August 03, 2005

Focus groups

(Notes / scribblings / thoughts)

Choice of methods
I chose to use a focus group for my research because the topic was particularly suited to interaction in a group context. It enabled me to gain insight from different perspectives and allowed flexibility to pursue topics which arose through previous discussion. Guiding, stimulating and facilitating the discussion was crucial to the focus group. Providing interesting material for research and ensuring all participants were sufficiently as ease to disclose occasionally quite private info was all crucial. The d. guide for the focus group was devised in a logical order, so that 'warm up' questions were placed at the beginning and built up to the more important and complex questions towards the end (d. guide in appendix).

Before the discussion I outlined the broad research aims, reiterated my position on confidentiality and anonymity, and checked that it was okay to record the discussion. I also encouraged them to express their views as freely as possible by stressing that there are 'no right or wrong answers'.

Practical considerations
Focus groups typically involve six to ten people. They need to be small enough to allow everyone to participate, but large enough to capture a variety of perspectives and enable people to bounce ideas off each other. Running focus groups requires the researcher to:

  • facilitate interaction
  • enable space of different group members to make their views known
  • keep the group discussion focused around the core themes (but still allow flexibility)
  • deal with dominant voices
  • sustain a pace of discussion that means core topics are covered but without constraining or rushing the talk (Tonkiss 2004: 204)

Focus groups may be used to supplement observation methods (as in my case where I used observation first to inform the focus group topic guide) – focus groups allow researchers to elicit info and explore attitudes not easily accessible through observation alone. Not all issues are always and easliy observable 'in the field', and focus groups allow members to define these issues in terms of their own understandings and concerns, producing shared and contested meanings:

Focus groups are an artificial intervention into a 'natural' observation setting, involving the researcher in a directive relation with their research subjects and with the process of data production (Tonkiss 2004: 197).

Methodological assumptions and issues
Key feature of focus group research is its interactive quality. Unit of analysis = group, not individuals. Focus groups are useful to theoretical research that seeks to explore social / cultural meanings, knowledges and discourses.

Underlying methodological assumption is that opinions, attitudes and accounts are socially produced and shaped through interaction with others (i.e. social constructivism?). Group context is important for exploring the way in which social and cultural knowledge and meanings are produced – we can see how people articulate and justify ideas in relation to others – emphasis is on social interaction and collective meanings.

Focus groups in this sense are not simply a means of interviewing several people at the same time; rather, they are concerned to explore the formation and negotiation of accounts within a group context, how people define, discuss and contest issues through social interaction (Tonkiss 2004: 194).
Focus groups capture the inherently interactive and communicative nature of social action and social meanings, in ways that are inaccessible to research methods that take the individual as their basic unit of analysis (ibid: 198).

Focus groups may be empowering, for 'if a group works well, trust develops and the group may explore solutions to a particular problem as a unit, rather than as individuals' (Gibbs 1997). However this benefit may not extend to all members, e.g. shy or inarticulate members. They may also become a 'forum for change' and means of achieving emancipation (ibid).

One methodological disadvantage is that researchers have less control over the data that emerge – a potential gain to participants (can be empowering), but a loss of power on the part of the researcher (issue to flag up in power/ethics section?). Another problem is that focus groups are an insecure basis for generalisation. Any claims I make in term of developing insights into social attitudes cannot be matched by ststematic claims about the representativeness of attitudes. Furthermore, although focus groups aim to reproduce the interactive aspect of naturally occurring social processes, they are not in themselves naturally occuring interactions. They offer no guarantee as to what people say, or how they interact, outside the research context.

Sampling – purposive sampling (participants selected on basis of having a significant relation to the research topic) – but known to each other beforehand as recruited from a particular training group. This may create problems with established relations of power, disagreement or consensus being brought into research setting (Tonkiss 2004: 201–2). Assertive voices are more likely to direct the group discussion (as I found). Furthermore, 'Familiarity can limit self-disclosure and discourage disagreement' (Litosseliti 2003: 54).

Can familiarity be seen as an advantage though? I would argue yes, it is a strength. The men had very varied responses to my questions and their friendship and familiarity allowed their differences in opinions and interpretations to be discussed openly as they knew each other sufficiently to show disagreement and to disclose private information. The fact that they had common experiences also meant that as a group they were able to give specific examples of events or people that they all knew about. I would suggest that the friendship dynmatics of the groups (many of the group saw each other socially after the course had ended), the subject of the research (their experiences of unemployment), a setting for the discussion with which they were all familiar by this stage (a room at the university of Leeds where they had already spent 4 days as part of the course), and the fact that I was known to them and had already built a friendly rapport with them – that all these factors helped the focus group to be successful as the participants felt at ease 'in the familiar' and were keen, lively and open in their interaction with me and the other participants.

Although they shared various commonalities (their age, gender and the fact they were long-term unemployed) their backgrounds were by no means homongenous. They had access to varying amounts of cultural, social and educational capital, and this affected their interpretations of their situations and meant that they often openly challenged other participants' interpretations. They had access to different discourses e.g. one member of the group was familiar and well-versed in pro-feminist discourses, and challenged other participants about their understandings and interpretations of women and work:

If mutliple understandings and meanings are revealed by participants, multiple explanations of their behaviour and attitudes will be more readily articulated (Gibbs 1997).


Gibbs, A. 1997. Social Research Update (19). Department of Sociology: University of Surrey. Available online at: link Accessed 03/06/05

Litosseliti, L. 2003. Using FocusGroups in Research. London and New York: Continuum.

Tonkiss, F. 2004. 'Using Focus Groups'. In Seale, C. (ed.). 2004. Reasearching Society and Culture (2nd ed.). London: Sage.

- No comments Not publicly viewable

Add a comment

You are not allowed to comment on this entry as it has restricted commenting permissions.

August 2005

Mo Tu We Th Fr Sa Su
Jul |  Today  | Sep
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31            

Search this blog

Blog archive


Most recent comments

  • Very nice, people must visit Bosnia more and more, espacially the part Hercegovina and and North Bos… by Me on this entry
  • That's not AT ALL like the Shepherdess I know. Mine is rather dark and has a foul mouth on her. Your… by Byron Getonefree on this entry
  • please send me if you have the panther picture but with a beater pixel for example 1024×786 or 1280×… by blondu on this entry
  • Well…? I take it all went ok considering your status on Facebook? (Not that I thought anything wou… by on this entry
  • If you liked Blood Meriden then you might also like this No Country for Old Men – I wrote a review a… by Neil Davis on this entry


Not signed in
Sign in

Powered by BlogBuilder