Process Over Product for Studying Computational Thinking Practices in Introductory Programming

• In Event: Poster Session 1
  In Poster Session: Discipline-Based Research in Computing and Robotics

Abstract

As K-12 computer science educators and researchers seek ways to support learning of computational thinking (CT) through programming in blocks-based programming environments, they are hampered by a lack of understanding of “programming process” that includes hard-to-see sequences of actions signifying problem-solving strategies. To support learning of CT practices such as problem decomposition, debugging, and iterative refinement, we need insights into this process. We present a preliminary framework that balances hypothesis- and discovery-driven data analysis approaches to understand programming process from data logs. In an ongoing pilot research with novice and expert programmers, we use a priori design patterns for CT practices developed using Evidence-Centered Design in conjunction with educational data mining techniques to analyze specific CT practices in programming logs.

Authors

• Shuchi Grover, SRI International

• Marie Bienkowski, SRI International