The Economics Network

Improving economics teaching and learning for over 20 years

2.2 Assigning students to seminar/tutorial groups, and organising groups in seminars/tutorials

There are wide-ranging views on whether students should be streamed according to their maths/stats ability when allocating students to teaching groups. Fundamentally the decision depends on the student body you are teaching. Your entry requirements and past experience may suggest that the students taking your maths/stats module are of a roughly similar ability. In this case it may be more appropriate to consider allocating students on the basis of other concerns around integration and team-building. In this case weaker students can be offered access to additional support outside of seminars and workshop activities.

However, typically, a key issue regarding teaching maths/stats modules is how to deal with a diverse student body in terms of maths/stats ability. In such circumstances, if groups are allocated on a basis other than ability some students may be reluctant to ask ‘simple’ questions in front of peers who they perceive to be more able. This can lead to the less confident students not making sufficient progress through the syllabus. Conversely, in such mixed sessions other more confident and able students may find the sessions ‘boring’ and therefore disengage, with similarly depressing effects on performance.

When streaming is considered there are always concerns among staff that some students will perceive that they have been labelled as weak because they are in the ‘bottom’ group. There is no denying that this may be a real issue. However, there are ways to mitigate the potential downsides, as discussed below. Essentially the problem can be met head-on by explaining the allocation of students to seminars in an open and frank discussion with the students themselves and by building an understanding of the fact that we are all good at different things. In addition it is important to remind students that the streaming may be incorrect and that they should request to change groups if they find their allocated group moving too ‘fast’ or ‘slowly’. By following these steps you can create an environment in which students themselves participate in deciding the best learning environment and as a result will not feel ‘stuck’ in a particular group.

There will also need to be an emphasis that all groups will cover the core material. Together, the emphasis that students are not ‘stuck’ in a group and that all groups will be taught the same core material should counter any negative connotations that students may have of streaming from school where these features are absent and different classes may indeed cover a varying syllabus. This approach finds support in the educational literature. Dancer and Fiebig (2004) note the usefulness of streaming to allow different styles of teaching to cater for varying maths ability, with streaming allowing staff to target at risk (of failure) groups. Similarly Alauddin and Butler (2004) note the benefit from streaming students and providing differentiated offerings, although the authors recognise that teaching to the median student in a lecture room is cheaper and that there are resource-induced tensions in the modern university.

If you decide to stream your students into groups based on a screening test you may also wish to consider sessions of different lengths. This is because although you may target a group who require additional foundational support using the test, there is still likely to be considerable diversity within this group (the greater the higher is the threshold pass mark in the screening test used for allocations). For example, you may want to allocate the more maths/stats able students to one-hour sessions and weaker students to two-hour sessions. This can help reinforce the fact that the same core material will be discussed in all sessions, albeit at different speeds across the groups. An alternative is to provide additional support sessions for students who wish to access this support. However those students who need this support - and recognise the fact - will often have other competing pressures on their time and therefore be less likely to make the time unless it is part of their scheduled activities. Accordingly, by formalising the additional support in different length sessions you can have more confidence that you will reach the students who may benefit most from it.

If streaming your students seems appropriate then the first practical consideration is what results to base this on. Assuming that the data is available then maths entry grades could be used for this purpose. However for some students these grades will have been obtained two or more years before starting university or be based on various overseas qualifications. Therefore it may be sensible to consider screening test prior to the start of the module to establish a current baseline of knowledge.

The next consideration is designing your test. A multiple choice test will allow for quick marking, as turn-around time is likely to be an issue, and you may consider on-line testing to avoiding marking altogether. It is perfectly acceptable to use a multiple-choice test in this setting as long as it is well designed. There is a range of guidance available to assist you, including that available via the Economics Network website. The main considerations are that enough answers are available as options and that these options are sensible in terms of the rational routes that students may have taken to solve the question.

It is worth bearing in mind that while on-line testing has obvious appeal there are a number of issues to be borne in mind. You will need to ensure that you can find a large enough computer room to carry out the testing. The other issue is that it takes time to become familiar with the on-line testing software available at your institution and you will need to make sure technical support is available on the day of the test. The reality is that you will also have some students arriving late and therefore a mixed economy of on-line and paper-based testing will nearly always be required. You will also want to make sure you have a paper-based back-up in case the system ‘crashes’ or some other unforeseen event occurs.

It is important to review your processes and to continue to tailor your approach to the needs of your students in the light of experience. In particular, the screening test threshold mark for allocating students to a preliminary support module or to longer workshop sessions will require review and reflection. As part of this process it is helpful to check that the test has appropriately streamed the students at the end of the maths/stats module. Checking the correlation between the screening results and final grades in the module concerned can be useful guide. Ideally streaming will have delivered added value bringing all students up to a similar level. However, in reality maths/stats learning takes more time a typical module allows to bring students to a similar level and so it is to be expected that there will remain some degree of positive correlation between screening results and end of module results.

If you decide to carry out streaming you may still involve students in group work within the streamed seminar classes. There needs to be thought given to determining the make-up of these within-class groups. Learners have different styles: activist, reflector, theorist and pragmatist. There is some evidence to suggest that it is useful to try and select students with similar learning styles into a group (Dunn et. al. 1995). Table 1 explains these types of learners. You may wish to set an exercise, or a learning style questionnaire (see, for example, Entwistle, 1981; Honey and Mumford, 1995), and then ask the students themselves to identify their learning style in order to help group formation.

Table 1 Seminar activities appropriate to each type of learner (according to ‘experiential learning’)

Activist style

Learns best from activities where:

  • there are new experiences/problems, etc.;
  • they can become engrossed in short tasks, games, competitive teamwork tasks, etc.;
  • there is excitement/drama/crisis and things chop and change with a range of diverse activities to tackle;
  • there is chance of limelight, e.g. leading discussions, giving presentations;
  • they are involved with other people, e.g. bouncing ideas off them, solving problems as part of a team.

Learns least from activities where:

  • learning involves a passive role, e.g. listening to lectures, reading, explanations;
  • they are not directly involved;
  • they are required to assimilate, analyse and interpret lots of data;
  • they are required to engage in solitary work, i.e. reading, writing, thinking on their own;
  • they are asked to repeat the same activity over and over again.

Reflector style

Learns best from activities where:

  • they are encouraged to watch/think/chew over activities;
  • they are able to listen/observe a group;
  • they can reach a decision in their own time without pressure and tight deadlines.

Learns least from activities where:

  • they are forced into the limelight;
  • they are worried by time pressures or rushed from one activity to another;
  • they are pitched into doing something without warning.

Theorist style

Learns best from activities where:

  • they are in structured situations with a clear purpose;
  • they are required to understand and participate in complex situations;
  • they have time to explore the associations and interrelationships between ideas, events and situations.

Learns least from activities where:

  • they have to participate in situations that are unstructured, where ambiguity and uncertainty are high, e.g. open-ended problems;
  • they are faced with a hotchpotch of alternative/contradictory techniques without exploring any in depth;
  • they find the subject matter platitudinous, shallow or gimmicky.

Pragmatist style

Learns best from activities where:

  • there is an obvious link between the subject matter and a problem set;
  • they are practising techniques with coaching/feedback;
  • they are given techniques that are applicable to the real world.

Learns least from activities where:

  • the learning is not related to an immediate need or relevance;
  • there is no practice or clear guidelines on how to do it;
  • they cannot see sufficient reward from the learning activity.

Source: Honey and Mumford (1995).