The integration of computational modelling in science classrooms provides a unique opportunity to promote key 21st century skills including computational thinking (CT) and collaboration. The open-ended, problem-solving nature of the task requires groups to grapple with the combination of two domains (science and computing) as they collaboratively construct computational models. While this approach has produced significant learning gains for students in both science and CT in K–12 settings, the collaborative learning processes students use, including learner regulation, are not well understood. In this paper, we present a systematic analysis framework that combines natural language processing (NLP) of collaborative dialogue, log file analyses of students’ model-building actions, and final model scores. This analysis is used to better understand students’ regulation of collaborative problem solving (CPS) processes over a series of computational modelling tasks of varying complexity. The results suggest that the computational modelling challenges afford opportunities for students to a) explore resource-intensive processes, such as trial and error, to more systematic processes, such as debugging model errors by leveraging data tools, and b) learn from each other using socially shared regulation (SSR) and productive collaboration. The use of such SSR processes correlated positively with their model-building scores. Our paper aims to advance our understanding of collaborative, computational modelling in K–12 science to better inform classroom applications.

No abstract

Researchers have long recognized the importance of using technology to support students' collaboration in learning and problem solving tasks. Recently, there has been a lot of research in capturing and characterizing student discourse and how they regulate each other when they perform learning tasks in pairs or in small groups. In this paper, our goal is to dive a little deeper into how students collaborate, and the learning behaviors they exhibit when working in pairs on a learning by modeling task, while also teaching a virtual agent in the Betty's Brain system. We report the results of a quasi-experimental study, where students were divided into two groups: one group worked in pairs and the other group worked individually. The results illustrate that students in the collaborative group built more correct causal maps than those working individually, and their pre-post test results show significantly higher learning gains in the science content. A differential sequence mining algorithm applied to their action sequences captured in log files showed differences in the learning behaviors between the two groups. The differences imply that the collaborative groups were better at debugging their evolving causal maps than the students who worked individually.