Unlike everyday conversations, research interviews are set up by the interviewer to elicit information from participants. The purpose of the interview is the discovery of informants’ feelings, perceptions and thoughts. Marshall and Rossman (2006) maintain that interviews focus on the past, present and, in particular, the essential experience of participants. The interview can be formal or informal; often informal conversations or chats with participants also generate important ideas for the project. Depending on the response of participants, researchers formulate questions as the interview proceeds rather than asking pre-planned questions. This means that each interview differs from the next in sequence and wording, although distinct patterns common to all interviews in a specific study often emerge in the analysis. Indeed, for many research approaches it is necessary that researchers discover these patterns when analysing data.

One interview, however, does not always suffice. In qualitative inquiry it is possible to re-examine the issues in the light of emerging ideas and interview for a second or third time. Seidman (2006) sees three interviews as the optimum number, but these require much planning in the short time span available to undergraduates for their project, so this is only possible for postgraduates. Many novice researchers therefore use one-off interviews although postgraduates and other more experienced researchers sometimes carry out more than one with each participant.

Pilot studies are not always used in qualitative inquiry as the research is developmental, but novice researchers could try interviews with their friends and acquaintances to get used to this type of data collection. We found that we lacked confidence when we started, and a practice run proved very useful. In our experience students become more confident as interviews proceed.

Most qualitative research starts with relatively unstructured interviews in which researchers give minimal guidance to the participants. The outcome of initial interviews guides later stages of interviewing. As interviews proceed, they become more focused on the particular issues important to the participants and which emerge throughout the data collection. Most qualitative studies do not only explore commonalities and uncover patterns, but they also describe the unique experiences of individuals particularly in one-to-one interviews.

One-to-one interviews are the most common form of data collection although researchers also use group interviews (see Chapter 8).

Researchers have to decide on the structure in the interview. There is a range of interview types on a continuum, from the unstructured to the structured. Qualitative researchers generally employ the unstructured or semi-structured interview.

The unstructured, non-standardised interview

Unstructured interviews start with a general question in the broad area of study. Even unstructured interviews are usually accompanied by an aide mémoire, an agenda or a list of topics that will be covered. There are, however, no predetermined questions except at the very beginning of the interview.

This type of unstructured interviewing allows flexibility and makes it possible for researchers to follow the interests and thoughts of the informants rather than follow their own assumptions. Interviewers freely ask questions from informants in any order or sequence depending on the responses to earlier questions. Warm-up and simple questions are generally asked first; however, if the interviewer leaves the essential questions till the end of the interview the participant may be tired and reluctant to discuss deeper issues.

Researchers also have their own agenda. To achieve the research aim, they keep in mind the particular issues which they wish to explore. However, direction and control of the interview by the researcher is minimal. Generally, the outcomes of these interviews differ for each informant, though usually certain patterns can be discerned. Informants are free to answer at length, and great depth and detail can be obtained. The unstructured interview generates the richest data, but it also has the highest ‘dross rate’ (the amount of material of no particular use for the researcher’s study), particularly when the interviewer is inexperienced.

The semi-structured interview

Semi-structured or focused interviews are often used in qualitative research. The questions are contained in an interview guide (not interview schedule as in quantitative research) with a focus on the issues or topic areas to be covered and the lines of inquiry to be followed. The sequencing of questions is not the same for every participant as it depends on the process of the interview and the responses of each individual. The interview guide, however, ensures that the researcher collects similar types of data from all informants. In this way, the interviewer can save time, and the dross rate is lower than in unstructured interviews. Researchers can develop questions and decide for themselves what issues to pursue.

The interview guide can be quite long and detailed although it need not – should not – be followed strictly so that the participant has some control. It focuses on particular aspects of the subject area to be examined, but it can be revised after several interviews because of the ideas that arise. Although interviewers aim to gain the informants’ perspectives, the former need to keep some control of the interview so that the purpose of the study can be achieved and the research topic explored. Ultimately, the researchers themselves must decide what interview techniques or types might be best for them and the interview participants. Our students and other researchers preferred good questions of medium length combined with the use of prompts and reported better results.

The structured or standardised interview

Qualitative researchers in general do not use standardised interviews as they are contradictory to the aims of qualitative research. In these, the interview schedule contains a number of pre-planned questions. Each informant in a research study is asked the same questions in the same order. This type of interview resembles a written survey questionnaire. Standardised interviews save time and limit the interviewer effect. The analysis of the data seems easier as answers can be found quickly. Generally, knowledge of statistics is important and useful for the analysis of this type of interview. However, this type of pre-planned interview directs the informants’ responses and is therefore inappropriate in qualitative approaches. Structured interviews may contain open questions, but even then they cannot be called qualitative.

Qualitative researchers use structured questions only to elicit socio-demographic data, i.e. about age, duration of condition, duration of experience, type of occupation, qualifications, etc. Sometimes research or ethics committees ask for a predetermined interview schedule so that they can find out the exact path of the research. For the purpose of gaining permission, a semi-structured interview guide is occasionally advisable for health researchers.

