Relations between modeling behavior and learn- ing in a Computational Thinking based science learning environment

Abstract

Computational Thinking (CT) is becoming a core element of current K-12 science curricula. In spite of the known synergies between CT and science education, few attempts have been made to leverage this synergy, especially at the middle school level. In previous work with the Computational Thinking for Simulation and Modeling (CTSiM) environment (Basu, et al., 2013; 2014; Sengupta, et al., 2013) we have demonstrated strong synergies between learning science content and CT, with students showing strong learning gains in kinematics, ecology, and CT concepts. However, when working in CTSiM, students also faced a number of difficulties in understanding science concepts and CT constructs, and had to be scaffolded 1-1 by the classroom teacher or the researchers conducting the study (Basu, et al., in review). To better scaffold their learning, we developed a set of hypertext resources and formative assessment quizzes in the system. This paper reports a teacher-led, multi-domain classroom study con- ducted with 5th grade students using CTSiM. Our results, based on pre- to post-test gains, and the accuracy of the students’ CTSiM models, demonstrate significant and synergistic learning of science concepts and CT skills in Kinematics and Ecology domains, along with transfer of CT skills across domains.