Analysing group data
The principles, processes and outputs described above relate to all forms of qualitative analysis concerned with interpreting meaning, irrespective of the type of data collection methods used. But there are some additional features that need to be considered when data have been collected through group discussions rather than individual interviews. We briefly consider these below.
The nature of group data
There are a number of ways in which group data differ from individual interview data but the following have particular significance for analysis:
• Group dynamics. A dynamic will occur within each group that will affect the way in which the subject is discussed. This will partly be manifest through what is said, or how it is delivered, but there will also be many non-verbal communications. The latter is additional evidence that the researcher will need to make explicit so that it is captured on tape or noted immediately after the discussion
• Interactions. There will be interactions between group members that may take the form of affirmations, disagreements, conflicts, or simply continuation of a previous contribution from another member. These are part of the 'data' and the way in which such interactions take place are a useful source of information. But they will also result in some incomplete or fractured elements in the discussion.
• Uneven coverage. The level and coverage of data available for each participant is likely to be uneven. This will occur within one discussion where each member of the group will speak at varying lengths on different topics, depending on the salience of that subject to them; and across different groups because each group will shape a slightly different agenda
depending on the contributions of individual members and the dynamic between them.
• Less extensive coverage. Whatever the level of contribution, there will be less coverage/depth of information for each participant than in an individual interview because time has to be 'shared' between the different discussants.
• The influence of other views. Group members have the opportunity to hear different or opposing views or other ways of expressing their own arguments.
As a consequence, they will modify, refine or extend what they say in the light of the other contributions. This process needs to be investigated in the group forum so that the ways in which views develop can be traced in analysis. It is for this reason that each contribution needs to be attributed to their originator in the verbatim transcript, although obviously in anonymised form .
Approaches to group analysis
There are two main ways in which group data can be analysed, the first of which is most commonly practised:
• Whole group analysis which treats the data produced by a group as a whole without delineating individual contributions. The group therefore becomes the unit of analysis and will be treated in the same way as a unit of individual data. Additional information (in the form of notes) about group interactions or the balance of individual contributions may be added to the data as part of the evidence.
• Participant based group analysis where the contributions of individual participants are separately analysed within the context of the discussion as a whole. This allows the information of each participant to be retained and for interactions between individual members to be noted as part of the recording of the group dynamic.
The advantage of participant based analysis over whole group analysis is that it allows more detailed evidence about similarities and differences between group members to be determined. It also allows certain types of analysis (such as associative analysis) to take place at an individual as well as a group level. The main disadvantage is that it may remove the immediate context in which the contribution was made, although there are ways to deal with this during the data management stage of analysis. It is also much more time-consuming than group based analysis because the contributions of eachmember have to be traced throughout the discussion. A decision about which approach to use will therefore depend in part on resources but will also be determined by the objectives of the research and the kind of analytical outputs that are required.
It is important to note that certain forms of analysis are more limited with group than with individual data. The identification of sectors and typologies has to happen at a more general level of assignment because different levels of information will be available about each individual. It cannot be carried out at all if group based analysis has been undertaken unless the groups themselves are very homogeneous in representing previously defined sectors of the population. Similarly, associative analysis is likely to be less refined than with individual data as it will have to take place either at the more global level of the group; or will be incomplete because of missing evidence.
Nevertheless, group discussions also have additional ingredients that are missing from individual data brought about through the interactions between group members. They can be extremely creative and may therefore be a rich resource for developing new strategies or generating hypotheses.
They can also help in the understanding of diversity by engaging people with different perspectives in debate and can thus have additional explanatory power. These analytic advantages and limitations have to be weighed in the context of the aims of the study when a choice about data collection methods is being made .
Source: Miles, M. B., Huberman, A. M., & Saldana, J. (2013). Qualitative data analysis. Sage.