This is a research project that would contribute to further understanding of STEM teaching and learning in informal settings by conceptualizing and measuring computational thinking for middle school students. Three research questions will be addressed: 1) What is an appropriate definition of computational thinking in middle school? 2) Does creating a computer game promote computational thinking? 3) Under what conditions does pair programming produce greater gains in computational thinking than solo programming? The study will be conducted over three years. Approximately 240 students will use Storytelling Alice to create 3-D animation games over 16 weeks. Participants will be drawn from extended learning programs in two school districts that include primarily Latino and white students. Classes will be randomly assigned to either pair or solo programming conditions. Paired students will choose their partners and statistical analyses designed for longitudinal dyadic interdependent data will be conducted. Data will include pre and post-test surveys, interviews, computer logging data, observations, students games, videotape, and performance assessments. They will also employ a group-randomized controlled trial; randomization will be by class to either a pair programming or solo programming condition. The qualitative and quantitative data will measure computational thinking, students experience programming, closeness with partner, affect, and behavior on the computer. The project will produce evidence about the strength of working in pairs compared with working on technology games alone and it is intended to inform discussions of the meaning of computational thinking.
Elective technology computer game design courses for middle school students are a promising strategy to introduce computer science concepts and ways of thinking to a broad population. The goal of this project was to contribute to research on how to engage middle school students in computational thinking by teaching them to design and program computer games. Over two years, 325 students completed the 20-hour class; the average age was 12 and 50% self-identified as Latino/a. Classes were held both after and during the school day, and followed a curriculum where students first learned to use the child-friendly programming environment called Alice, and then designed and programmed a game. Classes were randomly assigned to a pair or solo programming condition. Contributions include both research findings and resources for educators, researchers, and game programming tool developers. Research findings include a description of computational thinking (CT) in K-12, a range of assessment strategies to measure it, and research results that describe the aspects of CT that students engaged in, and the individual and social factors that explain variation in what students learned by programming computer games. Resources for educators include recommendations for the kinds of support and training that classroom teachers need to teach Alice, how to structure the game design and programming so it can address some of their instructional goals (e.g., engage them in logical thinking, provide individualized learning opportunities), and how to use pair programming to enhance learning outcomes. Resources for researchers include a parser that others can use to get meaningful information from the computer logs generated by each programming action; the parser and our approach for using it will contribute to the growing number of efforts to mine large datasets to better understand student learning, as well as the development of game programming tools that appeal to a more diverse population. In summary, the results contribute to efforts to describe and measure computational thinking in K-12, to understand how teachers can use computer game programming to promote children’s computational thinking, and to describe the conditions under which pair programming is more effective than working alone.
