Action research and evaluation on line

Session 9: Focus groups



This is Session 9 of areol, action research and evaluation on line, a 14-week public course offered as a public service by Southern Cross University and the Southern Cross Institute of Action Research (SCIAR). which two forms of focus group are described, one of them as focus groups are often conducted, the other a variation which is designed to increase both rigour and involvement


Take a moment to consider this before proceeding...  Most of us spend a fair amount of our time in unstructured discussion.  How effective is it, do you think, as a way of exchanging information? Are there occasions when you could improve it by introducing some structure into it? What types of structures do you think might be successful for what purposes?


In this session


Last session we talked about a particular interviewing technique: convergent interviewing.  This session examines a more involving activity which collects information in a group setting: the focus group.

A focus group is a "focussed group interview".  Compared to individual interviews, it more easily provides opportunities for participant involvement and interaction.

(Focus groups are also briefly discussed in one of the archived case studies.   )

I begin by describing, briefly, a typical focus group.  I then discuss an alternative version.  It is used to illustrate some general principles about process management and data interpretation. 


Focus groups

As I said, focus groups are group interviews.  The researcher frames and asks the questions.  The informants respond to that question and to each others' responses.

Focus groups are commonly used for marketing research.  They appear to be spreading to other fields of qualitative research, and to be more common than they were even a few years ago.  (Or at least re-spreading; qualitative research was their original application, I think.)  There is a growing literature specifically on focus groups.

As often conducted, the group consists of 8 to 12 people, perhaps a few more or less.  The participants (the "panel") may be chosen to be fairly similar, being drawn from a particular category of informant.  One or two hours is a typical duration.

Food to nibble, and drinks, are often provided.  Participants are often paid or offered a gift of some sort for taking part.  The venue is comfortable.  The participants are treated well.

(The inducements are intended to attract informants into taking part.  In work settings material rewards may be less necessary.  It is often enough to conduct the focus group in working time.)

A skilled facilitator asks the questions.  Questions are usually framed to relate to one concept -- a planned new product; a particular community issue; a problem...  The information collected is usually about the informants' attitude to that concept or idea or product.

The facilitator then keeps the conversation flowing.  As you can imagine, the quality of information depends a lot on the facilitator's skills.

The entire group session is most often recorded.  Video is common.  Audio can be used instead.  Whether video or audio the transcript or tape of the session can be analysed in depth. 


An alternative focus group

My intention in the rest of this session is to apply some of the ideas from earlier sessions.  Focus groups will be used as a vehicle for doing this.

If you do not know much about focus groups this session will introduce you to a form of focus group which is designed as action research.  Along the way we'll also summarise or expand on some earlier material.


Before you read on, you may wish to consider how you would structure focus groups to improve the quality of both involvement and data.  You might start by considering these questions:

How could you involve informants more (for example in interpreting the information)?

How could you improve the data (for example by building in dialectical processes)?


The description which follows is one way of conducting a focus group as part of an action research project.  Here in overview is a possible structure you might use:

1   Preparation.   Choose the panel members and approach them; plan the process; select the venue.

2   Set the scene.   Welcome people; explain the purpose and process; invite people to introduce themselves.

3   Context.   Facilitate a brief initial discussion which is slightly broader than the issue or concept that interests you.  For example, if the topic is a particular form of training within an organisation you might begin by discussing training in general.

This gives you a chance to warm people up to the important discussion to follow.  It also provides you with an opportunity to learn how the group interacts and plan your facilitation accordingly.

4   The discussion.   This is the heart of the process, and can contain a number of steps:

  • explain the concept, and what you want from the discussion
  • allow people a few minutes to think about their response to the concept or issue
  • an initial open discussion follows; during this, encourage people to identify themes and note them down for the next part ofthe discussion
  • ask people to report what themes they identified; capture these on butcher paper or electronic whiteboard (electronic whiteboards are great for this purpose!)
  • then facilitate a discussion on the relative importance and meaning of the themes;  capture the key aspects of this on butcher paper or electronic whiteboard.

In other words, use a general approach which begins with initial thinking.  Follow that up with a discussion.  Then ask people to extract themes from the discussion.  Finally, discuss the meaning and significance of the themes.

If people are slow to warm up, it is useful to follow the initial thought with discussion in pairs.  This enables people to "try out their words" in relative private before they have to express them publicly.  (I suggest you give them a few minutes to get acquainted first.)

A more detailed description of this style of focus group can be found in the archived file focus (see below).

