|
Abstract:
|
We examine three years of cross-institutional data (pre/post tests) from a wide variety of introductory statistics courses, exploring student gains in learning, including courses that revolve around simulation-based inference. This includes comparison of student attitudes towards statistics and student retention 4-months and 16-months post course. Hierarchical models explore student-level characteristics (e.g., first statistics course, first generation, prior mathematical performance) and instructor-level characteristics (e.g., type of institution including high school and community college, years teaching, familiarity with GAISE guidelines). We find that simulation-based inference courses consistently show larger gains than non-SBI curricula, regardless of institution, year, student pre-test score and a host of other institutional and student characteristics. Further research is needed to better understand what aspects of simulation-based inference curricula are directly leading to the observed patterns in student learning.
|
