This is a session of Topic 5: Statistics education in the disciplines and the workplace

Bridging the gap between current statistical practice in the workplace and modern statistics

Organizers

• Jennifer Brown (New Zealand) : Session chair
• John Maindonald (Australia)

Abstract

For the purposes of this session, modern statistics is a comprehensive term that has in view the mathematical theory, its adaptation to handle practical data analysis, and the assistance and insights that are available from the use of computing technology. Simulation, and the use of resampling methods, are increasingly important in the matching of theory to the demands of practical data analysis.

Barriers to bringing modern statistical ideas and practice into the workplace include a conservatism that prefers to stay with familiar tools and approaches, the need for time and training that will allow the mastery of new ideas and approaches, and a resistance in some applied disciplines to any methodology that strays outside of bounds that have become rigidly fixed. Modern insights are especially needed in those cases where practices that have become traditional are now known to be less than optimal, or even likely to mislead.

Additionally, technology has made available data of an extent and/or types that were previously unknown. The new analysis demands thus created have grown much faster than the supply of trained statisticians. Inevitably, much of the needed analysis will be carried out by individuals with very limited statistical training.

How can university teaching and research best respond to these changing demands? What is the role of in-service training and learning? Where should effort be focused in the development of new educational resources? How can we leverage new technology to assist these tasks? These are some of the questions which will be addressed in the framework of this session.

Papers