Scientific Program > Topic 7 > Session 7F >
Presentation 7F3. Designing Statistical Software: Lessons from StatPlay

Sue Finch (Australia)


Presentation Abstract
StatPlay is interactive software designed to allow students at an introductory tertiary level explore some basic concepts of descriptive and inferential statistics. The broad principles of StatPlay's design were strongly influenced by the work of Barbara White who used computer simulations to help overcome physics misconceptions, and the writing of Edward Tufte on graphical display of information. White's work suggested that students would benefit from multiple representations of concepts that were dynamically linked and easily manipulated; learning environments should be flexible and allow exploration. Tufte's influence was toward a spare, clean user interface.

In practice, the design and development of StatPlay benefited from input from students, teachers of statistics, philosophers, cognitive psychologists and computer programmers. A program of curriculum development and empirical research into the use of StatPlay accompanied the development work. In this paper, we consider the lessons from the StatPlay project for the key questions asked in this session.

Over several years, we had opportunities to observe students at different year levels, in a range of disciplines and in a variety of learning contexts using StatPlay. We also carried out a number of different types of research studies to investigate the effectiveness of StatPlay in overcoming some common misconceptions about statistical concepts. We discuss how the learning context and the task that students are set are important to effective use of software. We believe interactivity of the software, and interactions between students are key factors. StatPlay and other statistical software like it offer new opportunities to explore and interrogate the properties of statistical concepts in a natural and easy way.

Designing statistical software for StatPlay required development of a framework for describing and linking simulations relating to descriptive and inferential statistics, as well as choosing appropriate representations for particular statistical 'objects' or concepts. These tasks can challenge our own understanding of statistics. We believe that good representations can provide students with important building blocks for knowledge that they otherwise find difficult to articulate.

The development of a demonstration tool - SAM - that allowed recording and playback of multimedia demos with StatPlay was an important step forward in the design of flexible curriculum materials. Classroom activities built around these multimedia demos also proved to be useful research tools. We discuss the research potential of a tool that allows direct recording of interactions with the software.

Work on StatPlay involved an interplay of software design, classroom use and empirical research. The potential for learning offered by interactive software seems great and observations of students' activities often suggest that real insights are achieved. However evaluation of the effectiveness of statistical software, like that of other methods of teaching statistics, is often limited by the particularities of the context in which the evaluation is made. However, there are many other opportunities such software offers to learning about how people understand statistical concepts.


Download in Word format (DOC).
Download in Adobe Acrobat format (PDF).


Back to top

ICOTS-6, The Sixth International Conference on Teaching Statistics - International Program Committee (IPC) Website.
Copyright © 2001 by the IASE. All rights reserved. This information is subject to change without notice. This page was last modified on July 2, 2002.
For questions or comments, contact the Webmaster, Dagan Ben-Zvi.