Getting started
Talk based methods
What are talk-based methods?
Talk-based methods are qualitative research approaches that involve having conversations with people to understand their experiences, views, and perceptions1.
These methods allow participants to describe their experiences in their own words. In a public health context, talk-based methods are often used to:
oExplore lived experience of services or systems and understand topics from different perspectives (e.g., that of residents and professionals)
Understand barriers and facilitators to engagement with services or interventions
Provide depth and context alongside numerical data and survey findings in mixed-methods projects
Insight and engagement uses pragmatic approaches to engaging residents and professionals which generate knowledge to inform the design and delivery of specific interventions, services and policies.
Research and evaluation uses systematic and robust approaches to generate knowledge which helps us test ideas and understand the delivery or impact of interventions, services or policies.
All of these activities can involve using talk-based approaches. Research and evaluation involves planned sampling, structured topic guides, formal consent, and systematic analysis of data. Whilst engagement tends to be less rigorous, there is still value in using some of the approaches and practices outlined in this guide.
This guide focuses primarily on interviews and focus groups, as these are the most commonly used talk-based methods across the department. The guide will reference a few other approaches, but they are not covered in detail.
When should I use interviews and focus groups?
Interviews and focus groups are usually most appropriate when your project aims to understand how and why, rather than how many.
They may be a good choice if you want to:
Explore lived experience, feelings, or perspectives
Understand decision-making, behaviours, or journeys through a service
Understand why patterns shown in quantitative data are occurring
Understand groups who may be underrepresented in local or national data and services
Do not use interviews and focus groups if you need:
Statistically representative findings
Population-level prevalence estimates
Quick, low-resource data collection
In most of the work we do in Health & Wellbeing, talk-based methods work best alongside other methods (e.g., surveys, routine data analysis), rather than as a standalone approach.
Types and uses
There are several different types/approaches to interviews and focus groups. Below are summaries of different approaches that are helpful for our work in public health.
Structured interviews
Structured interviews involve asking questions in the same order in every interview you carry out, with less flexibility to probe or follow up on an interviewee’s responses. This approach allows for participant answers to easily be directly compared. Depending on the structure of the questions asked, data generated can be analysed quantitatively.
We can think of this method as a type of verbal questionnaire or survey. In public health and local authority settings, this approach might be used to:
Ask residents about their awareness of or satisfaction with a local service
Capture self-reported health behaviours across a community
Gather standardised information from practitioners about their knowledge of a particular pathway or protocol
Semi-structured interviews
Semi-structured interviews are the most commonly used interview approach in public health. They involve asking a limited number of open-ended questions based on the topic you are researching, and allow flexibility to:
Probe2 interesting points in more depth
Encourage interviewees to elaborate on their points
Adapt language or the order of questions based on participant responses
Semi-structured interviews are guided by a topic guide, which outlines the questions you will ask. We use topic guides flexibly to help steer the conversation. The sequence of questions might vary between interviews, as well as the level of probing.
This approach balances consistency and flexibility. Using a shared topic guide supports comparability across interviews, while still allowing space for participants to raise issues on the topic that matter to them.
This approach works particularly well for:
Exploring experiences of services or interventions
Understanding barriers and facilitators (e.g., to accessing services, adopting behaviours, implementing change)
Capturing professional knowledge and understanding decision-making
Early or exploratory work (e.g., where little is known about a topic)
Semi-structured interviews involve skilled facilitation. As an interviewer, you need to listen actively, probe appropriately, and be reflexive about how your questions and assumptions might shape the discussion. Practical guidance can be found in the interviewssection.
Narrative interviews
Narrative interviews are a type of unstructured interview that allow a participant to share their stories and accounts in their own words3. The defining feature of this approach is that the participant largely shapes the structure and direction of what is discussed.
Instead of using a detailed topic guide, the interviewer encourages the participant to describe their experiences freely (e.g., “tell me about your experience of…”) and follows the participant’s lead, using prompts where needed. The interview will often be guided by a broad list of topics to explore, but the wording and order of the questions are shaped by the participant’s responses.
This approach is grounded in the idea that how people construct and tell their stories is in itself analytically important, not just the facts they tell us.
