Interviews in Social Research: Advantages and Disadvantages

Last Updated on October 10, 2026 by Karl Thompson

An interview involves a researcher asking questions verbally to a respondent. Interviews involve a more direct interaction between the researcher and the respondent than questionnaires, and they can be conducted face to face, by phone, by video link or over social media.

There are five things you need to know about interviews for A-level sociology:

1. The main types of interview – structured, semi-structured, unstructured, and group interviews (including focus groups)
2. The key strength of unstructured interviews is validity – they let respondents explain themselves in their own words
3. The key limitation is reliability – every unstructured interview is different, and the interviewer can affect what respondents say
4. Sociological perspectives – interpretivists and feminists favour unstructured interviews, while positivists prefer structured ones
5. How interviews come up in the exam – on Paper 3 and on Paper 1, including methods in context questions

Interviews are part of the Research Methods topic within Theory and Methods. Theory and Methods is examined on Paper 3 (Crime and Deviance with Theory and Methods, 7192/3) and on Paper 1 (Education with Theory and Methods, 7192/1), where interviews can also come up applied to education in the methods in context question. Most students meet the methods questions on Paper 3 first.

This post has primarily been written for students studying A-level sociology, but it should also be useful for students studying methods for psychology, business studies and maybe other subjects too!

Types of interview

Structured or formal interviews are those in which the interviewer asks each respondent the same questions in the same way. This will typically involve reading out questions from a pre-written and pre-coded questionnaire, which forms the interview schedule. The most familiar form of this is market research, where you may have been stopped on the street by a researcher ticking boxes based on your responses. Structured interviews are a type of social survey, and they have their own post: Structured Interviews in Social Research.

Unstructured or informal interviews (sometimes called in-depth interviews) are more like a guided conversation. Here the interviewer has a list of topics they want the respondent to talk about, but the interviewer has complete freedom to vary the specific questions from respondent to respondent, so they can follow whatever lines of enquiry they think are most appropriate, depending on the responses given by each respondent. Most of this post is about unstructured interviews.

Semi-structured interviews are those in which the interviewer has a list of questions, but is free to ask further, differentiated questions based on the responses given. This allows more flexibility than the structured interview yet more structure than the unstructured interview. Semi-structured interviews have their own section below.

Group interviews – interviews can be conducted either one to one (individual interviews) or in a group, in which the interviewer interviews two or more respondents at a time. Group discussions among respondents may lead to deeper insight than just interviewing people alone, as respondents ‘encourage’ each other.

Focus groups are a type of group interview in which respondents are asked to discuss a particular topic. Both are covered in their own section below.

Interviews: key terms

The interview schedule – a list of questions or topic areas the interviewer wishes to ask or cover in the course of the interview. The more structured the interview, the more rigid the interview schedule will be. Before conducting an interview it is usual for the researcher to know something about the topic area and the respondents themselves, and so they will have at least some idea of the questions they are likely to ask. Even if they are doing unstructured interviews, an interviewer will have some kind of interview schedule, even if it is just a list of broad topic areas to discuss, or an opening question.

Transcription of interviews – transcription is the process of writing down (or typing up) what respondents say in an interview. In order to be able to transcribe effectively, interviews will need to be recorded.

The problem of leading questions – in unstructured interviews, the interviewer should aim to avoid asking leading questions, which suggest the answer the interviewer expects.

The advantages and disadvantages of unstructured interviews

The mind map below summarises the strengths and limitations of unstructured interviews. Each one is explained underneath, along with the counter-arguments you’ll need for evaluation.

The strengths of unstructured interviews

Good validity. The key strength of unstructured interviews is validity – they should get closer to what respondents really think and feel than a questionnaire can. For this to happen, questioning should be as open-ended as possible, to gain genuine, spontaneous information rather than ‘rehearsed responses’, and questioning needs to go far enough to get in-depth answers rather than glib, easy ones. HOWEVER, validity is not guaranteed just because the interview is unstructured. As the limitations below show, the interviewer can affect what respondents say, so validity depends heavily on the skill of the interviewer.

