Product Information:-

  • Journals
  • Books
  • Case Studies
  • Regional information
Request a service from our experts.
Visit the JDAL journal page.

How to... conduct a focus group

Options:     Print Version - How to... conduct a focus group, part 3 Print view

Design issues of focus groups

The focus group within the overall research design

Whether you are using just focus groups as a research method will depend on whether the technique can answer all your research questions by itself, and whether it alone is an effective way of reaching your target population. It is in any case highly likely that you will want to triangulate it with another method in order to increase reliability.

Which other method you choose will depend on the type of data you are trying to collect. You may for example wish to explore some issues in further depth through individual interviews, or you may wish to combine the focus groups with a survey (see our accompanying How to.. design a survey guide), for which the former act as a method of exploring the issues.

In "Consumer acceptance of online banking: an extension of the technology acceptance model" (Internet Research, Vol. 14 No. 3), Tero Pikkarainen et al. use focus groups to test a questionnaire, and to verify the hypothesis that affective factors were relevant to online banking acceptance.

"Probing user perceptions of service quality: using focus groups to enhance quantitative surveys" (Gwyneth H. Crowley and Charles L. Gilreath, Performance Measurement and Metric, Vol. 3 No. 2) describes the use of this technique to gain greater depth of data following a survey.

One focus group is difficult to generalize from, and you should have at least two, and more if you want to examine the views of more than one target population, say for example employed and unemployed, or people at different levels in an organization. Generally, fewer groups are necessary for more structured, exploratory work, and more for unstructured work.

In "Application of the means-end value hierarchy model to understanding logistics service value" (International Journal of Physical Distribution & Logistics Management, Vol. 27 No. 9/10), Mentzer et al. explore the value of logistics in terms of enhancing value perceptions to customers. They held 13 different focus groups each featuring a different product but covering the same broad areas. The moderator introduced a few general topics to get the conversation started, but afterwards let the discussion roam freely as this unstructured technique was felt to be most appropriate to gather underlying themes.

Sampling considerations, group size and composition

For recruitment, convenience sampling (see our companion "How to... design a research study" guide), whereby participants are recruited based on their easy accessibility to the researcher, will be cheaper than probability sampling, although you should consider whether or not this will provide sufficient rigour for your design. Snowball sampling, whereby participants recruit others, may also be possible, and purposive sampling may also be appropriate.

See "Community policing and social capital" (Nathan W. Pino, Policing: An International Journal of Police Strategies & Management, Vol. 24 No. 2) for an account of how he recruited participants from neighbourhood groups as these were more likely to be sufficiently informed.

Care should be taken over the wording of the invitation to attend, particularly if inviting members of an organization as the invitation can seem like an instruction.

Estimates of desirable group size vary between four to 12 and six to eight. It's a good idea to invite more people that your target number in case some don't turn up. A small group may be biassed by existing relationships among participants and silent members, whereas a large group may be difficult to control and splinter into sub-groups. Less than four people will lose the character of a group. The size will vary according to a number of factors:

  • sensitivity of the issue – for a sensitive issue a small group is better
  • the breadth/depth of data required – if you want breadth, go for large, if you want depth, go for small
  • the target population – if this is highly articulate, it will probably be better to go for small
  • the structure of the focus group – if the group is to undertake relatively structured work say in a workshop, then it's better to have a larger group.

In determining group composition, it's important to maintain a balance between homogeneity and heterogeneity. In studying disadvantaged groups, reticence may be lessened by ethnic and linguistic homogeneity, and it is generally better to avoid large differences in status and power. For example, when researching issues in an organization it is best to have different focus groups for different levels, and to select a horizontal slice through the organization. Homogeneity is particularly important when studying sensitive topics – for example in a study of people's attitudes towards threatened redundancy, you should not include people who don't feel threatened.

See "Focus group exploration of firm-employee relationship strength" (Carmel Herington et al., Qualitative Market Research: An International Journal, Vol. 8 No. 3) for a discussion of the use of focus groups and how the authors justified size and composition.

Group members should all have a degree of commonality in terms of how they relate to the topic, for example if researching reactions to a service, all group members should be users. On the other hand, too much homogeneity can impede discussion as agreement may make people fail to articulate views sufficiently clearly.

It is also important to avoid tokenism, for example just one male, one member of an ethnic group. If the group have a demographic feature in common, then it is better that the leader also shares the feature, e.g. is of the same ethnic origin as the group, or, in the case of a group studying issues of female managers, the leader should also be female.

It may be desirable in some instances that the group are strangers to one another, as this may make discussion easier. If you find that some people know one another, then get them to sit apart. On other occasions, for example if you are investigating issues within an organization, members of the group will already know one another.