Types of questions in qualitative interviews

When asking questions, interviewers use a variety of techniques. Patton (2002) lists particular types of questions, for example experience, feeling and knowledge questions.

Spradley (1979) distinguishes between grand-tour and mini-tour questions. Grand-tour questions are broader, while mini-tour questions are more specific.

The sequencing of questions is also important.

Practical considerations

In qualitative studies questions are as non-directive as possible but still guide towards the topics of interest to the researcher. Researchers should phrase questions clearly and aim at the various participants’ levels of understanding. Ambiguous questions lead to ambiguous answers. Double questions are best avoided; for instance it would be inappropriate to ask: How many colleagues do you have, and what are their ideas about this?

The researcher must be aware of practical difficulties in the data collection phase, particularly when interviewing in hospital. The routine of the hospital is disrupted by the presence of the nurse or midwife researcher whose activities might be viewed with suspicion by colleagues. A quiet place for interviews cannot always be found, and therefore the privacy of patients may be threatened. The ward might be full of noise and activity, and the researcher does not always find a convenient slot for interviewing without being interrupted by nursing activity, consultant round, cleaners, meals and so on. In the community, interviews are often interrupted by children or spouses and by the visits of friends or relatives.

During the interviews researchers can use prompts or probing questions. These help to reduce anxiety for researcher and research informant. The purpose of probes is a search for elaboration, meaning or reasons. Seidman (2006) suggests the term ‘explore’ and dislikes the word ‘probe’ as it sounds like an interrogation, and is the name for a surgical instrument used in medical or dental investigations and stresses the interviewer’s position of power.

Exploratory questions might be, for instance: What was that experience like for you? How did you feel about that? Can you tell me more about that? That’s interesting, why did you do that? Questions can follow up on certain points that participants make or words they use. The researcher could also summarise the last statements of the participant and encourage more talk through this technique.

Participants often become fluent talkers when asked to tell a story, reconstructing their experiences, for instance a day, an incident, the feeling about an illness. Unfortunately the data from interviews are sometimes more fluent or extensive when the participants are articulate, and occasionally researchers may choose those who have language and interaction skills. This may create bias in the interviews however and is not a good strategy.

The social and language skills of the researcher often make a difference to the outcome of the interview.

Non-verbal prompts are also useful. The stance of the researcher, eye contact or leaning forward encourages reflection. In fact, listening skills, which some nurses and midwives already possess from the counselling of patients, will elicit further ideas. Patients often give monosyllabic answers until they have become used to the interviewer, because they are reluctant to uncover their feelings or fear that judgements might be made about them. When participants do not understand the interview question, the researcher can rephrase them in the language they understand.

Interviews must be seen in the social context in which they occur; this affects the relationship between researcher and research as well as the data generated by it (Manderson et al., 2006). The setting is of particular importance; if interviews take place in the home of the participants, they are more relaxed, the researcher might gain richer data and the participant is in some position of control. On the other hand, this setting can be a difficult choice for the researcher as there might be many distractions such as children or spouses who interrupt the proceedings. Sometimes a neutral place such as a corner in a café or park, or an academic environment can be appropriate.

The researcher has to reflect on time and location and the persons involved in the interaction. Experience, background and characteristics of the researcher, as well as the participants’ group membership such as age, gender, class or ethnic group might also influence the interaction. Manderson et al. (2006) suggest that changes in any of these factors might generate different interview data as the social dynamics of the interview vary; indeed Roulston et al. (2003: 654) stress ‘the socially constructed nature of interview talk’. When sensitive topics are discussed, researchers have to use their own judgement whether their gender or ethnic membership might interfere with the research relationship. In some situations it is more sensitive and even useful when researcher and participant are of similar background. This is by no means always so. One of our students, a very young woman, interviewed older people about their lives. This study elicited more data than would be usual. The participants provided very rich data and deep thoughts – perhaps because they did not feel threatened.

When patients are interviewed, they might ask the researcher about advice on their condition or treatment. It is best to separate the researcher and professional roles, although this cannot always be done. It is best to point out a professional source of information or put the participants in touch with an expert who can answer their questions. In the case of very vulnerable people and sensitive topics, the researcher might seek advisors or experts before the research starts and ask for permission to contact them if necessary. If an emergency occurs during the interview, the researcher has to adopt the best way to assist the patient.

Certain commonalities exist between qualitative and therapeutic interviews. However, researchers and therapists have different aims; the researcher’s aim is to gain knowledge while the therapist’s aim is to assist in the healing process. Several studies have shown, however, that qualitative interviews might be beneficial for the participants, especially after they have gone through a traumatic experience (for instance, Colbourne and Sque, 2005 among others). Kvale (1996) argues that among other elements of interviews, interaction with others and remembering the past might be therapeutic.