(Mis)match of Students’ Country of Origin and the Impact of Collaborative Learning in Computer Science

Considerable research has demonstrated the educational efficacy of active and collaborative learning in various postsecondary contexts. However, group composition and intergroup dynamics have the potential to shape the quality of student experiences and the outcomes of collaborative learning. Previous studies have extensively examined the impact of interactions across some forms of difference (e.g., race/ethnicity), but very little research has explored the role of college students’ country of origin or language in collaborative learning contexts. This issue is particularly important within increasingly international and multicultural societies throughout the world. Therefore, the present study randomly assigned partners to students for participation in pair programming, which involves close collaboration to complete a computer science coding task. Within a sample of 819 responses from 369 undergraduates in the United States (US), non-US citizens benefitted from having a partner from another country (primarily the US) in terms of the amount of lab assignment completed, belief that the assignment was virtually error-free, and confidence in quality of the submitted assignment; however, these students were also less involved in writing code during pair programming when they had a partner from a different nation than from their own nation. In contrast, the national origin of US citizens’ partners generally did not affect their outcomes. These findings support the use of mixed-nationality partners in pair programming and potentially other collaborative learning contexts, which diverges from prior studies that suggest facilitating similarity between partners in terms of demographics, personality, and prior programming experience.