There has been much debate about whether group work is an effective learning strategy. Harry Webb (@webofsubstance) wrote an excellent post about it recently. For me, the main consideration when teachers are deciding to use group work is whether working in a group will mean that the children will achieve more than if they were working individually.
This was brought home to me in a Y6 class I was in. The children were building Mindstorm robots in readiness for programming them in groups. It was a complicated building process. The group that made the most progress with the robots was the group where different individuals within the group had specific tasks that were then put together at the end to finish the group task effectively. The groups that made the least progress were those groups where the members behaved as individuals just doing their own thing. The result of this was that they achieved no more than if one pupil had tried to build the robots on their own.
If the group work task is set up in a way that this doesn't happen, why bother? Of course there are the possible social benefits of allowing children to work together i na group. However, I would argue that time spent on this should be minimal and probably done at the start of the school year when relationships between the children need to be renewed.
In Harry Webb's post, he suggests that group work is only effective if there is individual accountability. I couldn't agree more. If not, children can either dominate (hogs) or coast (logs).
A technique that I was introduced to by a fantastic headteacher I have worked with on numerous occasions (Kathryn Coiffait) is 'Envoying'. I believe that enjoying meets both of the considerations mentioned above. Please bear with me and I'll explain one example of how this technique could work in a primary classroom.
First of all the children are organised into groups of mixed ability. The groups don't need to be the same size which allows for flexibility in managing potential disputes between children that don't get along. Then the teacher introduces the task, say for example, writing a multi-sensory story setting (an Alan Peat technique). Each group is given the task of collecting words/phrases for a given sense e.g. one group would be doing smell, another sounds etc. Obviously, an engaging stimulus will be given for the children to draw on. Before the children start to work in their groups, the teacher says that one of them is going to be an envoy. As the children don't know who it is going to be at this stage, they know they must all make sure they have the full range of words/phrases collected by the group therefore they cannot coast or dominate. This should take about 10 minutes.
After a period of time, the teacher selects an envoy from each group. This envoy then visits the other groups sharing his or her group's words/phrases and at the same time collecting words/phrases from the group he/she has visited. The members of the group being visited by the envoy have to record for themselves the information the envoy is providing. This is repeated until all the envoys have visited all the groups. This part of the process should only take about 10 minutes.
Finally the envoys return to their groups with a wealth of information for themselves and all the other group members have a wealth of information from the other envoys. The children can then move to individual work to complete their setting descriptions. For this to work, the dividing up of the task for each group has to be thought out carefully. I have also seen it work well in maths where a class were asked to investigate properties of 3D shapes. Each group was given a couple of shapes to investigate in detail. The teacher then used enjoying as a quick way of transferring each group's learning to all of the children in the class.