Time, place and other practicalities

In setting the time for the focus groups, you need to consider potential clashes with major events, for example, sporting fixtures or major holidays, as well as the needs of your target population. (If researching in an organization, there may be particular days of the week which are allocated to meetings and it is best to avoid these.) For example, mothers with school age children are more likely to come to an event during school hours, busy managers more likely to show up if the event is organized around lunch time, especially if lunch is provided (see below). If planning more than one focus group, have these at different times to accommodate people's schedules.

The venue also needs to be appropriate to the participants. If using a pre-existing group, then there is some advantage to meeting on familiar territory, for example a church hall, or meeting room of a place of work (although in the latter case, the setting should be as neutral as possible). If there are no such places available, then you will have to give some thought to hire of a room in, say, a pub or local hotel – in which case, you need to make sure that you have the funds in your budget.

The room needs to be of adequate size and quiet, so that group members are not distracted. Pay attention to the arrangement of the seating – are the chairs arranged in such a way as to facilitate discussion? If you are planning to have the groups back to back, you need to have a second room for the next lot of participants to wait in.

It is quite common to offer some sort of incentive to attend the focus group, such as refreshments (which should be offered anyway). For example, publishers of school textbooks offer vouchers for educational material.

"Using focus groups to assess student needs" (Britt Anna Fagerheim and Sandra J. Weingart, Library Review, Vol. 54 No. 9) describes how the authors went about recruiting students to their focus group, and how they organized the practical issues.

Focus groups online

Needless to say, once you remove focus groups from the conventional physical environment there needs to be considerable differences in the design, and despite the growing sophistication of Internet technologies, focus groups in a virtual environment are still a relative novelty. As with any form of virtual communication, the loss of non verbal clues creates a challenge, with the use of probes being particularly problematic as silence has different connotations.

Advantages are that people do not have to travel to a particular location, and those who are shy may have a greater degree of confidence. There is the very real plus that data is automatically recorded.

Care needs to be taken to create a genuine "virtual space", which should have its own design, make people feel at home and be dedicated, so that it can be free of other distractions.

The following three articles, though none of them very up-to-date, discuss virtual focus groups.

"Focus groups in cyberspace: using the Internet for qualitative research" (Henrietta O'Connor and Clare Madge, Qualitative Market Research: An International Journal, Vol. 6 No. 2)

"Designing and conducting virtual focus groups" (Casey Sweet, Qualitative Market Research: An International Journal, Vol. 4 No. 3)

"On-line focus groups: conceptual issues and a research tool" (Mitzi Montoya-Weiss et al., European Journal of Marketing, Vol. 32 No. 7/8)

Recording

An important consideration is, how will you record the data? There are various ways:

  • Relying on memory.
  • Taking notes, in which case you will need a second facilitator who sits next to you, as you will be too tied up with moderating the group to take notes as well.
  • Tape recording. This is probably the best way, but you should remember the length of time it takes to transcribe the results. You should also have a good quality tape recorder with a multidirectional microphone placed on the table, and a good quality tape close to the recorder (trying to transcribe from a poor quality tape is a very frustrating experience). You should also ask permission before you record, and check that the recorder is functioning before you start.
  • Video recording. The advantage here is that you can record body language as well as verbal data, but it's very intrusive and you will definitely need permission.

Analysis

The level of sophistication of analysis may vary – you may feel that a full-scale content analysis (which we shall deal with as a separate topic) is called for, or it may be sufficient to do a more "informal" analysis of the themes.

Whichever method you choose, it is advisable to listen to the tape recording or look at your notes, and analyse the main themes and sub-themes, as well as the participants' characteristics. You need to think about how you are going to substantiate your findings: what is the real evidence? Are there particular characteristics of the participants which might affect the data? Are there any particularly strong statements? Does a consensus emerge?

Having done this, you can then summarize in a series of statements expressing the main themes.

Most of your data will be in the form of text; if you use numerical analysis it should be to give an impression, for example a certain percentage felt... It is not usual to subject the findings of focus group research to statistical tests, although recently analytic methods for qualitative data have become more sophisticated.

When you write your report, you should organize according to major themes, noting any significant differences between groups but not writing individually on each group. More detailed analysis should go in an Appendix.

In "Workplace health concerns: a focus group study" (Journal of Management in Medicine, Vol. 13 No. 2), Annette M. Jinks and Ruth Daniels describe how they analysed their data using content analysis and "meaning units" or words or phrases that characterized categories.

"Recruiting and retaining scarce information technology talent: a focus group study" (Diane Lockwood and Al Ansari, Industrial Management & Data Systems, Vol. 99 No. 6) provides an example of reporting the findings of focus groups.

"Using an ANN-approach for analyzing focus groups" (Marcus Schmidt, Qualitative Market Research: An International Journal, Vol. 4 No. 2) describes a quantitative approach to analysis of focus group data.