An analysis of student’ patterns of exploration with NetLogo models embedded in the Connected Chemistry environment

How can we characterize the ways in which students explore computer models to find needed information? Do the types of goals that guide their exploration affect these characteristics? Is the path of exploration affected by the model’s interface tools? By the kinds of mathematical relationships governing the model’s target phenomenon? In the Modeling Across the Curriculum project, we enable students’ exploration of computer models that are embedded in a supporting script. The Connected Chemistry learning environment (Levy & Wilensky, 2004; Levy, Novak & Wilensky, 2005; Stieff & Wilensky, 2003) is one such model-based curricular unit. Connected Chemistry employs multi-agent NetLogo (Wilensky, 1999a) models to empower the students’ manipulation and observation of chemical “entities” at the molecular level as well as the resulting aggregate patterns. These models are embedded in Pedagogica (Horwitz, 2002) scripts that guide the model exploration as well as asking students questions about their exploration and findings. The first Connected Chemistry unit is on the topic of gases: Gas laws, and Kinetic Molecular Theory. In this paper, we provide an analysis of student’s explorations within computerized models, as derived from computer logs of their actions and the model’s changing properties. We have conducted four studies if of patterns of students’ model exploration. The studies differ in (a) We observe students employing four distinct patterns; (b) Students are consistent in their use of a specific pattern; (c) Some specific features of these patterns change when the goals change; (d) Half of the students’ took advantage of the affordances of more powerful exploration tools to Levy & Wilensky, 2005 4/10/05 2 improve their search for information; (e) Most of the students’ adapted their exploration strategy to the underlying mathematical relationships. The results are discussed in terms of science inquiry skills, styles in information search in digital spaces and the educational implications of these findings. We then discuss the planned future work in tying these findings in with the students’ prior knowledge and learning outcomes.