Understanding Distributions by Modeling Them


Authors: 
Cliff Konold , Anthony Harradine and Sibel Kazak
Volume: 
12(3)
Pages: 
Online
Year: 
2007
Publisher: 
International Journal of Computers for Mathematical Learning
URL: 
http://www.springerlink.com/content/lx2213721548n428/
Abstract: 

In current curriculum materials for middle school students in the US, data and chance are considered as separate topics. They are then ideally brought together in the minds of high school or university students when they learn about statistical inference. In recent studies we have been attempting to build connections between data and chance in the middle school by using a modeling approach made possible by new software capabilities that will be part of TinkerPlots 2.0 (TinkerPlots is published by Key Curriculum Press and has been developed with grants from the National Science Foundation (ESI-9818946, REC-0337675, ESI-0454754). Opinions expressed here are our own and not necessarily those of the Foundation.). Using a new Sampler object, students build "factories" to model not only prototypical chance events, but also distributions of measurement errors and of heights of people. We provide the rationale for having students model a wide range of phenomena using a single software tool and describe how we are using this capability to help young students develop a robust, statistical perspective.

The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education

register