Capturing Computational Making: What Screencasts Tell Us About Learning in a Digital Making Activity
American Educational Research Association Annual Meeting
American Educational Research Association
New York, NY
Within the maker movement, there is growing interest in computational making activities through which youth construct artifacts that bridge the physical and digital worlds (Rode, et al., 2015). Inspired by Papert’s (1980) Logo Turtle, which provides students access to tangible, concrete feedback of their digital constructions, other computational making activities that bridge the physical and digital have emerged such as Scratch+Makey Makey (Lee, Kafai, & Vasudevan, 2014) and e-textiles (Buechley, 2006). As scholars begin to explore the computational learning that happens across these digital making activities (e.g., Dr. Scratch, Moreno-León, Robles, & Román-González, 2015), we contribute timely insights to a new methodological approach for capturing computational skills within these activities. In this project, we present our methods examining screencasts as a tool to measure learning, particularly computational thinking skills (Grover & Pea, 2013), in a digital making activity. Our aim is to understand what, where, and how youth exhibit computational thinking skills in making games on Augmented Reality and Interactive Storytelling (ARIS), a narrative-based programming platform for non-programmers (Holden, Dikkers, Martin, & Litts, 2015). We completed an after school workshop consisting of six 2-hour sessions with 9-15 year olds during which they tackled the design challenge of using ARIS to “Make Your Own Pokemon-Go.” We collected and coded over 50 hours of screencasts of students’ digital construction, which captured all on-screen activities in the ARIS editor. Our analyses entailed coding step-by-step on-screen actions of student game construction (initial coding) as well as CT skills (theoretical coding) as described by Grover and Pea (2013). Screencast data offer deep insights for real-time patterns of digital making. There are two unique components to screencast data: (1) on-screen activity and (2) audio activity. For example, Gracie was working on building a factory, the ARIS element that uses algorithmic thinking to enable items to spawn in the game world, for the “PUPPY!” item in her game. She wonders aloud “what does ‘maximum in game at any time’ [a variable in the factory] mean?” A nearby instructor hears her query and reinterprets the variable meaning by asking “how many are going to be spawning on your app at one time?” Grace then enters two, but quickly says “no” and changes it to one (Screencast, 03/16/17). With on-screen activity, we gained unique insights to the Gracie’s technical game creation, editing, and debugging. The audio data provide the contextual interactions that surround the technical game creation including think aloud narration and conversations with peers or instructors. Unlike log data or interviews, screencasts offer a more contextualized insight to youths computational learning throughout the design process. Hence, they offer valuable in-situ perspectives on the digital making process by synchronizing on-screen making patterns with the explanations of design decisions, questions about game variables, and collaboration between peers, as well as demonstrations and walkthroughs of their games. Our next step is to triangulate log data, screencasts, and other qualitative data sources (e.g., fieldnotes, interviews) to further contribute to understandings of computational thinking skills and collaboration practices within digital making contexts.
Mortensen, C.K, Litts, B.K., *Lewis, W.E., & *Benson, S.R. (2018, April). Capturing Computational Making: What Screencasts Tell Us About Learning in a Digital Making Activity. American Educational Research Association Annual Meeting: New York, NY.