Are Undergraduates Ready for Generalized Linear Models and Correlated Data Methods?

• Activity Number: 414
• Type: Topic Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Education
• Author(s): Paul Roback*+
• Companies: St. Olaf College
• Keywords: undergraduate curriculum ; generalized linear models ; correlated data methods ; multilevel models

Abstract

Undergraduate courses in applied regression techniques have become increasingly common with advances in student preparation and interest in statistics. Yet in our experience, student projects at the end of regression courses often are limited in scope, because the real data that students want to analyze (from public health, psychology, education, etc.) involve non-normal responses, correlated data, or predictors at several different levels. Therefore, GLMs and correlated data methods fit well as part of a year-long sequence in applied modeling, building upon a course in applied regression. In fact, "Advanced Statistical Modeling" is one of two core courses taken by all undergraduate statistics concentrators at St. Olaf College. Over eight years, we have found this material to be both appropriate and exciting for our undergraduate statistics concentrators, half of which are mathematics majors and half of which are majors in other disciplines with potentially minimal backgrounds in mathematics. In this talk, we will describe our experiences teaching Advanced Statistical Modeling at St. Olaf--intended learning outcomes, topics covered, and lessons learned over several iterations.