-level progress toward these goals, it is important to monitor trends in educational outcomes over time, and this is particularly true given the COVID-19 disruption on education systems around the world.  The purpose of this paper is to demonstrate how optimally predictive growth models can be constructed in order to monitor the pace of progress at which countries are progressing toward the education sustainable development goals. A number of growth curve models can be specified to estimate the pace of progress, however, choosing one model and using it for predictive purposes assumes that the chosen model is the one that generated the data, and this choice runs the risk of “over-confident inferences and decisions that are more risky than one thinks they are” (Hoeting et al., 1999). A classical approach to addressing this type of model uncertainty is Bayesian model averaging (BMA) (e.g. Madigan & Raftery, 1994) .  However, BMA rests on the assumption that the true data generating model is in the set of models that are being averaged.   To mitigate this problem, we adapt and apply Bayesian stacking to form mixtures of predictive distributions from an ensemble of individual models specified to predict country-level growth rates. Bayesian stacking relaxes the true data generating model assumption of BMA and does not even require that a true data generating model exists.  We demonstrate Bayesian stacking on country-level data from the Program on International Student Assessment. Our results show that Bayesian stacking yields better predictive accuracy than any single model as measured by the Kullback-Leibler divergence.  On the basis of the ensemble average calculated from the stacked predictive distribution, we show a forecast plot for one PISA cycle ahead – namely PISA 2025.

About the Speaker

David Kaplan is the Patricia Busk Professor of Quantitative Methods in the Department of Educational Psychology at the University of Wisconsin – Madison. Dr. Kaplan holds affiliate appointments in the University of Wisconsin’s Department of Population Health Sciences and the Center for Demography and Ecology. Dr. Kaplan’s program of research focuses on the development of Bayesian statistical methods for education research. His work on these topics is directed toward applications to large-scale cross-sectional and longitudinal survey designs. Dr. Kaplan is an elected member of the National Academy of Education; a recipient of the Samuel J. Messick Distinguished Scientific Contributions Award from the American Psychological Association (Division 5); a past-President of the Society for Multivariate Experimental Psychology; a fellow of the American Psychological Association (Division 5); a recipient of the Alexander Von Humboldt Research Award;  a fellow of the Leibniz Institute for Educational Trajectories; and was a Jeanne Griffith Fellow at the National Center for Education Statistics. Dr. Kaplan received his Ph.D. in education from UCLA in 1987.