The archived file presents what you might call a mid-range description -- not too detailed, not too brief.  Here are some issues that warrant more detailed attention.  These also illustrate some general principles about information collection, especially when it is done in a group setting. 


Selecting the panel

For the usual form of focus group the panel is typically not very diverse.  A homogeneous group allows easier discussion.  People are more likely to build on one another's ideas without undue conflict.

Because of the more controlled processes in a structured focus group, greater within-group variety can be managed.  A structured focus group also allows more variation in numbers of participants.

You may be able to create a climate where differences are explored with interest rather than confrontation.  If so, even greater variety may be useful.  (I like to use highly-diverse groups, such as those you might put together as a maximum-diversity sample.  At the same time, I give a lot of attention to creating a climate where people feel able to disagree without having to arrive at a conclusion or win others over to their point of view.)

If well facilitated, diverse panels can offer a number of potential advantages.  Here are two in particular:

  • You can cover a greater range of the views within the wider community or organisation.
  • More diversity means more differences of opinion.  In turn, that can allow deeper interpretation.  It is often the exploration of differences that can lead the participants (and the facilitators) to a better understanding.  But of course this requires more skilled facilitation. 


Setting the scene

You can have more influence over the process if you have established the beginnings of a person-to-person relationship with each of the panel members.  (This is true of action research generally, not just focus groups.)

For this reason I try to make contact with each participant just before the session begins.

To this end, I often ask participants to arrive a little before the planned start.  Coffee, fruit juice and biscuits are available.  I use this time to talk briefly with people as they arrive.  I also memorise their names.  (I'm poor at remembering names.  I have to make a special effort to do this.  But I believe it's worth the effort.)

This initial contact also gives the panel members a chance to ask any questions they have about the process or the purpose.

Some people feel anxious until they have made a connection with at least one other person.  It is often useful, after you have introduced yourself to someone, to introduce them to someone else. 


Creating a climate

During the discussions, differences are likely to arise.  The climate of the whole focus group depends a lot on how these are dealt with.

I believe this is so important that I try to do several different things to encourage the expression of differences.  For example...

Using processes which explicitly encourage the identification of different views:

"I'll ask each of you to present your opinion, and I'll record it as you do.  For the moment, don't worry about whether it is the same as, or different to, anyone else's.

"When I've recorded everyone's opinion, I'll ask you to help me identify the agreements and disagreements.  Then we can try to work out what those disagreements mean."

Giving instructions which encourage mutual inquiry:

"I encourage you to speak your mind.  I'd prefer that you were not too belligerent about it.  But I'd rather that you were tactless than dishonest."

Encouraging an inquiring reaction to disagreement, treating it as data:

"We don't have to reach agreement.  In fact, it's important to me to know what different views there are, and how much disagreement there is.

"You can help me by speaking honestly.  I also appreciate it if you notice disagreement and bring it to my attention.

"The existence of disagreement is important information."

Responding to disagreement with interest and enthusiasm:

"Thank you, Jane.  It seems that you and Fred have different opinions.  That's great! If we can understand how those differences arise, that will be valuable indeed."

Discouraging open confrontation in a way which still elicits open discussion:

"Good, there's another difference of opinion.  Thank you, Jack.  That's useful.

"I'd also encourage you to find some way of saying it that informs us, rather than trying to persuade us.  If you had said to me what you said to Alex, I'd want to defend my point of view.  And I don't think that's useful for present purposes."


Pursuing explanations

I think that some general principles of qualitative data use are relevant here.  I've touched on them before; they are important enough to warrant some repetition.

Idiosyncratic information

When only one person from a large community or organisation tells you something, it may not be very informative.  In general, then, I suggest you largely ignore idiosyncratic information if the other participants also ignore it.

(If that person is a key stakeholder you may choose not to ignore it.  On balance, though, I think you will learn more from pursuing the agreements and disagreements.)


When several people agree, I try to find out how widely that applies.  So I look for exceptions.  In effect, I am trying to find the boundary of the agreement.  (There is an example given in the next section.)

This is also part of the strategy of challenging your emerging data and interpretations.  By the end of the study, it's good to be able to say that your interpretations survived many vigorous attempts to disconfirm them.


Agreements are useful.  But for me, disagreements are the key.  They lead me to a deeper understanding.  They can often be the source of new and creative ways of thinking about what is happening.  They also help me to identify my own assumptions.

Sometimes disagreements emerge spontaneously.  Person A says "X is good"; person B says "X is bad".  I can then explore why that is so.  I can also, in a process akin to dispute resolution, help A and B to explore why.

