Reasoning with Multivariate Evidence.

Ridgway, J.; Nicholson, J.; McCusker, S.

Abstract

Reasoning with evidence about realistic situations - education, health, crime, social change, climate change - is inherently problematic. Evidence is often multivariate; relationships between variables are rarely linear; variables interact, and show effects of different sizes, over different timelines. If people are to use evidence to inform and understand public debate, to function in complex, fast-moving commercial environments, and to make important personal decisions, these sophisticated statistical notions need to be part of 'common sense'. For this to happen, we need to understand just how difficult it is to acquire these sophisticated notions, we need curriculum activities that develop these ideas, and we need some ways to assess the acquisition of these skills, if they are to be taken seriously by students and teachers. Here, we focus on the difficulty of reasoning using multivariate data, and explore possible consequences in the development of reasoning from multivariate data.

Citation

Ridgway, J., Nicholson, J., & McCusker, S. (2007). Reasoning with Multivariate Evidence

• Journal Article Type: Article
• Online Publication Date: Mar 4, 2008
• Publication Date: 2007-10
• Journal: International Electronic Journal of Mathematics Education
• Peer Reviewed: Peer Reviewed
• Volume: 2
• Issue: 3
• Pages: 245-269
• Public URL: https://durham-repository.worktribe.com/output/1564219