A focus group is a moderated group discussion, typically of six to ten carefully recruited participants, in which a trained moderator uses a structured set of questions to explore how people think and feel about a defined topic. Unlike a one-to-one interview, the method deliberately harnesses the interaction between participants — their agreements, disagreements, and the reasons they give — to surface shared understandings, points of divergence, and the language a community uses to make sense of an issue. Focus group research is therefore best suited to exploratory questions where the social dynamics of an opinion matter as much as the opinion itself.
Use a focus group when you want breadth of perspective quickly, when you are scoping a topic before designing a survey, or when the meaning a group co-constructs is itself the object of study. Use individual interviews instead when the subject is sensitive, when you need depth on each person’s private experience, or when group pressure would distort what people are willing to say.
What a focus group is (and is not)
A focus group is a qualitative data-collection method in which a small group of participants discusses a topic under the guidance of a moderator who follows a planned discussion guide. The defining feature is not simply that several people are interviewed at once to save time; it is that the interaction among participants becomes part of the data. When one person voices an opinion, others react — endorsing, qualifying, or challenging it — and in doing so they reveal the range of views in the wider population, the norms that govern what can be said, and the reasoning behind positions. This is what methodologists mean when they describe the focus group as a method for studying “group effects.”
It helps to be clear about what a focus group is not. It is not a group interview in which the moderator simply asks each person the same question in turn (that wastes the method’s central advantage). It is not a consensus-building or decision-making meeting; disagreement is welcome, not a problem to be resolved. And it is not a substitute for a representative survey: with six to ten people per group you are sampling perspectives, not measuring their prevalence. If you find yourself reporting that “70% of the group agreed,” you have misunderstood the unit of analysis.
Focus groups sit within the broader family of methods of data collection and are frequently contrasted with their closest relative, the qualitative research interview. Choosing well between them is the first methodological decision you must justify.
When to use a focus group versus individual interviews
Both methods produce rich qualitative data, but they are good at different things. Focus groups excel when interaction adds value — when you want participants to spark ideas off one another, when you are exploring shared cultural meanings, or when the topic benefits from the “group as a whole” generating a richer account than any individual would alone. Individual interviews excel when you need depth on each person’s biography, when the topic is private or stigmatised, or when you need to attribute every statement confidently to a specific person.
| Consideration | Focus group | Individual interview |
|---|---|---|
| What it captures best | Shared meanings, range of views, group interaction and norms | Personal experience, individual reasoning, life history |
| Sensitive topics | Risky — social desirability and disclosure concerns | Strong — confidential, safer for personal disclosure |
| Speed / efficiency | High — many voices per session | Lower — one participant per session |
| Depth per person | Limited — airtime is shared | High — full attention on one person |
| Idea generation | Strong — participants build on each other | Weaker — no cross-stimulation |
| Main risk | Groupthink and dominant voices | Interviewer effects, no triangulation in the room |
A useful rule of thumb comes from the literature on qualitative design: choose focus groups when the collective meaning is the prize, and interviews when the individual account is. Many strong dissertations combine the two — focus groups to map the territory, then interviews to probe depth.
Group size and composition
The conventional recommendation is six to ten participants per group. Fewer than five and the discussion can feel thin and exposed; more than ten and you lose control, airtime per person shrinks, and quieter members disappear. Many applied researchers favour the smaller end — six to eight — for complex or emotive topics, because smaller groups give each person room and are easier to moderate. Always over-recruit by one or two, because no-shows are normal.
Composition is just as important as size. The key decision is whether to make groups homogeneous (members share relevant characteristics) or heterogeneous (members differ).
- Homogeneous groups — similar age, status, role, or experience — put people at ease, encourage candid talk, and let you compare cleanly across groups. This is the default for most student projects.
- Heterogeneous groups deliberately mix perspectives to provoke debate and surface contrast, but they risk power imbalances (for example, mixing managers with junior staff can silence the juniors).