Respondent led. Unstructured interviews are ‘respondent led’ – the researcher listens to what the respondent says and then asks further questions based on it. This should allow respondents to express themselves and explain their views more fully than with structured interviews.

Flexibility. The researcher can change their mind about what the most important questions are as the interview develops. Unstructured interviews thus avoid the imposition problem – respondents are less constrained than with structured interviews or questionnaires, in which the questions are written in advance by the researcher. HOWEVER, flexibility comes at a direct cost to reliability: every time the researcher follows a new line of questioning, the interview becomes harder to compare with the others and harder for anyone else to repeat.

Rapport and empathy. Unstructured interviews encourage a good rapport between interviewee and interviewer. Because of their informal nature, like guided conversations, unstructured interviews are more likely to make respondents feel at ease than the more formal setting of a structured questionnaire or experiment. This should encourage openness, trust and empathy. HOWEVER, the closeness that produces rapport is the same thing that produces interviewer effects. A respondent who likes the interviewer may be more inclined to give the answers they think the interviewer wants to hear.

Checking understanding. Unstructured interviews also allow the interviewer to check understanding. If an interviewee doesn’t understand a question, the interviewer is free to rephrase it, or to ask follow-up questions to clarify parts of answers that were not clear the first time.

Good for sensitive topics. Unstructured interviews are more likely to make respondents feel at ease with the interviewer, and they allow the interviewer to show more sympathy (if required) than the colder, more mechanical quantitative methods. A classic example is Dobash and Dobash’s research on domestic violence, published as Violence Against Wives (1979), which drew on in-depth interviews with 109 women in Scotland who had been beaten by their husbands. It is hard to imagine a questionnaire getting the same depth on a topic like this. HOWEVER, a comfortable atmosphere does not guarantee disclosure. Respondents may still withhold things that are painful, embarrassing or that clash with how they see themselves.

Good for finding out why respondents do not do certain things. Questionnaires struggle to explain the absence of a behaviour – why someone doesn’t vote, apply for something or take up a service – because the reasons are often complicated and personal. Unstructured interviews give respondents room to explain. Because they are spoken rather than written, they are also better suited to researching people who may have low literacy skills.

Empowerment for respondents. The researcher and respondents are on a more equal footing than with more quantitative methods. The researcher doesn’t assume they know best, and this empowers the respondents. Feminist researchers in particular believe that the unstructured interview can neutralise the hierarchical, exploitative power relations that they see in the more traditional interview. The best-known version of this argument is Ann Oakley’s chapter ‘Interviewing women: a contradiction in terms?’ (1981), based on her experience of interviewing women in a study of the transition to motherhood. Oakley criticised the conventional textbook model of interviewing, in which the interviewer directs the questioning, stays detached and treats the interviewee as a source of data. In her alternative, a woman recounts her experiences in her own words, with the interviewer acting as a guide to the account, and a more reciprocal relationship between the two is welcomed rather than avoided. HOWEVER, critics such as Phoenix (1994) argue that the idea of feminist interviewing as a cosy, equal exchange based on shared gender ignores the differences between women in terms of ethnicity and class. And however equal the interview itself feels, it is still the researcher who decides what to ask, what gets published and how it is interpreted.

Practical advantages. There are few practical advantages with this method, but compared to full-blown participant observation, unstructured interviews are a relatively quick method for gaining in-depth data. They are also a good method to combine with overt participant observation, to get respondents to explain the meanings behind their actions. So in short, they are impractical, unless you’re in the middle of a year-long participant observation study (it’s all relative!).

The limitations of unstructured interviews

Lack of reliability. The main theoretical disadvantage is the lack of reliability. Unstructured interviews lack reliability because each interview is unique – a variety of different questions are asked and phrased in a variety of different ways to different respondents.

