Replacing traditional sections with Teams-based groupwork: Remote learning and beyond

The sudden shift to online teaching and learning brought about by the Covid-19 pandemic provided an opportunity to consider alternatives to entrenched teaching practices. Making use of the private channel function in Microsoft Teams, I replaced traditional sections in an introductory linguistics course with asynchronous groupwork. This enabled students to form learning communities that facilitated peer learning and support in spite of remote learning, while unexpectedly connecting students with instructors in more personalized ways than typically witnessed in traditional sections. The medium allowed the teaching team to provide tailored feedback on each group’s work, as well as point out errors that were common across groups. I reflect on some of the problems encountered and consider how these might be addressed in the future.

1. Introduction. Discussion sections are commonplace in large university courses, and for good reasons: not only can they provide opportunities for active learning that supplement traditional lectures (e.g. Buckley et al. 2007), they also serve as useful platforms for novice instructors to develop their practice (Thomson & Zamboanga 2008). However, the actual management of sections can be challenging in practice. I report on a teaching intervention that was implemented in an undergraduate introductory linguistics course taught at the National University of Singapore (NUS), where I replaced traditional sections with groupwork operationalized through Microsoft Teams.
1.1. BEFORE COVID-19. The course typically enrolls between 250 to 350 students and had always been taught in a lecture-discussion section format. Prior to the intervention, sections 1 were conducted in-person either by the lecturer or graduate student teaching assistants (GSTAs) and were capped at 25 students per section. Sections met once a week, for a duration of 45-minutes, and involved instructor-led discussions of problem sets that students were expected to attempt prior to the session. Attendance was mandatory and students were graded on the basis of their level of participation and the quality of their contributions. While this model provided a physical space that connected students with one another and with the instructors, there were limitations that were at odds with the goals of justice, equity, diversity, and inclusion. First, there were issues with managing and assessing participation. Students in Singapore tend to be rather shy, so class participation is typically imbalanced: a few outspoken students would dominate the discussions while many others would be unincluded, in addition to being penalized for their lack of class participation. Classroom management was challenging for inexperienced GSTAs, who often faced difficulties moderating discussions due to the uneven levels of participation. Because of the group setting, we were also unable to provide detailed feedback that was tailored to students' individual weaknesses, and shy students tended to avoid seeking clarifications.
1.2. COVID-19 AND THE SHIFT TO REMOTE LEARNING. In the semester that the course was to be conducted, the University mandated a campus-wide shift to remote learning due to the pandemic. This introduced new issues for the teaching team. First, the pre-existing difficulties with managing participation in a physical classroom would likely be compounded if we were to attempt to replicate the existing model in a synchronous online setting. Moreover, given the importance of community in online learning arrangements (e.g. Angelino et al. 2007, Liu et al. 2007, I was concerned that students, particularly freshmen with no prior experience in higher education, who make up the bulk of the enrolment, might become disengaged from the learning environment due to the lack of in-person interaction.

Intervention.
With these considerations in mind, I decided to modify how sections would be conducted: instead of its traditional function as a forum for small-group instruction and discussion, I decided to use the section as an administrative unit for the organization of smaller groups. Students in each section randomly formed sub-groups of three to four members, which we referred to as 'gangs', 2 and membership was maintained throughout the semester. A Microsoft Team 3 (henceforth 'Team') was created for each section, and within each Team, a private channel was created for each gang. Only gang-members and the teaching team had access to their private channel, making it a dedicated space for each gang to work in. Students were able to leave real-time messages on a message board, initiate video calls, and create/edit documents within their private channels. Figure 1 provides an example of a Team created for a section, where all students in the section/Team had access to the 'General' channel, while Figure 2 illustrates an example of a private channel within the Team. There were six gangs (A-F) in this Team, each with its own private channel. There were 13 such Teams in total.
Every week, students would conduct video calls within their private channels and create a new document to work together with their 'gangmates' on the week's problem set (Figure 3). At the heart of this activity was the notion of peer teaching and learning, which "may be particularly relevant when one seeks to maximise the student's responsibility for his own learning and active participation in the learning process, and to enhance the development of skills for cooperation and social interaction" (Goldschmid & Goldschmid 1976: 29). Problem sets were assigned after lectures, 4 which were conducted synchronously over Zoom, and students could choose whether to use the scheduled section time to work on their solutions. After each week's deadline, the teaching team graded and provided feedback on the solutions.