On other occasions disagreements emerge because I go searching for them.  Suppose that there seems to be agreement that "X is good", for example.  I can ask for exceptions.

"When isn't X good?" "Who doesn't think X is good?" "What's not good about X?"

And so on.  When an exception is identified, I can then try to explain it.

A caution...  "Why" questions are tricky.  Asked "Why?", most people try to give an explanation.  The difficulty is that, lacking an explanation, they may invent one.  If you and they then believe it, you may lead each other on a fruitless detour.

I usually prefer to find some other way of asking why.  For example...  "What are the important differences between those who believe X and those who believe Y?" Or "How do those differences arise?"


Introducing other information

You may often use a number of focus groups to explore a single topic or issue.  The later groups then give you a chance to challenge the data and interpretations emerging from earlier groups.

You may combine focus groups with some other form of data collection.  This is most commonly done for the purposes of triangulation: when two different techniques yield the same conclusion, you can have more faith in that conclusion.

In both instances it is often useful to introduce information and interpretations collected earlier.

When do you introduce new information? This is an important question, as it can have an influence on how the information is treated.  I prefer to start the current focus group as described above.  After they have provided their own themes, I then summarise for them the findings from earlier groups.

Alternatively, you can complete the interpretation phase, and then present other information.  The panel can then be asked to respond to the other information, and explain the differences.

Again, you can give particular attention to agreements and disagreements.  You can again seek explanations for the disagreements. 



  1. See case study 1 in the archive.  The URLs are
    [ back ]


Archived resources

There is a more detailed description of structured focus groups archived as focus on the arlist archive.  The URLs are:

An important part of facilitation is modelling an open and non-defensive style of interaction.  The archived file communicn, mentioned in an earlier session, is relevant here.  The URLs are:

I've already mentioned (in the previous session) datadriv, an account of the data-driven procedure of seeking exceptions to agreements, and explanations for disagreements: 


Other reading

Sage (who are probably the leading US publishers of qualitative research resources) publish three useful books on focus groups:

Krueger, R.A.  (1988) Focus groups: a practical guide for applied research.  Newbury Park: Sage.

Morgan, D.L.  (1988) Focus groups as qualitative research.  Newbury Park: Sage.

Stewart, D.W.  and Shamdasani, P.N.  (1990) Focus groups: theory and practice.  Newbury Park: Sage.

and, so Sage don't have the field entirely to themselves:

Greenbaum, Thomas L.  (1988) The practical handbook and guide to focus group research.  Lexington, Mass.: Lexington Books.

Edmunds, Holly (1999) The focus group research handbook. Chicago, Ill.: NTC Business Books in conjunction with the American Marketing Association.



A thought experiment

For the next few days, make a note of occasions when you find yourself disagreeing with people.  When you have some quiet time, reflect on their beliefs, and yours.  What can you learn about the assumptions that underpin your beliefs? How can you explain why you and others sometimes think differently? How can you engage more constructively with people when you and they disagree?

An individual activity

This can usefully follow the individual activity above.  For the next few days, explore ways of turning potential conflicts into dialectical discussions:

Keep an ear open for anyone who says something you disagree with.  When this occurs, listen to what they have to say before you respond.  Use a non-confronting form of inquiry to learn more about their views.  Then find a way of stating your view so that you engage them in mutual inquiry.

From time to time, disputes may erupt between other people.  Watch for these.  When you notice one, experiment with ways of using your own curiosity and non-defensiveness to defuse their conflict.  Note: some ways of defusing conflict do so by cutting off the debate.  The aim here is to convert the debate into something non-adversarial.

For your learning group

Choose a facilitator.  Choose a topic (perhaps one that is relevant to one of your projects).  The facilitator then conducts a structured focus group on that topic.

Build into the process a number of review periods.  After the initial discussion might be one such place.  After deciding priorities and meanings might be another.


Let's practice action research on areol.  What ideas do you have for improving this session? What didn't you understand? What examples and resources can you provide from your own experience? What else? If you are subscribed to the email version, send your comments to the discussion list. Otherwise, send them to Bob Dick


In summary

I've described a focus group which uses structure to enhance the quality of information.  The discussion has also provided a vehicle for talking about some of the principles of process management.

Next session we'll examine some other data collection techniques.  See you then.  -- Bob



Copyright © Bob Dick 2002.  May be copied provided it is not included in material sold at a profit, and this and the following notice are shown.

This document may be cited as follows:

Dick, B.  (2002) Focus groups Session 9 of Areol - action research and evaluation on line.





Maintained by Bob Dick; this version 11.04w; last revised 20020712

A text version of this file is available at