Difficult to repeat. They are also difficult to repeat, because the success of the interview depends on the bond of trust between the researcher and the respondent – another researcher who does not relate to the respondent in the same way may get different answers.

Validity can be undermined in several ways:

• respondents might prefer to give rational responses rather than fuller emotional ones (it’s harder to talk frankly about emotions with strangers)
• respondents may not reveal their true thoughts and feelings because they do not coincide with their own self-image, so they simply withhold information
• respondents may give answers they think the interviewer wants to hear, in an attempt to please them!

Interviews only tell you what people say. We also need to keep in mind that interviews can only tap into what people SAY about their values, beliefs and actions. We don’t actually get to see these in action, as we would with observational studies such as participant observation. What people say they do and what they actually do can be quite different.

Interviewer bias might undermine the validity of unstructured interviews – this is where the values of the researcher interfere with the results. The researcher may give away whether they approve or disapprove of certain responses in their body language or tone of voice (or in the wording of probing questions), and this in turn might encourage or discourage respondents from being honest.

The characteristics of the interviewer might also bias the results and undermine validity – how honest the respondent is in the course of an hour-long interview might depend on the class, gender or ethnicity of the interviewer. Sudman and Bradburn (1974) reviewed a large body of research on these ‘response effects’ and found that answers varied with the characteristics of the interviewer relative to the respondent – and that an interviewer’s characteristics were most likely to affect answers to questions connected to that characteristic. Race of interviewer, for example, mattered most on questions about race.

A well-known illustration comes from the linguist William Labov (1969). In interviews with African-American children in Harlem, children interviewed one to one in a formal setting by white interviewers gave short, one-syllable answers and could easily have been judged as having poor verbal skills. When the setting was changed – a Black interviewer from the local area, with a friend present – the same children were highly talkative. The ‘finding’ about the children’s language turned out to be a product of the interview situation.

Unstructured interviews lack representativeness. Because they are time-consuming, it is difficult to get a large enough sample to be representative of large populations.

It is difficult to quantify data, compare answers and find statistics and trends, because the data gained is qualitative.

Practical disadvantages. Unstructured interviews may take a relatively long time to conduct, and some interviews can take hours. They also need to be recorded and transcribed, and in the analysis phase there may be a lot of information that is not directly relevant to the research topic that needs to be sifted through.

Interpersonal skills and training. A further practical problem is that some researchers may lack the interpersonal skills required to conduct unstructured interviews. Training might need to be more thorough for researchers undertaking unstructured interviews – to reduce the problem of interviewer bias.

Ethical issues with unstructured interviews

Unstructured interviews are not usually ethically problematic in the way that covert research can be, because respondents know they are being interviewed. HOWEVER, the depth that makes them valuable also raises ethical issues of its own.

Informed consent. Respondents should know what the research is about and agree to take part. With unstructured interviews this is less straightforward than it sounds, because the conversation can drift onto topics nobody planned. Consent is better thought of as ongoing – respondents should feel able to decline a question or stop the interview at any point.

Confidentiality and anonymity. The researcher is getting more in-depth data – more of an insight into who the person really is – and this means the information could do more harm to the respondent if it got into the wrong hands. Detailed quotations can also make people identifiable even when names are removed, especially in a small community or a single school.

Harm and distress. Talking about sensitive experiences – abuse, bereavement, crime – can be upsetting. The researcher needs to think in advance about how to handle distress and whether to point respondents towards support. How serious all of this is depends on the topics discussed and the exact content of the interviews.

The British Sociological Association’s Statement of Ethical Practice sets out the standards sociologists in the UK are expected to follow.

Semi-structured interviews

Semi-structured interviews sit between structured and unstructured interviews. The interviewer has a set of questions that every respondent is asked, but is free to probe, follow up and ask further questions depending on the answers.

