Who Takes CBL Courses in Statistics? Could Selection Bias Be a Good Thing?

• Activity Number: 238
• Type: Topic Contributed
• Date/Time: Monday, August 10, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Scientific and Public Affairs Advisory Committee
• Author(s): Lynne Schofield*
• Companies: Swarthmore College
• Keywords: Community Based Learning ; Data Analysis Courses ; Statistical Education

Abstract

This paper focuses on comparing students in two different data analysis courses, taught in the same semester by the same professor. One course included a community-based learning (CBL) component and the other did not. Results suggest that students who chose to take the CBL course were more likely to have less mathematical and statistical background (as measured by the number of previous college-level mathematics and statistics courses), were more interested in community service, and wanted a statistics course that aligned closely with their career goals and aspirations. In comparison, students in the more traditional course had stronger mathematical and statistical skills, were more interested in a traditional methods course that employed a variety of "case-studies," and had less time for and interest in a community project. The results suggest there is a place for both CBL and traditional courses in college curricula. The students tended to appropriately self-select into the course that met their needs. CBL courses may also be a good place to draw in students who might otherwise not choose to take a data analysis course in college.