Using 'Real Data' for Teaching in the Health Sciences: Benefits, Challenges, and Opportunities

• Activity Number: 127
• Type: Topic Contributed
• Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Teaching of Statistics in the Health Sciences
• Author(s): Ronald Gangnon* and Amy Nowacki* and Steven Grambow* and Laila Poisson* and Jacqueline Milton*
• Companies: University of Wisconsin and Cleveland Clinic and Duke University and Henry Ford Health System and Boston University School of Public Health

Abstract

The ASA-endorsed Guidelines for Assessment and Instruction in Statistics Education (GAISE) College Report encourages instructors to "use real data" in teaching. In the health sciences, real data is more interesting and motivating for students. In addition, real data can help students learn to deal with messy data, to read and use codebooks, and to connect data analysis with published results. However, it is not easy to find health sciences datasets for use in teaching. Useful datasets are often widely dispersed on various websites and difficult to find efficiently. Additional challenges include datasets for published studies that are not publicly available, inadequate documentation, HIPAA issues, proprietary information issues, formatting complexities, etc. The speakers in this panel will discuss the benefits and the challenges involved in using real data in teaching in the health sciences, and will describe some approaches to make real data more readily available for teaching.