Using Dynamic Network Analysis to Understand Idea Diffusion in Open-Ended Makerspaces With App Inventor

• In Event: Structured Poster Session: Analytical Frameworks to Advance the Study of Making: Themes in Research and Development on Makerspaces

Abstract

Objectives or purposes
When understanding learning in open-ended, exploratory maker activities, the processes of creating are often more important than learners’ final products (Vossughi et al., 2013; Daskolia, & Kynigos, 2012). While collaboration is touted as an important element in maker activities (Petrich, Wilkinson & Bevan, 2013), capturing and understanding how this collaboration manifests itself in learners’ individual creative processes is often difficult. This paper introduces a mixed methods approach combining interaction analysis with learning analytics to understand how knowledge spreads within a group of makers engaged in an open-ended exploratory activity focused on the Internet of Things (IoT).

Theoretical Perspectives
This paper examines making and learning through a divergent inquiry perspective (Tissenbaum et al., 2017a), which aims to understand how collaboration happens within open-ended and exploratory environments. In divergent inquiry, learners are encouraged to follow their own pathways and to discover multiple ways of understanding a particular problem and its possible solutions (Lyons et al., 2015). Divergent inquiry recognizes that collaboration can be fruitful even when participants move away from shared understandings of a goal through adaptation and differentiation, as it provides opportunities for learners to compare or contrast their own tinkering to their peers’ (Tissenbaum et al., 2017a).

Methods, techniques, or modes of inquiry
This study engages twenty students in a summer camp aimed to support underrepresented high school students in developing math and science literacies. In the 5-week (two hours each week) maker curriculum, students use App Inventor, a blocks-based programming language, to develop IoT solutions to issues in their communities. Students are encouraged through an iterative design process to self-identify these issues and their solutions. To understand how collaboration helps foster problem solving, we employ a mixed methods approach which combines social network analysis (SNA – de Laat et al., 2007) and interaction analysis (Jordan & Henderson, 1995) to recognize when and how code moves across leaners’ projects.

Data sources
Data is drawn from the incremental log files from App Inventor (which creates snapshots of users’ code every time a change is made), and on video camera that captured whole class interactions and two cameras that focused on two specific groups to give us finer grained understandings of discourse and collaboration.

Results
Analysis is currently underway, however building off previous work that has highlighted the ways that learners collaborate within exploratory and divergent inquiry spaces (Tissenbaum et al., 2017b), we expect to reveal the nuanced ways learners share ideas and adapt the work of others into their own designs and problem solving. By applying SNA to the entire class’ coding projects, we can also get a better sense of the knowledge diffusion of particular programmatic approaches.

Scholarly significance
The findings in this study are of particular relevance to the field as it shows how combining learning analytics and qualitative interaction analysis can provide a more nuanced understanding of how collaboration supports knowledge diffusion across learners in open-ended, divergent inquiry. Further, the use of SNA to specifically track code adoption among learners is a particularly novel approach.

Authors

• Mike Tissenbaum, Massachusetts Institute of Technology

• Mark Sherman, Massachusetts Institute of Technology

• Josh Sheldon, Massachusetts Institute of Technology

• Hal Abelson, Massachusetts Institute of Technology