What’s missing in longitudinal studies conducted in the U.S. with implications for mathematics education?

ABSTRACT

‘What works’ policies are the latest incarnation of best practices in educational research. Instituted by various organisations internationally, they define what kind of research counts as ‘evidence’ for reform-oriented decision-making. While some countries rely on systematic analyses and meta-analyses, the U.S. favours primary quantitative longitudinal research designs as evidence. Critics claim that institutionalising methods will distort educational research. This commentary discusses these two concerns about the research designs favoured in the American ‘what works’ milieu: missing research and missing data. To explore whether those concerns are warranted, we focus on mathematics education, a national priority in the U.S. After establishing enduring questions in mathematics education, we provide primers on quantitative longitudinal research designs and the analysis of missing data due to attrition. Then, we address these questions about the state of research: As a research community, are we missing research with implications for mathematics education? Are we missing data with implications for the inferences that can be made? This review of research questions and analyses in recent studies suggests missing data poses a greater threat than missing research. Finally, we make recommendations for minimising the risk of missing evidence that will be of interest to international educational researchers in any discipline.

KEYWORDS:

• Longitudinal design
• missing data
• mathematics education
• research policy