Bayesian Analysis for Assessing Equivalence in Delivery of Graduate Statistics Education Between Synchronous Distance Learning Versus Traditional Face-to-Face Learning Students

• Activity Number: 185
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Teaching of Statistics in the Health Sciences
• Author(s): Milind A. Phadnis* and Jo Wick
• Companies: University of Kansas Medical Center and University of Kansas Medical Center
• Keywords: hierarchical model ; test of equivalence ; Bayesian analysis ; distance learning ; statistics education

Abstract

Empirical evidence on the equivalence of delivering graduate statistical education to distance learning students versus traditionally trained face-to-face students is limited mainly due to a lack of randomized controlled studies. Despite this, there is an ever increasing demand for dissemination of online statistics education owing to the flexibility that distance learning offers in terms of overcoming numerous geographical barriers and conflicting student schedules. To meet this demand, a university that offers synchronous learning options to distance learners in addition to the traditional face-to-face training to in-class students has to ensure that the two modes of delivering education are equivalent in terms of student performance and that such equivalence remains continually effective. To attain this goal, we present analysis using a Bayesian hierarchical model and perform a formal test of equivalence between the two modes of learning. The analysis is done both at individual student level as well for grouped course data and evaluates equivalence in student performance both in terms of their continuous overall course scores as well as their final categorical course grades.