An Integrated Faculty Professional Development Model Using Classroom Sensing and Machine Learning to Promote Active Learning in Engineering Classrooms

点击翻译按钮获取中文摘要

This project aims to serve the national interest by enhancing the teaching effectiveness of engineering faculty. Faculty training is known to be a key factor for successful integration of evidence-based teaching practices such as active learning in STEM classrooms. However, faculty also need opportunities for frequent feedback and reflection on the effectiveness of their use of active learning strategies. The project will combine both training and feedback in a faculty development model it calls TeachActive. The TeachActive model will be implemented with 30 engineering faculty across multiple semesters. TeachActive will use the EduSense open source platform to automatically gather real time descriptions of student and teacher behaviors in the classroom. These classroom analytics will be displayed in a dashboard to provide faculty with feedback about the level of active learning in their classrooms. It is expected that this feedback, together with faculty reflection, will promote systematic improvements in evidence-based teaching by the faculty participants. As a result, the project has the potential to accelerate engineering educators’ adoption and effective implementation of active learning strategies in engineering classrooms, which, in turn, can positively influence student engagement and learning.The TeachActive professional development model will embed automated classroom observation and analysis within a theoretically grounded and evidence-based professional development framework. Integration will include three consecutive components: (1) active learning training; (2) four-week sessions of automated classroom observations with a one-year follow-up; and (3) feedback and reflection. The research team will collect quantitative and qualitative data to monitor changes in instructors’ use of active learning, teaching beliefs, facilitation strategies, and reflective practices. The data for the project will be collected in four stages: (1) pre-post teaching beliefs survey; (2) classroom analytics; (3) reflection prompts; and (4) semi-structured interviews of faculty. The Approaches to Teaching Inventory instrument will be used to assess faculty beliefs about pedagogical strategies before and after their participation in TeachActive. Classroom analytics tracked by EduSense (an NSF-funded camera-based classroom sensing system developed at Carnegie Mellon University) will reveal behavioral indicators of active learning facilitation strategies in classrooms as well as students’ behavioral engagement data. The machine learning techniques within the EduSense system will be validated for the specific context at Iowa State and behavioral features of interest (e.g., sit vs stand; hand raises; kinesthetic patterns) thus providing automated context-sensitive feedback via the TeachActive dashboard. This project is supported by the NSF IUSE: EHR Program, which supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.