This paper is from Session 3H: Distributions to spatial statistics: New approaches to teaching statistic
which comes under Topic 3: Statistics education at the post-secondary level
Paper 3H1 (Monday 9th, 16:00-17:30)
Developing students’ causal understanding of sampling variability: A design research study
Presenter
- Ethan Brown (University of Minnesota, United States)
Co-author
- Robert delMas (University of Minnesota, United States)
Abstract
Students struggle with understanding sampling variability. Recent transfer and conceptual change literature suggests a new pedagogical approach: supporting a causal explanation of why sampling variability decreases as sample size increases. We designed technological tools, new representations, and a hypothetical learning trajectory to help students understand causal mechanisms of sampling variability. Two causal explanations were targeted: “swamping”, extreme deviations from the mean are less influential in larger samples, and “heaping”, values near the population mean become more probable as sample size increases. Preliminary pilot testing of five students who recently completed an introductory statistics course demonstrated the potential of the activities for deepening students’ thinking, as well as the challenges of managing and articulating the abstraction of swamping and heaping.