A Systems Approach to Analyzing Design-Based Research in Robotics-Focused Middle School STEM Lessons through Cognitive Apprenticeship

摘要

Abstract Under the design-based research (DBR) process, the design and implementation of a lesson undergoes several iterations, the outcomes of each iteration are analyzed, and the necessary changes and refinement in the design are proposed for implementation and adaptation in the next iterations. Such a strategy may ultimately improve learning outcomes and help yield novel learning theories and artifacts. The DBR approach is especially important to design STEM lessons with robotics because the flexibility and uncertainty of robotics applications require investigations on various alternative approaches to identify the most appropriate, feasible, reliable, and cognitively appealing approach for using robotics in the lesson design that promote students' intrinsic and extrinsic motivation to engage in the lesson. Moreover, application of DBR in STEM education with robotics provides numerous opportunities examine the feasibility and benefits of incorporating constructs of various learning theories such as: cognitive apprenticeship, situated cognition, problem-based learning, and inquiry-based learning. However, application of DBR in STEM education using robotics has not received significant attention yet. From the review of available literature, it is evidenced that while rich qualitative observations and analysis of outcomes guide DBR iterations, a systematic quantitatively-guided approach is lacking. The prevailing DBR strategy may not fully capture true scenario of an iteration and may fail to suggest appropriate modifications for next iteration. This may limit the effectiveness of DBR as currently practiced. This paper is based on our collaboration with a school teacher who is designing and implementing 7th grade math lessons using robotics. First, students build robots based on the instructions of the teacher and then the teacher designs math lessons using robotics and implements them in classes. The learning activities are sequentially performed in several sessions, which we treat as 'iterations.' In each iteration of robot building, the students discuss among themselves and with the teacher and researcher under a student-teacher-researcher co-design approach, which fosters collaborative learning and co-generation. The students receive researcher's expert opinion that provides the advantages of cognitive apprenticeship. In each iteration, two separate groups of students work towards building two identical robots. For one group, the teacher and researcher use traditional qualitative observation, brainstorming, discussion, questionnaire, and feedback methods to analyze the outcomes of the iteration. For the second group, in addition to traditional methods, they follow some advanced systems engineering approaches under the cognitive apprenticeship of the expert researcher. The DBR is treated as a continuous improvement (CI) method, and resembles as the Deming or Plan-Do-Check-Act cycle. They observe the outcomes in each iteration and analyze them using cause and effect diagrams (fishbone analysis). Then, they apply the Pareto principle (80-20 rule) to identify the vital few causes that contribute to the major outcomes. At the end of an iteration, outcomes for the systems approach are compared to that for the traditional approach. The results show that the system approach is more effective. The results are novel that enrich the DBR method and improve its efficacy in designing STEM lessons using robotics for middle school classrooms.