Group work: What is the best way to form the groups?

There are three basic methods of deciding who goes into which group:

  • Friendship - allowing students to self-select
  • Random - alphabetical, or truly random for instance by using the Moodle auto-groups feature
  • Tutor assignment - based on perceived strengths or weaknesses, for instance, or shared interests, or subject options

Each of these approaches has its advantages and disadvantages.


Although this may be the easiest option for the tutor, think carefully before issuing the instruction “come back next week in a group of 6”. For one thing, you may be reinforcing the formation of groups which don’t have enough diversity to be effective in terms of experience and competence. Self-selection doesn’t have to be based on friendship, or acquaintance, but that’s what will happen in general if you leave people to form their own groups. Colorado State University provides some useful guidance for students if they are able to choose their own groups, based on ‘when they are available to meet, individual strengths and weaknesses, diversity and commitment’ . This is a useful suggestion, but we know that left to themselves, people will tend to select other people who are like them. According to (Hinds, P. et al. 2000) “people strive for predictability” when choosing work group members. This is not only based on similarities, but also on

“a reputation for being competent and hard working, and … [people] with whom they have developed strong working relationships in the past.” (Hinds, P. et al. 2000)

First years in particular may not yet know enough people on the programme or their reputation for competence to make a group up easily, so it may be difficult for them to find a group to join. Being the last person to be selected for a team because you are different is something which many people will have hoped to leave behind at primary school.

It’s also possible that people in friendship groups may struggle to be honest with each other because their studying/working environment is too close to their personal environment.

There are obvious situations where students might benefit from self-selecting – final year options where there is a shared interest in a topic, where you have given them a thorough induction into the risks and benefits of self-selection, or where you set it up with a structured process of self-evaluation and comparing strengths and weaknesses.

In summary, if you would like to give students an experience of working collaboratively and developing skills of performance evaluation and feedback, then you may be providing them with more opportunities if you don’t let them self-select. You will also be supporting the institution’s commitments to encouraging diversity and inclusivity.


Randomly Allocated

You can use tools such as the Moodle auto-groups tool to form genuinely random groups.

To do this, click on ‘groups’ in the right hand menu, then on ‘auto create groups’ on the groups tab. See also the helpsheet on creating and managing groups in Moodle (Moodle login required).

Of course you can also use the class list - but remember that picking in alphabetical blocks is not truly random as you are likely to get six Smiths in one group and six O’Reillys in another – they might be sisters, or at the least they are likely to share many of the cultural similarities which you’ve hoped to avoid by doing random allocations. If you do use the class list, sort it via student ID rather than name to get the best chance of mixing names up.

Members of randomly allocated groups can have difficulties getting to know each other and getting down to the task. However, this can be alleviated by providing clear induction and some structured introductory activities.


Tutor Assignment

If you know a group well – let’s say a group of final year students on an option – you may feel confident in allocating them to groups based on your perceptions of their strengths and skills. This may work well, but remember that we are as prone as anyone else to selecting people who resemble us (perhaps even more so, as our peer working environment is not very diverse) – so make the criteria clear and open.

It’s more likely that you would use this approach to put all students taking a particular combination of options together, or using declared information which is relevant to the task. For instance, if you were looking at cultural businesses you might want one group of students to visit a museum, one to go to a football club, one to a club venue etc, you might choose students who have already declared interests in those areas. Or you might decide that it would be best to send people who have no interest and previous experience in those areas to get a fresh look.

If you use this approach, the chances are that they will need the same initial support as the members of randomly allocated groups: clear induction and structured introductory activities.

next: Induction »