Learning to Code in Early Childhood

In the NSF Coding in Kindergarten (CiK) project, designed and directed by Jody Clarke-Midura (Utah State University) and Co-Pis Jessica Shumway (USU) and Victor Lee (Stanford University), I served as the postdoc studying a popular computational environment for early computing: commercially available tangible robot coding toys. The project used evidence-centered design of assessments and design-based approaches to classroom curriculum implementation to explore how young children are developing computational thinking. Not much is known about early CT, and much of what we do know is based on expert models. In our research, we are operationalizing what is involved for pre-literate children thinking computationally as they learn to program robots to move around their classrooms or create instructions for an agent to navigate around a gridded space.

Extending our definition of CT beyond the cognitive skills involved in learning to program, I led analyses of the social-computational dimensions of CT: affect, equity, social positioning, division of labor, and an ethic of care. We identified these as important after qualitative analyses of Kindergarteners in 5 classrooms showed how collaborative coding requires much more than cognitive skill. It also involves creating a floor-based task-space conducive to all children’s learning, distributing tasks between peers and computational agents, distributing bodies around a classroom, and figuring out how to keep technologies in working order. When tasks- or robots- do break down, children are adept at performing computational caregiving, an important dimension of learning in a digital age.