Strengths. Semi-structured interviews give some of the comparability of a structured interview, because everyone answers the core questions, while keeping some of the flexibility and depth of an unstructured interview. They are a popular compromise in education research, where researchers often want to compare the views of different groups – pupils, teachers, parents – while still letting people explain themselves.

Limitations. They inherit a weaker version of both sets of problems. The follow-up questions differ between respondents, so they are less reliable than structured interviews. The fixed questions mean the researcher still partly imposes their own framework, so they are less respondent-led than unstructured interviews. And the interviewer effects described above apply just as much.

For an example of semi-structured interviews in use, see Semi-Structured Interviews to Research Education, which summarises Archer, Halsall and Hollingworth’s (2007) research with working-class pupils in London schools who had been identified as at risk of dropping out.

Group interviews and focus groups

A group interview is where the researcher interviews several respondents at the same time. A focus group is a type of group interview in which the group discusses a particular topic in depth, and the researcher is interested in how people respond to each other as members of a group, not just in what each individual says (Bryman, 2016).

Group interviews are widely used in education research, where young people may be more comfortable talking alongside their friends than alone with an adult stranger. Paul Willis’s research for Learning to Labour used group discussions with ‘the lads’ alongside observation.

The strengths of group interviews

Interaction between respondents. Discussion between respondents can spark ideas and conversations that the interviewer hadn’t anticipated, which can then be probed further. Respondents can bounce ideas off each other, build on each other’s answers and even challenge each other.

A more natural, less intimidating setting. For some respondents – especially pupils – a group of peers can feel safer than a one-to-one interview with an adult, and peer support can reduce the power imbalance between the interviewer and the respondents. If the researcher is skilled, they may even drift into the background.

Observing group dynamics. The researcher can see how views are formed and negotiated in a group, which is useful if the topic is something people experience collectively, such as peer group culture.

Time efficiency. Several respondents can be interviewed in the time it takes to interview one, although transcription takes longer because several voices need to be separated.

The limitations of group interviews

Peer pressure and conformity. Respondents may go along with the group view, or give answers that will make them look good in front of their peers, rather than saying what they really think. They may also encourage each other to exaggerate or show off.

Dominant individuals. One or two confident people can dominate the discussion while quieter respondents say little. One person can change the dynamic of a group of three or four people enormously.

Even harder to repeat. Group interviews are especially difficult to repeat, because the dynamics of the interview are influenced not just by the researcher but also by the mix of people in the group, so they are low in reliability.

Confidentiality. The researcher can promise to keep what is said confidential, but cannot guarantee that the other members of the group will.

For a worked exam example, see Evaluate group interviews to investigate subject choices made by pupils, and for interviews applied to education more generally see Using interviews to research education.

Sociological perspectives on interviews

Positivists

Positivists are not against interviews as such – the structured interview is one of the standard ways of carrying out a social survey. What positivists want is standardisation: every question asked to every respondent in the same way, by an interviewer who stays neutral, shows no emotion, avoids suggesting replies and does not skip questions. This produces reliable, quantifiable data that can be compared across a large sample.

From a positivist point of view, unstructured interviews are the problem case. There is no guarantee that responses aren’t products of the interview situation – the particular interviewer, the particular conversation – rather than a reflection of the underlying social reality.

Interpretivists and interactionists

Interactionists see things the other way round. Silverman (1993), drawing on Denzin (1970), summarises the interactionist view of the interview as a form of social interaction in its own right. The context of the interview is part of understanding the responses, and no sharp distinction is drawn between a research interview and any other social interaction. Interview data are valid when the interviewer and respondent reach a mutual understanding of what is being said.

Interactionists therefore prefer non-standardised interviews, because they allow respondents to shape the interview according to their own view of the world.

Feminists

Feminist researchers have been some of the strongest supporters of unstructured interviews, for the reasons covered under empowerment above. Oakley’s (1981) argument is that the traditional, detached interview reproduces the hierarchy between researcher and researched, and that a more equal, reciprocal relationship produces better research as well as more ethical research. HOWEVER, not all feminist researchers reject quantitative methods, and critics such as Phoenix (1994) question whether shared gender is enough to create an equal relationship.

