Constructing Scientific Arguments Using Evidence from Dynamic Computational Climate Models

Pallant, A., & Lee H.-S. (2015). Constructing scientific arguments using evidence from dynamic computational climate models. Journal of Science Education and Technology. 24 (2-3) 378-395. doi 10.1007/s10956-014-9499-3.

Lead Organization(s): 
Year: 
2014
Short Description: 
Modeling and argumentation are two important scientific practices students need to develop throughout school years. In this paper, we investigated how middle and high school students (N=512) construct a scientific argument based on evidence from computational models with which they simulated climate change. We designed scientific argumentation tasks with three increasingly complex dynamic climate models. Each scientific argumentation task consisted of four parts: multiple-choice claim, open ended explanation, five-point Likert scale uncertainty rating, and open-ended uncertainty rationale. We coded 1,294 scientific arguments in terms of a claim’s consistency with current scientific consensus, whether explanations were model based or knowledge based and categorized the sources of uncertainty (personal vs. scientific). We used chi-square and ANOVA tests to identify significant patterns. Results indicate that (1) a majority of students incorporated models as evidence to support their claims, (2) most students used model output results shown on graphs to confirm their claim rather than to explain simulated molecular processes, (3) students’ dependence on model results and their uncertainty rating diminished as the dynamic climate models became more and more complex, (4) some students’ misconceptions interfered with observing and interpreting model results or simulated processes, and (5) students’ uncertainty sources reflected more frequently on their assessment of personal knowledge or abilities related to the tasks than on their critical examination of scientific evidence resulting from models. These findings have implications for teaching and research related to the integration of scientific argumentation and modeling practices to address complex Earth systems.
Resource Type: 
Publication