Children's Constructions of Data, Chance, and Inference in Design Contexts

We used a clinical methodology to explore elementary students' reasoning about data modeling and inference in the context of long-term design tasks. Design tasks provide a framework for student-centered inquiry (Perkins, 1986). In one context (Study 1), a class of fifth-grade students worked in six different design teams to develop hypermedia documents about Colonial America. In a second context (Study 2), a class of fifth-grade students designed science experiments to answer questions of personal interest. In the first, a hypermedia design context, students compared the lifestyles of colonists to their own lifestyles. To this end, ten "data analysts" developed a survey, collected and coded data, and used the dynamic notations of a computer-based tool, Tabletop (Hancock, Kaput, & Goldsmith, 1992), to develop and examine patterns of interest in their data. Tabletop's visual displays were an important cornerstone to students' reasoning about patterns and prediction. Analysis of student conversations, including their dialog with the teacher-researcher, indicated that the construction of data was an important preamble to description and inference, as suggested by Hancock et al. (1992). We probed students' ideas about the nature of chance and prediction, and noted close ties between forms of notation and reasoning about chance. In the second study involving the context of experimental design, we consulted with two children and their classroom teacher about the use of a simple randomization distribution to test hypotheses about the nature of extra-sensory perception (ESP). Here, experimentation afforded a framework for teaching about inference grounded by the creation of a randomization distribution of the students' data. We conclude that design contexts may provide fruitful grounds for meaningful data modeling.