Abstract:
|
Post the Data Science explosion, we are faced with the dual challenges of building statistical and data science pedagogical materials for large populations with a wide range of backgrounds – many of which have little to no coding experience - while trying to understand and develop best practices and methodologies for the complex data science workflow. In short, we need to both teach and research data science, and we need to do it at scale. Here we present ISLE (Interaction Statistics Learning Environment), a browser-based platform that reduces the computing cognitive load and lets students explore Statistics & Data Science concepts in both structured and unstructured ways. We track every click, word, and decision made along the data analysis pipeline, studying “the science of data science”. Platform features also include learning dashboards for students and instructors, a rich case study repository, data provenance, peer review, and real-time collaboration with contribution tracking. We share insights gained from the use of ISLE with hundreds of students in both Statistics and non-Statistics courses at multiple universities as well as corporate executive education training.
|