Tori Ellison (University of Illinois Urbana-Champaign)
Abstract
As the data science industry rapidly evolves, graduates are increasingly expected to teach themselves new techniques after their formal coursework has ended. Yet prior studies of informal machine learning learners show that challenges consistently emerge across all three stages of self-directed learning: diagnosing learning needs and goals, choosing effective learning strategies and resources, and evaluating progress. We explore how statistics instructors can better prepare students to teach themselves new data science concepts after graduation. The session will showcase an example “future learning roadmap” for a particular data science topic designed to address common pitfalls students experience when teaching themselves new machine learning concepts. Participants will brainstorm skills, questions, and scaffolds that can help data science students become more adaptive, lifelong learners. The goal is to spark a community-wide conversation about how we can intentionally design courses that prepare students not just to use current tools, but to learn future ones.