The safest approach is to make each group internally homogeneous but to vary the composition between groups — a strategy called segmentation. For instance, run one group of first-year students and a separate group of final-years rather than mixing them, so any difference between year groups becomes visible in your analysis. How you decide who is eligible draws directly on your sampling strategy; purposive sampling is the norm.
How many groups, and reaching saturation
A single focus group is rarely enough — one group can be idiosyncratic, dominated by a strong personality, or simply have an off day. The widely cited guidance is to plan three to five groups per segment and to keep running groups until you reach saturation: the point at which new groups stop producing new themes and you are only hearing variations on what you already know. In practice, many projects find that thematic saturation arrives after three to four well-run groups per segment, but you should treat that as a working estimate, not a quota — decide saturation from the data, not from a number fixed in advance.
“The hallmark of focus groups is the explicit use of group interaction to produce data and insights that would be less accessible without the interaction found in a group.” (Source: Morgan, 1997)
The moderator’s role and skills
The moderator (or facilitator) makes or breaks a focus group. The role is not to lecture or to extract answers, but to create a permissive climate in which people feel safe to speak, to keep the discussion on the topic without railroading it, and to manage the group dynamic so every voice is heard. Good moderation is mostly listening.
The core skills are:
- Neutrality — not signalling approval or disapproval, so participants do not simply tell you what they think you want to hear.
- Active listening and probing — following up with “Can you say more?” or “Does everyone see it that way?” rather than moving on too quickly.
- Managing dominant and quiet members — gently redirecting a monopoliser (“Thank you — I’d love to hear from someone who sees it differently”) and inviting silent members in by name.
- Comfort with silence — a pause often draws out the most considered comment.
- Flexibility — following an unexpected but fruitful thread while still covering the guide.
A second person, the note-taker or assistant moderator, is invaluable: they capture seating, non-verbal cues, and who said what, freeing the moderator to focus on the conversation.
Designing the topic (discussion) guide
The discussion guide is your roadmap: a short, ordered set of open questions, usually five to eight, that moves from broad and easy to specific and probing. It is a guide, not a script — you should feel free to reorder, skip, or add follow-ups in the moment. A common and effective structure funnels the conversation:
- Opening question — quick, factual, gets everyone talking (“Tell us your name and one word for how you find online learning”).
- Introductory question — broad and easy, frames the topic.
- Transition questions — move the group toward the heart of the matter.
- Key questions — two to five questions that drive at your research objectives; these get the most time.
- Ending questions — “Is there anything we’ve missed?” and a summary check.
Keep questions open (“How…?”, “What…?”, “Why…?”), avoid leading wording, and ask one thing at a time. If your wider project also uses a survey, the same care that goes into a qualitative questionnaire applies here: pilot the guide on a couple of friendly volunteers and revise anything that confuses them.
Steps to plan and run a session
The procedure below takes you from recruitment to analysis. Treat it as a checklist for each group you run.
- Recruit and screen. Define eligibility, recruit purposively against your segments, over-recruit by one or two, and send a reminder the day before. Confirm informed consent and ethical approval are in place.
- Prepare. Finalise the discussion guide, book a quiet, neutral room with a round-table layout, arrange refreshments, test your audio recorder (plus a backup), and prepare name cards and consent forms.
- Warm-up. Welcome participants, explain the purpose, confirm confidentiality and recording, set ground rules (one voice at a time, no right or wrong answers), and run a quick round-the-table opener so everyone has spoken once.
- Core questions. Work through the key questions, probing for reasons and inviting contrasting views. Watch the airtime, draw in quiet members, and gently manage anyone dominating.
- Wrap-up. Summarise the main points back to the group and ask whether you have captured it fairly, then ask the final “anything we’ve missed?” question. Thank participants and explain what happens next.
- Analyse. Transcribe promptly and analyse thematically (below), attending to interaction as well as content.
