Opinionated Practices for Teaching Reproducibility: Motivation, Guided Instruction and Practice

• DOI: 10.1080/26939169.2022.2074922
• Publication Date: 2022-06-23T18:16:26Z
• Authors: Joel Ostblom; Tiffany Timbers
• Keywords: FOS: Computer and information sciences; 0303 health sciences; LC8-6691; 4. Education; Special aspects of education; Reproducibility; QA273-280; Data science; Education; Methodology (stat.ME); Computer Science - Computers and Society; 03 medical and health sciences; Computers and Society (cs.CY); Curriculum; Probabilities. Mathematical statistics; Statistics - Methodology

In the data science courses at University of British Columbia, we define as study, development and practice reproducible auditable processes to obtain insight from data. While reproducibility is core our definition, most learners enter field with other aspects in mind, for example predictive modelling, which often one interesting topic novices. This fact, along highly technical nature industry standard tools currently employed science, present out-of-the gate challenges teaching classroom. Put simply, students are not intrinsically motivated learn this topic, it an easy them learn. What can a educator do? Over several iterations focused on workflows, have found that providing extra motivation, guided instruction lots key effectively challenging, yet important subject. Here examples how deeply motivate, guide provide ample opportunities engage learning about topic.