Evaluation.
As students were graded on the basis of their solutions rather than class participation, this eliminated prior problems associated with managing and assessing participation. The small gang sizes also meant that shy students, who would otherwise not speak up in traditional sections, would face less public/social pressure and could be more proactive in sharing their views. 5 The medium allowed the teaching team to provide high-quality, tailored feedback on each gang's solution, which we could not do in the old model - Figures 4 and 5 illustrate the extended feedback that the students received. In addition, because each section was organized as a Team, we were able to address common errors in the general channel, as shown in Figure 6: every week, the teaching team discussed common mistakes and highlighted them to their respective sections/Teams in the general channel. This allowed gangs that did not make such mistakes to be aware of common pitfalls as well, ensuring equal access to general feedback. In addition to the practical benefits mentioned above, the use of groupwork and the provision of a bespoke space for each gang to work in helped to foster learning communities, in spite of the remote learning situation. For instance, besides meeting weekly in their private channels to discuss the problem sets, some gangs also initiated meetings to revise for midterm tests, demonstrating that they viewed their gang as a learning community whose utility was not limited to working on the problem sets (see Figure 7).    An unexpected, but welcome, outcome of the intervention was that we saw closer interactions between students and instructors than in traditional sections. For example, gangs engaged us with questions about the problem sets in various ways allowed by the medium, e.g. by using the message board in their private channel, requesting video calls with the instructor within their private channel, or inserting questions within their submitted documents. Some students even felt comfortable asking questions about their individual assignments within their gang's private channel. This significance of this cannot be understated, as students are graded on an enforced curve and NUS students are notoriously grade-conscious: the fact that students did not mind their gangmates potentially benefitting from (the instructors' responses to) their questions about their individual assignments demonstrates the extent to which they treated their gang as an in-group learning community. Some of the comments that students left for the instructors in their solutions to the final problem set showed that students appreciated the close student-instructor interactions and saw their gang's private channel as a safe space for learning (see Figures 8 and 9).

Drawbacks.
As noted earlier, one benefit of having traditional sections is that they provide a forum for instructional development on the part of GSTAs (Thomson & Zamboanga 2008). On the one hand, this new arrangement entailed that our GSTAs did not have any practice with actual classroom management; on the other hand, they gained valuable experience in terms of learning to provide detailed, constructive feedback. In order to help GSTAs hone their classroom management skills, I plan to explore a hybrid model in the future, when face-to-face teaching resumes, that incorporates occasional physical meetings for purposes complementary to groupwork on the problem sets, e.g. as review sessions for examinations.
One issue that affected several gangs was that of 'sleeping' or uncooperative gangmates, which is perhaps unavoidable with group assignments. At some level, this is more unconstructive than having nonparticipators in traditional sections. In the case of the latter, discussions may be imbalanced, but at least active students were not unfairly penalized; having a 'sleeping' gangmate could, however, have a negative impact on the entire gang's performance (and thus grade) on the problem sets. Unfortunately, such complaints tended to be raised late in the semester, so when cases were reported, all the teaching team could do was investigate the claims and impose penalties on the offending student where appropriate. In the future, it would be wise to emphasize the need to surface such issues early, so that the instructor can intervene swiftly to limit the handicap to rest of the gang.

Conclusion.
The sudden shift to remote learning precipitated by the Covid-19 pandemic has in many ways negatively impacted students' learning experiences. But it has also given us pause to reflect on previously entrenched practices. My experience with asynchronous groupwork facilitated by Microsoft Teams is that it helps to foster learning communities while addressing many of the issues that I had previously faced in traditional sections that oppose the goals of justice, equity, diversity, and inclusion. Together with other reports on asynchronous activities in the remote teaching of introductory linguistics, e.g. having students contribute to class wikis (Bjorndahl 2021) and having them respond to one another's responses to weekly discussion prompts (Curtis 2021), these collective experiences highlight asynchronous activities as valuable tools for engaging students and fostering learning communities, which may be worth retaining even after life returns to normal.