This is a session of Topic 4: Statistics education at the post secondary (tertiary) level

Other sessions under topic 4 ... 4B: Less parametric methods in statistics 4C: Methods for ordinal data analysis 4D: Innovations in teaching statistics at the tertiary level 4E: Heterogeneity of student levels 4F: Sensible use of multivariate software 4G: Learning statistics through projects 4H: Integrating consulting with graduate education 4I: Integrating Bayesian methods with traditional statistics education 4J: Sampling populations

(Monday 12th, 11:00-12:30)

A taxonomy of statistics courses

Organizer

• Alison Gibbs (Canada)

Abstract

A traditional course of study for undergraduate students specializing in statistics may begin with training in mathematics and probability followed by a theoretical introduction to testing and estimation based firmly in parametric methods. A traditional syllabus for a statistics course for students from other disciplines forgoes the theory, instead offering a series of recipes for data analysis emphasizing parametric methods based on the normal distribution. Consideration of the development of the quantitative skills required for efficient citizens is often relegated to a rarely required course in statistics and society. Does our course organization allow us to adequately train our students?

Papers in this session will address how we organize and classify our statistics course offerings and how this organization should respond to new approaches to data analysis. Issues to be considered include:

• Should our courses be classified by student? Does the dichotomy of courses for experts and practitioners make sense? Should we introduce our future experts to statistics with a different approach than we use for future practitioners from other disciplines? And do practitioners in biology require a different introduction than practitioners in business or practitioners in other disciplines?
• Should our courses be classified by method? Does an immersion in data through case studies sufficiently serve our students? Or do we need more structure based on statistical topics? Which of the 1000’s of potential methods should we teach? Should training in statistics be training in a sequence of methods, or development of a broader methodology?
• Should our courses be classified as theoretical or applied? What statistical theory should be isolated to separate courses in mathematical statistics?

Papers

Paper	Title	Presenter(s) / Author(s)
4A1	Banishing the theory-applications dichotomy from statistics education	Larry Weldon (Canada)
4A2	Accommodating specialists and non-specialists in statistics courses	Kevin Keen (Canada)
4A3	Specialized basic courses for engineering students: a necessity or a nuisance?	Lena Zetterqvist (Sweden)

Other sessions under topic 4 ... 4B: Less parametric methods in statistics 4C: Methods for ordinal data analysis 4D: Innovations in teaching statistics at the tertiary level 4E: Heterogeneity of student levels 4F: Sensible use of multivariate software 4G: Learning statistics through projects 4H: Integrating consulting with graduate education 4I: Integrating Bayesian methods with traditional statistics education 4J: Sampling populations