Interviews

An interview is a self-report method using a structured, semi-structured or unstructured approach.

Interviews are becoming increasingly popular as a research tool. Since the earliest day of psychology, the interview has been seen as a useful way of collecting data but the method was not seen as ‘scientific’ by the behaviourist school, although even behaviourists acknowledged that what they called ‘verbal behaviour’ sometimes provides useful research data.

An interview occurs when a participant is asked questions that have been designed to elicit particular types of information. Interviewing is a skilled activity, because of its interpersonal nature. People are very good at reading non-verbal signs, i.e. the small changes in behaviour that indicates how information is being received. These are often unconscious on the part of the interviewer, but they can have a profound effect on the way people respond. In a face-to-face situation, for example, most people like to be co-operative, and to avoid interpersonal conflict. So an unconscious indication from an interviewer that they disagree with what the person has said, or with a particular option in a question, can lead a respondent to change their answers to something that they feel is more socially acceptable.

Any social situation also means that people will bring their wider social knowledge and habits into the situation, and this can affect the answers they give. It is not uncommon that people respond differently to male interviewers than they do to female interviewers. Age and ethnicity may also have an effect. Basically, people adjust their responses according to what they consider appropriate for the person asking the questions.

Interviewer effects of this kind are something, which need to be taken account of in any interview study. If the study is concerned with obtaining representative views from a large number of people, then it is more likely that more than one interviewer is needed, and then it is appropriate to counterbalance interviewer effects by varying age, gender and ethnicity so that it is balanced out in the sample as a whole.

The interviewer asks the questions and the interviewee (respondent) answers the questions. Interviews may be formal, directive interviews and more informal interviews.

Structured (directive) interview

• The questions have to be answered in a preset order (strict schedule: same questions and same order).
• The answers can be quantified (put into numbers)
• And statistical analysis can be performed.

Advantages/disadvantages of structured interview:

• Fixed questions easier for the interviewer to conduct (so easier to conduct survey with large samples)
• More objectively verifiable and generalisable.
• A problem with the interview approach: sampling bias (voluntaries)
• Problems with validity

Problems in esp. structured interviews: lack of reliability

• When different interviewers use same structured questions (the interviewers must be trained in the same way because they need to use the same techniques in order to have a consistent approach to the respondents).
• When interviewers depart from the interview schedule, they may elicit information that biases further answers. This may produce intervening variables that can affect the data obtained.
• Inter-interviewer reliability is obtained if same training has been given to the interviewers and that they all follow the interview schedule in structured interviews. It may also help to make pilot studies to train the interviewers.

Semi-structured interview

• Focused interviews (developed to combine advantages of structured and unstructured interviews)
• Involves use of additional questions
• Gives researcher the opportunity to hear the participant talk about a particular aspect of their life or experience (questions triggers participant to talk)
• Important to take interviewer identity into consideration (to avoid bias) and important to familiarise with respondent’s cultural milieu and the status of ‘the interview’ within this milieu (an interview with a young person may invite to a more informal approach with an interviewer around the same age, and interview with a middle-age professional may invite to a more formal style).
• One of the most widely used methods of data collection in qualitative research in psychology.
• Important that interviewer is aware of linguistic variability (i.e. it is important to understand the meaning of what the respondent says e.g. in that particular context).
• Important to establish a rapport between interviewer and respondent (establish interview agenda) and observe ethical considerations (e.g. in spite of good atmosphere between researcher and respondent, the interview should not abuse the informal ambience to make the respondent reveal more than he/she is comfortable with after the event).
• Interview agenda: a relatively small number of open-ended questions that allow the researcher to identify respondent’s own ideas and terms into the interview, so that questions become more relevant to the respondent. Careful reflection on what good questions are, and good idea to restate respondent’s comments and integrate them in later questions in order to show that the interviewer is listening (active listening technique). Also good to express ignorance because a naive interviewer encourages respondent to ‘state the obvious’. It is important that the questions are meaningful to the respondent.

