Is Carla grumpy? Analysis of peer evaluations to explore microaggressions and other marginalizing behaviors in engineering student teams

• Published in: Journal of engineering education (Washington, D.C.) D.C.), 2024-07, Vol.113 (3), p.603-634
• Main Authors: Dickerson, Darryl A., Masta, Stephanie, Ohland, Matthew W., Pawley, Alice L.
• Format: Article
• Language: English
• Subjects: Algorithms; coded language; Engineering; Engineering education; equity; Evolutionary algorithms; gender; Mathematics; microaggression; Peer Evaluation; qualitative (free‐text) marginalization in teams; Qualitative analysis; race/ethnicity; Social exclusion; Students; Taxonomy; team dynamics; Teams
• ISSN: 1069-4730; 2168-9830
• DOI: 10.1002/jee.20606

Background Teamwork has become a central element of engineering education. However, the race‐ and gender‐based marginalization prevalent in society is also prevalent in engineering student teams. These problematic dynamics limit learning opportunities, isolate historically marginalized students, and ultimately push students away from engineering, further reinforcing the demographic imbalances in the profession. Purpose While there are strategies to improve the experiences of marginalized students within teams, there are few tools for detecting marginalizing behaviors as they occur. The purpose of this work is to examine how peer evaluations collected as a normal part of an engineering course can be used as a window into team dynamics to reveal marginalization as it occurs. Method We used a semester of peer evaluation data from a large engineering course in which a team project is the central assignment and peer evaluation occurs four times during the course. We designed an algorithm to identify teams where marginalization may be occurring. We then performed qualitative analyses using a sociolinguistic analysis. Results Results show that the algorithm helps identify teams where marginalization occurs. Qualitative analyses of four illustrative cases demonstrated the stealth appearance and evolution of marginalization, providing strong evidence that hidden within language of peer evaluation are indicators of marginalization. Based on the wider dataset, we present a taxonomy (eight categories) of linguistic marginalization appearing in peer comments. Conclusion Both peer evaluation scores and the language used in peer evaluations can reveal team inequities and may serve as a near‐real‐time mechanism to interrupt marginalization within engineering teams.