Narrative interviews are especially well-suited to projects where:
The topic is exploratory or poorly understood, and participant-led accounts are more likely to surface what matters versus researcher-led questioning.
Meaning, context, and personal or professional experience are central to the research aims, rather than frequency or prevalence
The aim is to understand how people make sense of an experience. For example, a resident describing their pathway through mental health services, or a professional recounting how an organisational change affected their practice.
Narrative interviews can produce very rich data, but they require a high level of skill and might be less suitable for projects which are time or resource limited which is why this approach is less commonly used in public health and local authority settings. However, you may want to consider it if depth and participant-led accounts are central to the aims of your project.
In-depth interviews
In-depth interviews are a type of unstructured or loosely structured interview, where your typically begin with a broad, open question and then encourages the participant to speak openly about the topic. Like narrative interviews, they are relatively personal to each participant and shaped by the direction of the conversation, but you play a more active role in guiding exploration of the topic or issue.
They allow you to explore participants’ experiences, perspectives, behaviours, and interpretations in detail, with a focus on understanding underlying concepts and processes. This approach is particularly useful where a topic is complex or poorly understood, where we need much more detailed insight into how a system works or how decisions have been made, and when working with participants who have specialist knowledge or long-term involvement (e.g., senior professionals or sector experts).
This approach is not limited to any one type of participant. In-depth interviews can be used equally effectively with:
Residents - for example, exploring how families navigate accessing dental services, or how people living with a long-term condition understand and manage their health.
Professionals and practitioners - for example, understanding how school nurses make decisions about referring pupils, or how commissioners have responded to changing public health guidance.
People with lived experience of a service or system – for example, where you want to go beyond surface-level responses to understand what an experience actually felt like and meant to a participant.
‘Unstructured’ does not mean unplanned, informal, or without preparation, and an unstructured interview is not just ‘a conversation’.
What ‘unstructured’ refers to is the format of the interview. Specifically, questions are not fixed in wording or order in advance, and the direction of the interview is shaped by the participant’s responses rather than a predetermined script.
In practice, unstructured and semi-structured interviews typically involve significant preparation. The researcher will have developed a carefully considered topic guide or list of topics to explore, thought through how to open the interview and build rapport, anticipated any areas of sensitivity, and planned how to probe effectively.
Using this method requires a lot of skill, in order to make a structured, purposeful conversation feel open and participant-led, which can often be much harder than following a fixed script.
Paired, dyadic and joint interviews
Paired, dyadic or joint interviews involve interviewing two people together, usually because they share a relationship, role, or experience that is relevant to the research.
Paired interviews can be particularly useful for exploring shared experiences or journeys, and where participants have an established relationship (e.g., a parent and their child, or two colleagues).
Paired interviews should not be seen as ‘mini focus groups’4 – they are a method in their own right, with distinct benefits and challenges.
Paired interviews can generate insight not only from what participants say, but from how they respond to each other’s responses. This can help us understand areas of agreement, difference, or shared understanding in ways that might not emerge in a typical one-to-one interview.
Paired or dyadic interviews are not suitable for all topics, particularly where topics are highly sensitive or where there may be safeguarding concerns. As with any talk-based method, their use should be guided by the aims of the project and the needs and safety of participants.
When evaluating a local authority-commissioned dementia support service, the evaluation team choose to use paired interviews to collect data from service users and families. A paired interview with a person living with dementia and their family carer can be especially useful for exploring the shared experience of care.
Interviewing them together can surface how decisions are made and negotiated in everyday life, such as managing appointments, medication, home safety, or social activities, and how each person experiences the same support differently. For instance, a carer may describe the service as “a lifeline” whilst the person with dementia may emphasise feeling “talked over” in appointments or may recall different aspects of what helped. The interaction itself (prompting, reassuring, disagreeing) can reveal practical dynamics that are hard to capture in separate one-to-one interviews.
Focus groups
Focus groups are a form of facilitated group interview where data is collected from multiple participants at the same time, guided by a skilled facilitator using a topic guide. Focus groups can reveal perspectives that might not always emerge in one-to-one interviews, but can become visible through discussion, disagreement, and collective reflection5.