Different kinds of questions can be used

• Descriptive questions prompt the respondent to give a general account of ‘what happened’ or ‘what it feels like to..’, anecdotes, life histories etc.
• Structural questions: prompt respondent to identify structures and meanings to use to make sense of the world (e.g. what does it mean to your life to suffer from AIDS?)
• Contrast questions allows the respondent to make comparisons between events and experiences (e.g. Did you prefer being in that company or the other?)
• Evaluative questions are about the respondent’s feelings about someone or something. (E.g. did you feel afraid when the HIV test was taken?)

Advantages/disadvantages of semi-structured interview:

• More flexibility in the interview but still the advantages of the a structural approach
• Allows for analysis in a variety of ways because it is compatible with many methods of data analysis (discourse analysis, grounded theory etc.)
• Easier to arrange than other forms of data collection (fewer logistical difficulties to arrange a series of semi-structured interviews with a small number of participants than to design a longitudinal study

Unstructured interview (or in-depth interviews)

• More like a conversation
• Interviewer work from a list of general topics but greater freedom to explore areas of interest

Advantages/disadvantages of unstructured interview:

• Permits full exploration of ideas and beliefs, i.e. it is a more valid account of social life
• Flexible: allows interviewer to pursue interesting points
• Problems of reliability, replication and time costs
• The respondent is too much in control

One-to-One interview

An interview conducted by one person to another. This is the most common method.

Conversational interview

• The conversational interview is more of a discussion (no predetermined set of questions)
• This interview style is chosen by the interviewer mainly because they can get more information from people because they tend to be more honest and open because it is a relaxed atmosphere.
• Gives maximum flexibility to be able to pursue questioning in whatever direction appears to be appropriate
• Requires an interviewer knowledgeable and experienced in the content area and strong in interpersonal skills, since he or she will have considerable discretion in directing the interview.

Advantages/disadvantages of conversational interview

• Conversational interview is highly individualized and relevant to the individual.
• Likely to produce information or insights that the interviewer could not have anticipated.
• Generates less systematic data that is difficult and time consuming to classify and analyse.
• Since different information is collected from different people, this kind of interview is not systematic or comprehensive, and it can be very difficult and time-consuming to analyse the data

Small group interview (focus group)

• Focus group discussions (group depth interviews) among the most widely used research tool (takes advantage of the interaction between a small group of people).
• Participants will respond to and build on what others in the group have said: synergistic approach believed to generate more insightful information and encourages discussion participants to give more candid answers.
• Focus groups are further characterized by the presence of a moderator and the use of a discussion guide.
• The moderator should stimulate discussion among group members rather than interview individual members, that is to say every participant should be encouraged to express his/her views on each topic as well as respond to the views expressed by the other participants.
• In order to put focus group participants at ease, the moderator will often start out by assuring everyone that there are no right or wrong answers, and that his/her feelings cannot be hurt by any views that are expressed since s/he does not work for the organization for which the research is being conducted.
• Focus groups can be homogenous (they share key features) or heterogeneous (they are different, pre-existing (friends or colleagues) or new. They can also be concerned (i.e. interested in the subject matter in question, e.g. abortion) or naive (not committed in the question).

Advantages/disadvantages of focus group interview

• Relatively easy to use and can be done quickly
• Social interaction in the group provides more free and complex answers
• Researcher can ask for clarification
• High face validity
• Requires a skilled moderator and difficult to assemble group
• Problems as to whether the group represents a larger population

Triangulation is desired in focus groups due to the large size of the groups, usually ranging from 6-10. Since such interviews are in large groups, observation of the participants could be helpful in understanding and evaluating his or her actions and replies

Email and telephone interview

Email interviews may be used when conducting an interview in person, or contacting via telephone, is inappropriate due to location, schedule conflict, or different time zones.

Telephone interview: Interviewers make phone calls to contact people (interview/survey).

