Abstract: The purpose of this paper is to demonstrate the use of the statistics of relative risk and odds ratios in mathematics education. These statistics are widely used in other fields (especially medical research) and offer a useful but currently under-utilised alternative for education. The demonstration uses data from a longitudinal study of students' understanding of decimal notation. We investigate the statistical significance of results related to the persistence of misconceptions and the hierarchy between misconceptions. Relative risk and odds ratio techniques provide confidence intervals, which give a measure of effect size missing from simple hypothesis testing, and enable differences between phenomena to be assessed and reported with impact.
Steinle, V., & Stacey, K. (2005). Analysing longitudinal data on students' decimal understanding using relative risk and odds ratio. In H.L. Chick & J.L. Vincent (Eds.), Proceedings of the 29th Conference of the International Group for the Psychology of Mathematics Education, (Vol. 4, pp. 217–224). Melbourne: PME.