Learner teams in courses

Teams can provide high-quality interaction even in large courses.

November 28, 2022 - Richard Darst


Part of a series on the Future of Teaching

In April/May 2020, when we started doing "large" (~100 people) online courses, we wanted a way to be more interactive. This was before the time of co-teaching and barely at the start of parallel chat, so we were focused on the personal experience. Our solution was learner teams.

How it worked

The basic mechanics was this:

  • Learners are grouped into teams of ~5 people each. (We try no more than 6 people that don't know each other already, if a team already knows each other then we take whatever size they came with).
  • Some learners registered as teams (in this case, we kept them together).
  • We would also accommodate individual learners by finding suitable teams for them - and this did work well.
  • Each team has an exercise leader assigned to it. It might be someone who separately volunteered, or it might be someone who is already part of the group.
  • Teams are pre-assigned and static over the entire workshop. This takes more work, but means that a community forms - randomly assigning teams each day would not work as well.
  • There are lecture parts, and there are exercise parts. During the exercise parts, everyone is moved to a breakout room. The exercise leaders work with their team to do the exercises and in general be friendly.
  • We have other course staff around ("expert helpers") and they rotate between different rooms and make sure the expert helpers are doing well, if they need any help, and if they are keeping a safe and welcoming course environment.

We have a short ~1-hour training course for the exercise leaders, where we try to motivate their value, give hints on promoting interaction, hints on what happens if things go badly, and especially about motivation of learners and safe learning environments.

We would encourage entire groups to sign up together as a team. Research has shown that when two people (instead of one) learn some new skill, it is much more likely to be adopted by a group. This is the natural extension of that, and it becomes very easy to take entire groups into a course.

For exercise leaders, detailed knowledge about the course material wasn't needed, exercise leaders could come by and help with that. The more important role of the exercise leader was social: being able to keep everyone engaged and make sure that no one felt left out. Knowing a little bit about command line interfaces and not being scared of reading error messages is a nice bonus, but not strictly needed.

Teams seemed to work best when it was actively managed. This doesn't mean forcing everyone to be a part of a team, but if someone is on a team, it's clear they are expected to take part in it. There should be a clear "do you want to take part in a team?" during the registration phase.

Evaluation

Teams worked great for us to scale to ~100-120 people. We could have one instructor per lesson with plenty of staff to support the teams. We could scale up the number of people we could teach at once much more than the traditional 3-instructor model in a classroom.

In principle, this is a lot like "work tables in a classroom". In some ways, it wasn't as nice since it was online. But in other ways, screensharing can allow everyone to see the active screen - something not easily possible with this sized groups in a classroom. Everyone had a clear team, and exercise leaders were clearly responsible for the people on their teams - which meant that fewer people were at a table but not really participating in the group.

When teams already knew each other in advance, it worked exceptionally well. They would usually speak the same domain language and programming language, and use the same tools. Often, one team member had more experience and became the exercise leader. Even when teams didn't know each other in advance, if we could try to put people together based on these criteria, by day 2 they were working very well together.

Benefits

Teams allowed us to scale to an even larger number of people. Our registration was: "We take people up to the capacity of exercise leaders we have. However, we take any team that comes with their own exercise leader, even if we are over our basic capacity". This worked well.

Decrease of teams with the rise of livestreaming

By 2022, the roles of teams has been decreasing (though this isn't exactly a good thing). Our developments in co-teaching, parallel chat, and livestreaming allows us to break free of the limits of zoom. Co-teaching provides engagement without needing two-way communication, parallel chat allows a way for everyone to ask questions at the same time, and with that livestreaming goes bigger than even the team-based approach. In our recent attempts at teams, even when we provided an in-person team room with staff, there were very few attendees. After spending large amounts of time setting up the teams, this was a bit disappointing. Still, this doesn't discourage us overall - if people find the mass communication to work better, that's fine! We can be available for those who want something else. Thus, or latest strategy is "livestream for the masses, higher-quality teams for those who want them."

Downside: amount of organizational work

The biggest downside is the overwhelming amount of effort needed to assign and manage teams. The more work done to make good teams, the more effort needed, and it almost needs a full-time person to manage it. This means we need to re-think our registration to make it more sustainable in the long-term.

The advance assignment was doable, but handling last-minute changes to keep the teams balanced, or handling no-shows, was the worst part. In order for the team concept to work best, we needed to handle these cases, since a team of too few people, or without an exercise leader, or changing every day didn't work well.

Summary

In summary, teams allowed us to make a more interactive and engaging course than many others could online. It's similar to how tables organized themselves in groups in classrooms, but by putting more attention to the arrangement, we could ensure fewer people were left out than in-person. Yet, practical difficulties and the benefits of a livestream strategy mean that teams have a less central role than they used to. In the near future, we will put extra effort into simplifying the registration system so that they can co-exist with the large livestream courses, since they are worth it when they work.

See also


Funding

CodeRefinery is a project within the Nordic e-Infrastructure Collaboration (NeIC). NeIC is an organisational unit under NordForsk.

Privacy

Privacy policy

Follow us

Contact

support@coderefinery.org