Advantages and disadvantages of email and telephone interview

• Quick results and relatively inexpensive
• Problems to establish a proper sampling frame and this kind of contact may have cultural limitations (e.g. not considered appropriate to contact people privately in China)

Verbal protocols

• Also called ‘think aloud protocols’
• Participants are asked to report by thinking aloud verbally while they perform a task, i.e. verbally describing what is going through the mind while performing the task.
• Newell and Simon pioneered and championed the use of verbal protocols. They felt that the systematic collection of these types of observations could be used to test information processing models of human reasoning.
• Analysis of verbal protocols on the basis of ‘chunking’ the protocol into units (unitising) and categorisation (analysing units for usability)

Advantages and disadvantages of verbal protocols

• Many of the cognitive processes are not conscious
• Subjective verbal reports are not reliable

Clinical interview/informal interview

• Used by doctors, psychologists etc. to gather personal information
• Questions are usually not following a preset order (inspired from the respondent’s previous answer).

Advantages of the informal, non-directive (unstructured) interview

• People speak more freely + truthfully about themselves (high ecological validity),
• Clinical interview may give valuable information about the person’s mental states (diagnosis).

Recording and transcription of interviews

In order to carry out a full analysis of the data, it is necessary to audio- or video-record and transcribe the interview. Most qualitative methods of analysis require that the material is transcribed verbatim, or near verbatim. Taking notes during the interview may interfere with eye contact and non-verbal communication, so it distracts the interview. However, taping the interview may also affect what is being said, because participants are not entirely comfortable and relaxed in the presence of tape-or video recorder. It is important that the researcher explains why the recording is being made and how it is going to be used. It is also a good idea to offer the respondent a copy of the transcript of the interview if possible.

If the interview is being recorded, the researcher needs to make sure that the recorder is placed in a position where it will record clearly and so that the interviewer has eye contact with the respondent, and the recorder should be working so that the interview is not spoiled by bad technique (e.g. interruption or that you cannot hear what is on the tape afterwards).

All types of transcription constitute a form of translation of the spoken word into something else, and an interview transcript can never be the mirror image of the interview. Different ways can be used in transcription of an interview, i.e. if we are interested in the subtleties of communicative interaction between interviewer and interviewee, we must transcribe the words as well as the way in which they are spoken. This means including pauses, interruptions, intonation, volume of speech and so on. These various features of speech are represented by the signs of the transcription notation. A commonly used form of notation for such detailed transcription was developed by Gail Jefferson (e.g. (.) just noticeable pause). Detailed transcription is required for conversation analysis and some types of discursive analysis. If we are interested only in the content of the interview, we need not transcribe non-linguistic features of the interview. In this case, it is sufficient to transcribe what is being said (i.e. the words alone). This would be appropriate for grounded theory analysis . However, even here it is important to decide what to include (e.g. incomplete sentences, false starts, laughter and repetition of words) or if we want to ‘tidy up’ the transcript. It all depends on what we want to use the transcript for so basically the decision about what type of transcription to use depends upon the research question and the method of analysis chosen.

A description of the interview and transcription process from here.

In the case of face-to-face (FTF) interviewees, a cassette recorder was used to record all verbal exchanges between researcher and participants. For the first two interviews all talk between interviewer and interviewee were transcribed. Pauses, laughter, sighs and other non-verbal sounds were noted alongside the text, which was supplemented by notes made during and immediately following the interview regarding such detail as gestures used, and movements such as standing and 'displaying' (e.g. hitching trousers to show the interviewer a prosthetic leg). The rest of the interviews were repeatedly listened to, themes identified and transcribed along with accompanying non-verbal sounds. Therefore, only partial talk between interviewer and interviewee were transcribed in subsequent interviews. These partial transcriptions were also supplemented by notes taken during the course of the interview. The punctuation used in the transcripts, such as full stops and commas, was an interpretative approximation by the interviewer. While such transcription is problematic, the fact of the interviewer and transcriber being the same person is advantageous.

Here is an example of a transcription including response analysis.

Post-modern transcripts means that everything is transcribed, even utterances like aha, coughing, movements done and non-verbal communication. Post-modern transcriptions appear to be linked to social constructionism and discourse analysis.

Return to Qualitative Methodology
Return to the Homepage