In the exam, these perspectives are your main source of evaluation: the positivist case against unstructured interviews is essentially the interpretivist case for them, turned upside down.

How interviews come up in the AQA exam

Interviews come up in the Theory and Methods sections of Paper 3 (7192/3) and Paper 1 (7192/1), and in the methods in context question on Paper 1.

Paper 3 – Theory and Methods. There are two types of methods question:

• 10-mark ‘Outline and explain two…’ questions with no item – for example, two advantages or two problems of a particular method. Remember that there are no marks for evaluation on the no-item 10-markers: explain each point in depth instead.
• 20-mark ‘Evaluate…’ questions with an item – here evaluation is essential, and the strengths and limitations above, together with the perspectives section, are your raw material.

Paper 1 – Education with Theory and Methods. The same methods material can come up in the 10-mark Theory and Methods question. Interviews can also come up in the 20-mark methods in context question, where you have to apply the method to a specific education topic and research context, not just list general strengths and limitations. See how to answer methods in context questions and the worked example on group interviews and subject choice.

Practice question and plan

This is a practice question written in the AQA style, not a question from an AQA past paper.

Outline and explain two theoretical problems of using unstructured interviews in sociological research. (10 marks)

A possible plan:

• Problem 1: low reliability. Each interview is unique – different questions, asked in different ways – so another researcher cannot repeat it and check the results. Develop it: this matters to positivists because they want findings that can be checked and compared; link it to flexibility, since the very thing that makes unstructured interviews useful is what makes them unrepeatable.
• Problem 2: interviewer effects undermine validity. The interviewer’s characteristics and reactions can shape what respondents say. Develop it: use Labov’s Harlem interviews or Sudman and Bradburn’s review of response effects as evidence; explain that the closer the rapport, the greater the risk of respondents saying what they think the interviewer wants to hear.

Because this is a no-item 10-marker, there are no marks for evaluating these problems – the marks come from explaining each one in depth.

Revision notes

If you like this sort of thing, then you might like my Theory and Methods revision notes bundle.

[IMAGE: Theory and Methods bundle cover, linked to https://revisesociology.sellfy.store/p/2qyB/]

Sources and further reading

• Bryman, A. (2016) Social Research Methods, 5th edition, Oxford University Press
• Denzin, N. K. (1970) The Research Act: A Theoretical Introduction to Sociological Methods, Aldine
• Dobash, R. E. and Dobash, R. P. (1979) Violence Against Wives: A Case Against the Patriarchy, Free Press
• Labov, W. (1969) ‘The logic of non-standard English’
• Oakley, A. (1981) ‘Interviewing women: a contradiction in terms?’, in H. Roberts (ed.) Doing Feminist Research, Routledge & Kegan Paul
• Phoenix, A. (1994), as discussed in Oakley, A. (2016) ‘Interviewing women again: power, time and the gift’, Sociology, 50(1)
• Silverman, D. (1993) Interpreting Qualitative Data, Sage
• Sudman, S. and Bradburn, N. M. (1974) Response Effects in Surveys, Aldine/National Opinion Research Center

Recommended further reading: Gilbert and Stoneman (2016) Researching Social Life.

Signposting and related posts

This post has been written for students studying research methods as part of the AQA’s A-level sociology course, examined on Paper 3 and Paper 1.

Interviews are usually studied alongside the other main primary methods. Structured Interviews in Social Research covers the standardised version of the interview, and Questionnaires in Sociology covers its written equivalent.

Participant observation is the other main qualitative method, and is often combined with unstructured interviews.

The concepts used throughout this post are explained in validity and reliability.

For interviews applied to education, see Using interviews to research education.

For more posts on research methods please see my main research methods page.

Please click here to return to the homepage – ReviseSociology.com