Focus groups use semi-structured topic guides, but the facilitator’s role goes beyond just asking the questions. They must manage the pace and direction of discussion, ensure everyone has the opportunity to share, handle disagreement well, and create an environment in which participants feel comfortable speaking openly. This requires preparation, skill, and attention to group dynamics.
Focus groups are especially well suited to:
Exploring shared or contested experiences
Examining how views on a topic are shaped by community context, shared identity, or cultural frameworks
Engaging communities who prefer collective discussion
Focus groups can be particularly valuable where there are power differences between participants and decision-makers, commissioners, or professionals. The group setting can support confidence, mutual validation, and participants might feel more able to voice critical or challenging perspectives when they are not alone, and when they can see that others share their views.
This method is widely used in a public health context, and can be a very valuable source of data, but good focus groups require careful planning, skilled facilitation, and consideration of group dynamics (at the planning, facilitation and analysis stages) and confidentiality. Practicel guidance can be found in the focus groups section.
Choosing between interviews and focus groups
Choosing between interviews and focus groups should be guided by the overarching aims and objectives of your research or evaluation.
Both of these methods involve facilitated conversation, but they are not interchangeable. They generate different types of data and have different purposes.
Your choice on what method you use should be informed by what you are trying to understand, whose voices you need to hear, how the data will be used, as well as practical considerations such as time, resources, and ethical requirements.
In many projects, there is no single ‘right’ method. Some research questions may be better answered through one-to-one interviews, others through group discussion, and in many cases through a combination of both.
Having clear project aims at the outset will help you choose the most appropriate approach and avoid collecting data that does not fully meet your needs.
Table 1 below can help you decide which method is most appropriate for your project.
In some cases, a multi-method approach may be appropriate. For example, using interviews to explore sensitive issues in depth, followed by focus groups to test or refine emerging themes.
Practical considerations also matter. For example:
Recruiting for focus groups can be very challenging
Focus groups require strong facilitation skills
Interviews may take longer overall if you are involving a lot of participants.
Later chapters of this guide provide more detailed practical guidance on planning and delivering both interviews and focus groups.
| Key question or consideration | One-to-one interviews | Focus groups |
|---|---|---|
| Is the topic(s) you want to discuss sensitive, personal, or potentially distressing? | ✓ | ✕ |
| Is privacy or confidentiality likely to be particularly important to participants? | ✓ | ✕ |
| Would collective reflection or idea generation add value to your project? | ✕ | ✓ |
| Do you want data in the form of detailed individual stories or journeys? | ✓ | ✕ |
| Are interactions between participants likely to generate valuable insight? | ✕ | ✓ |
| Might participants feel uncomfortable speaking openly in a group? | ✓ | ✕ |
| Are you exploring experiences that are highly individual or varied? | ✓ | ✕ |
| Do you want to explore commonality between individuals experiences? | ✕ | ✓ |
| Do you want to understand agreement, disagreement, or consensus within groups? | ✕ | ✓ |
| Are participants likely to feel more confident speaking together or with others? | ✕ | ✓ |
Interviews vs Focus Groups
Footnotes
UKRI. What is social science? Available from: https://www.ukri.org/who-we-are/esrc/what-is-social-science/qualitative-research/↩︎
Probes are designed to encourage deeper thinking and allow the interviewer to ask the participant to provide additional information about their response to a question.↩︎
Jovchelovitch, S. and Bauer, M.W., 2000. Narrative interviewing. Qualitative researching with text, image and sound, 57, p.74. Available from: https://eprints.lse.ac.uk/2633/1/Narrativeinterviewing.pdf↩︎
Morgan, D.L., Ataie, J., Carder, P. and Hoffman, K., 2013. Introducing dyadic interviews as a method for collecting qualitative data. Qualitative health research, 23(9), pp.1276-1284. Available from: https://journals.sagepub.com/doi/10.1177/1049732313501889↩︎
Braun, V and Clarke, V. (2022) Successful qualitative research: A practical guide for beginners. Los Angeles: SAGE. Available from: Braun & Clarke Textbook↩︎