Small Group Working

This is mainly about using smaller groups within the context of a larger class: some of this does not apply if you are blessed with a small teaching group to start with.

As ever, the rule is: task first, system follows. So what is it you actually want to do?

Use them to promote interaction within the class as a whole

Use small groups for syndicate discussions of questions or cases

Use small groups for substantial project work

Questions of group size are also relevant here.

Use small groups to promote interaction within the class as a whole

The tight deadline is paradoxical: it could be argued that it aborts the process of "forming" etc. So it does, but you don't want to develop allegiances to these small groups. but to the whole.In this case you will want a flexible structure in which the groups change all the time. This is relatively easy to achieve, but remember that every time a really small group (fewer than six, say) changes a member, the group development process starts all over again. It is important, therefore, to give them a clear task, and usually a tight deadline of 10 minutes maximum (five is preferable).

• Number individuals in the class, up to however many groups you want, and instruct all the "Ones" to go together, all the "Twos" together, and so on. This has the great advantage of breaking up cliques of students who habitually sit next to one another. However, it will also call for a lot of moving about the room, so pay attention to the physical setting. Or: 
• Start with small groups, of say four people, for the first stage of an exercise. For the next stage, one member from each group moves on to the next group, to share the first group's ideas. For the third stage, another member moves on, and so on. Or:
• Start with pairs, and then "snowball" them. Having decided in pairs on their answer to a question, for example, each pair meets up with another to form a four, and to share answers. They may reach a consensus and then meet up with another four. That is about as far as you can take it.

Remember that whatever method you use, you have to do something with the results. If they are not fed back, participants are likely to feel that they have simply been manipulated. See Reporting Back.

Use small groups for syndicate discussions of questions or cases

This is a frequent pattern on training courses, and here there are more complex questions:

• If you have members with different backgrounds, do you want to keep them together or to mix them up?
  • If you want to be able to compare answers from different perspectives in the plenary session, keep them together. The group session is shorter, the reporting back and plenary discussion longer  
    • This is the pattern to use if there is a power differential between the groups represented. There may even be occasions when it is more productive to have separate male and female, or black and white, groups. Since this is a sensitive issue, it is better to consult the group about such divisions, or at least to explain the logic of them. (And to be confident that you can handle the "matters arising".)
  • If you want them to explore differences face-to-face, mix them up: leave them for longer in the groups, but expect to get parallel rather than complementary reporting-back.
• If you have a "cliquey" group, then:
  • Keeping clique members together is the simplest option (allow them to form their own groups), but will probably result in more "off-task" activity.
  • Choosing groups and allocating participants to them may result in more time being spent on-task, but probably less commitment to that task. It could of course also lead to time being spent on group formation issues.
• If your group is more homogeneous, consider whether you want to promote sub-groupings for mutual help and support. If you do, be consistent with the syndicate groups: if not, change them about.  
• Nothing contributes more to the establishment of a certain basic group identity than gentle competition, but be careful how you use that.

Use small groups for substantial project work

The decisions about groupings assume more importance when they are going to work together on something substantial, possibly over almost a whole semester, and perhaps with an assessment linked to the final product. See group-based assessment and problem-based learning.

Apart from the purely social dimension—and this is clearly important in longer-term groups—the major question is around ability levels.

• Should you group students in mixed-ability groups, in the hope that the more competent members will help the less able?
• Or should you "stream" them, so that you can focus your support more effectively on the groups which need it most?

In part the answer depends on the nature of the task, and the roles required within the group. There is a danger that in a mixed-ability group, for example, with clearly-differentiated roles, the least able member will become merely the "gopher".

There is however substantial evidence that helping others is a very effective way to learn, so less-able students can make an important contribution to the learning of their more-able colleagues, as well as benefiting themselves. The limiting factor is the social climate of the group: a group which scapegoats its less-able members is clearly not a good idea. Fortunately group roles cover "maintenance" or social needs of the group as well as task-related ones, and it is often possible for a less-able student to find a niche on that basis.

Power imbalances do require attention, however. Women students do not often do as well as they might in a male-dominated group: men can sustain being in a much smaller minority. Ethnicity and religion may also be factors, and there may be problems in balancing a desire to provide—quite literally—equal opportunities, with a recognition of the reality of cultural conflicts and tensions. This is not a situation in which to be dogmatic.

↑ Top