Group projects in project-based courses are designed as opportunities for students to learn from each other in a collaborative setting. In a race to maximize efficiency, however, students often resort to divide-and-conquer strategies subverting the opportunity to learn from the collaboration. These challenges are exacerbated online as differing student schedules make it harder to work together on projects. This work investigates the use of a task structuring and role-assignment paradigm to support collaborative project-based learning that discourages divide-and-conquer. The paradigm, designed based on the Script Theory of Guidance, is embedded in synchronous programming activities offered in an online project-based software engineering course. This study then compares this paradigm to the use of additional conversational agent-based prompts to enhance argumentative discussion in an effort to improve learning over and above what can be achieved with the paradigm on its own.
Sreecharan "Sree" Sankaranarayanan is a 5th year Ph.D. student at the Language Technologies Institute of Carnegie Mellon University. His research focuses on the use of conversational agents to support group-work in online learning contexts. As a part of the Technology for Effective and Efficient Learning (teel.cs.cmu.edu) lab, his work has been applied to project courses offered to several hundred students each semester at Carnegie Mellon University and its worldwide branch campuses. As the co-director of Discussion Affordances for Natural Collaborative Exchange (dance.cs.cmu.edu), his prior work, in partnership with edX, brought conversational agent-based support to several thousand students online. He currently serves as the Founder Co-Chair of the International Learning Sciences Student Association (ILSSA), a committee of the International Society of the Learning Sciences (ISLS) for students of the learning sciences.