Recording and transcription
Always audio-record (with consent) — you cannot moderate well and take a verbatim record at the same time. Use two recorders in case one fails, and have your assistant keep a seating plan and contemporaneous notes that mark who is speaking and capture non-verbal reactions (nods, laughter, visible disagreement) that the audio will miss. Transcribe the recording in full, ideally within a day or two while memory is fresh, labelling speakers consistently (Moderator, P1, P2…) so you can later trace turn-taking and exchanges. A focus group transcript is harder to produce than an interview transcript because voices overlap; a clear seating plan and good notes are what make accurate attribution possible.
Analysing focus group data
Most focus group data are analysed using thematic analysis — the systematic coding of transcripts to identify, organise, and interpret patterns of meaning across the data. Following Braun and Clarke’s (2006) widely used six phases, you familiarise yourself with the data, generate initial codes, search for themes, review them, define and name them, and write up. What makes focus group analysis distinctive is the obligation to attend to group dynamics, not just what was said. You should ask not only “what themes appear?” but also: How did views shift as people responded to one another? Where was there consensus, and where did the group split? Did a dominant voice steer the discussion? Treating the group’s interaction as data — coding agreements, challenges, and reversals — is what separates a genuine focus group analysis from a stack of mini-interviews analysed together.
Working it through: She budgets 75 minutes per session. With 8 participants and roughly 60 minutes of active discussion, that is 60 ÷ 8 ≈ 7.5 minutes of airtime per person — enough for everyone to contribute, which is why she caps the group at 8 rather than 12 (which would give only 5 minutes each). She drafts a 6-question guide and pilots it on two volunteers, cutting one leading question. Running the first segment, no new themes appear by the third first-year group, so she judges that segment saturated and does not add a fourth. After transcribing all six sessions she codes the data in Braun & Clarke’s six phases, generating 34 initial codes that she collapses into 4 themes. Crucially, she notes a dynamic: in two groups, one confident student dismissed exam anxiety, after which quieter members went silent — so she codes “normalising of stress / silencing” as a dynamic in its own right and reports it, rather than concluding the group simply felt no anxiety. Her final analysis reports the 4 themes and how group interaction shaped them.
Strengths and limitations
Used for the right question, focus groups are powerful; used for the wrong one, they mislead. Weigh both sides honestly in your methodology chapter.
Strengths
- Interaction generates richer, more spontaneous data than the sum of individual answers.
- Efficient — you hear many perspectives in a single session.
- Excellent for exploring shared meanings and the language a community uses.
- Participants can prompt recall and ideas in one another that an interviewer never would.
Limitations
- Groupthink: participants may converge on a socially acceptable view and suppress dissent.
- Dominant voices: one or two confident members can skew the whole discussion.
- Confidentiality: you cannot guarantee that what is said in the room stays in the room, which makes the method unsuitable for sensitive disclosures.
- Not generalisable: small, purposive samples describe perspectives, not their prevalence.
- Harder to analyse: overlapping talk and managing interaction make transcription and analysis more demanding.
Common mistakes to avoid
- Treating it as a group interview — asking each person the same question in turn and ignoring the interaction that is the point of the method.
- Making groups too large, so airtime per person collapses and quieter people drop out.
- Mixing incompatible participants (e.g. staff with their managers), creating power imbalances that silence people.
- Running only one group and over-claiming from it.
- Asking leading or double-barrelled questions that steer the group.
- Letting a dominant participant run the room without redirecting.
- Counting heads (“70% agreed”) as if the group were a survey.
- Reporting only what was said and ignoring how the group’s dynamics shaped it.
How to do it well
The best focus group studies are planned like experiments and facilitated like good conversations. Segment your groups so comparisons are meaningful, recruit purposively against clear criteria, pilot the guide, and prepare the room and equipment so nothing distracts from the talk. In the room, lead lightly: ask open questions, sit comfortably with silence, protect quieter voices, and let productive disagreement breathe. Afterwards, transcribe promptly, analyse the interaction as well as the content, and report both. Done this way, focus groups give you something no other method can — a window onto how people make sense of a topic together.
Designing focus groups for your dissertation?
Our academics can help you plan the segmentation, write the discussion guide, and analyse your transcripts thematically — chapter by chapter.
