Scientific Program > Topic 7 > Session 7F >
Presentation 7F5. Challenges in Designing Data Analysis Software for Young Students

Cliff Konold (USA)


Presentation Abstract
What kind of a tool do young students need for doing, and learning about, data analysis? Roshelle, Kaput and Stroup (in press) have described the process of designing educational software tools in a particular domain as a search for the "sweet spot," or fruitful intersection, of three spheres: a) students' prior domain knowledge, b) the unique affordances of the computer, and c) the conceptual structure of the target domain. In this paper, I describe our effort to locate this sweet spot in designing Tinkerplots, data analysis software targeted for students aged 11-14.

Research has shown that students come to the study of data analysis without knowing the difference between various data types (nominal, ordinal, ratio), without an explicit understanding of the difference between characteristics (tall) vs. variables (height), and without knowledge of the conventions of 2-D representations. Although these are important objectives of instruction, students should not need to understand these ideas to begin analyzing data. Tinkerplots comes with no built-in plot types such as bar graphs, pie charts, and scatterplots. When a data set is first opened, a plot window appears showing a haphazard arrangement of data icons on the screen, where each icon represents an individual case. Students then use primitive operators, such as separate, order and stack, to progressively organize the icons into arrangements that are useful in answering questions they have about the data. The other advantage of this design is that students are not as apt as they are with more conventional tools to adopt a trial-and-error approach in producing data representations.

Research has also identified as a major hurtle for students coming to perceive a data set as more than a collection of individual elements, but as a group with emergent characteristics. Tinkerplots attempts to support this change by providing smooth transitions between representations where each individual case is clearly visible (e.g., dot plots,) to representations that aggregate cases such that they become invisible (e.g., histograms and pie charts). It also includes visual and auditory supports that help structure students' perceptions of characteristics of distributions. Listening to the changes in a frequency distribution while sweeping a line from lower to higher values, or "replaying" a time series plot are examples of such supports. In the former case, a sound much like popcorn popping - starting slowly, growing rapidly to a crescendo, then dwindling off - may sensitize students to how it is they perceive group variability in real-life situations. This could, in turn, help them perceive it in graphic forms.


Download in Adobe Acrobat format (136 Kb).


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.