Icy (Yunyi) Zhang (University of Wisconsin-Madison), Ji Yun Son (California State University, Los Angeles)
Abstract
Students often perceive introductory statistics courses as covering a bunch of fragmented topics—descriptive statistics, t-tests, ANOVA, etc. Understanding each individual concept is already difficult, not to mention making connections between each concept. This workshop presents an alternative: reorganizing introductory statistics as statistical modeling, motivated by the Practicing Connections Hypothesis, a learning science framework about teaching in a way that promotes flexible, transferable learning. We also introduce a more hands-on pedagogy that will help students more easily understand abstract concepts by connecting them to physical experience in the real world and practicing applying them in Jupyter Notebooks.
Participants will explore how modeling can serve as a unifying core concept, with topics like data visualization, summary statistics, t-tests, confidence intervals, and ANOVA all relate to the General Linear Model (GLM). We’ll share research-backed pedagogical practices that emphasize on how to foster meaningful and connected mental representation. These practices have supported a diverse range of students from over 50 universities as they connect statistical models to real-world contexts, coding in R, and formal representations like GLM notation.
This hands-on, 6-hour workshop is designed for instructors of introductory statistics interested in modernizing both content and pedagogy. No prior experience with R programming or teaching from a modelling framework is required. Although some teaching experience in statistics is preferred, we also welcome participants who have an interest in teaching statistics but haven’t taught before. Participants should bring laptops with Google Chrome or Safari.