Sociology: Interviews

Sociological research employs two main types of interviews. These are:

• Structured interviews 

• Unstructured interviews (included groups interviews)

Sociologists sometimes also use semi-structured interviews, combining elements of both 

The structured interviews involves the face-to-face or over-the-phone delivery of a questionnaire. Structured interviews use an interview schedule- a pre-set list of questions designed by the researcher and asked of all interviewees in the same way. Interviewees then choose from a limited list of possible answers. Structured interviews are usually relatively brief 

Positivists start from the assumption that there is a measurables objective social reality. They take a scientific approach using standardised methods such as structured interviews to obtain quantitative data. Structured interviews employ fixed lists of closed-ended questions, so answers can be classified, counted and quantified. This allows the researcher to identify pattern and produce generalisations and cause-and-effect statements 

Positivists also prefer structured interviews because they are reliable and can produce fairly large-scale, representative data. however, interpretivists reject their use because they see them as lacking validity

Because they use a fixed list of questions, with pre-coded response categories, structured interviews produce easily quantifiable data. As every interviewee’s response is measured in the same way, structured interviews are a form of standardised measuring instrument. This means that the data from the different interviews is directly comparable 

Using the same fixed list of questions and possible answers also means that the structured interview is replicable and can be used to verify the results of earlier interviews or to identify changes over time  

compared with unstructured interviews, structured interviews are relatively quick to conduct. This means that a larger sample can be interviews, which is likely to produce more representative results, allowing the researcher to make generalisations. Structured interviews have a higher response rate then mailed questionnaires and this also helps with representativeness

Because they are quick to complete, structured interviews are the cheapest form of interview. Interviews also need only limited training 

Face-to-face interviews mean  a higher response rate is likely, because the researcher’s presence means that the research’s purpose and importance can be explained to potential interviewees

Interviewer effect occurs when the interview’s presence affects the interviewee’s responses and reduces validity. However, with structured interviews, interviewer effect will be far less that with open-ended, free-flowing unstructured interviews. This is because, in structured interviews, contact is limited to asking and responding to a fixed list of questions and possible answers. Some sociologists also try to minimise interviewer effect by a ‘deadpan’ delivery, showing no emotion, not stressing any wording etc    

Add extra evaluation by comparing interviews with another method such as observation or by comparing different types of interview

Interpretivists challenge the validity of data produced from structured interviews 

• Because the researcher decides the question in advance, and because the interview schedule cannot be altered once it has been finished, this prevents interviewees from raising new issues.  The researcher has limited what the respondent can talk about and this may exclude important aspects of the researcher issue   

• fixed-response questions may prevent the interviewees from saying what they really think- they have to force their respondents to fit the researcher’s categories. Anything that limits how the interviewee responds reduces the validity of the data generated 

• The wording of some questions may be open to interpretation and there is a limit to how far the interviewer can explain their meaning to the interviewee

• It is difficult to know if the respondent is being truthful. Although this is a problem with all interviews, it is particularly so with structured interviews, since the researcher usually cannot move away from the fixed interview schedule to check the answers they are given

• Structured interviews are not completely free from the interviewer effect. The interviewee may interpret the interviewer’s social characteristics- the age, gender, class etc- in ways that may influence their response

Interviewer effect may also reduce reliability because, although the same questions may be asked, interviews are going to have different social characteristics. Thus the interview will not be exactly replicable 

Employing and training interviews incurs a cost and even relatively brief structured interviews are not as cheap to carry out as mailed questionnaires

Asking someone a fixed list of questions in a deadpan manner can be quite off-putting and not particularly useful for investigating sensitive issues where a rapport is needed  

Positivists consider the quantitative data generated by structured interviews to be much more useful than the qualitative material that emerges from unstructured interviews. This is particularly true when the researcher wants to obtain basic factual information or gain some idea of general patterns of attitudes and behaviour. Feminists argue that structured methods are patriarchal as the interviewer not the female interviewee is in control, making it difficult for women to express their experience of oppression    

• Ask mainly open-ended questions, with no fixed set of questions to be asked of every  respondent 

• Produce qualitative data because the interviewee can respond in words that are meaningful to them  

• Are guided as much by the interviewee as by the interviewer

• Are informal and free-flowing, and more ‘normal’ than a structured interview- more like a guided conversation 

• Build a stronger relationship between researcher and research subject

Interpretivists seek to discover the meanings that underlie our actions and this means using open-ended research methods that produce valid, qualitative data. They prefer unstructured interviews because they give people the opportunity to talk openly, unrestricted by a fixed list of questions and possible responses. As a result it is likely that their meanings and worldview will emerge more clearly 

However, positivists reject unstructured interviews because their lack of standardised questions and answers means that reliable, quantitative data cannot be generated

Interpretivists claim that the main advantage of unstructured interviews is that they create data that is high in validity, for several reasons:

• The informal, conservational nature of unstructured interviews means that trust and rapport can develop between interviewer and interviewee. The more comfortable an interviewee feels, the more likely they are to ‘open-up’. This often helps when researchers sensitive issues and increase the chances of getting full and honest responses

• They avoid the danger of the sociologist imposing their ideas onto the interview process. With no set questions or fixed response categories, interviewees have the opportunity to reply in their own worms in ways that are meaningful to them. Unlike with structured research techniques, the interviewee has the opportunity to raise issues they think are important 

• The flexibility of unstructured interviews also adds to validity. The interviewer can follow up any issues raised by the interviewee, probing more deeply to get a truer picture 

• Open-ended questioning allows interviewees scope to give detailed, in-depth reactions. The more detailed the response, the greater the likelihood of the sociologist understanding the research subject’s worldview 

Unstructured interviews may not achieve the degree of validity that is sometimes claimed for them 

• The closer interviewer-interviewee bond may increase the chance of the respondent seeking to please by giving the answer they think the researcher wants to hear 

• There is also the issue of what to do with the huge amount of data that unstructured interviews produce. The researcher has to interpret it and be selective about what is presented in the final research report. In doing so, the researcher’s own perspective may distort the interviewee;s original meanings 

Many students assume that unstructured interviews are ‘valid’ but it can be claimed that some aspects of this approach reduce validity, so point out these arguments 

Positivists argue that unstructured interviews lack reliability. No fixed list of pre-set questions, plus the ability of interviewees to respond in any way they wish, makes it impossible to classify and count their responses. This means statistical evidence cannot be created from informal interviews, preventing comparisons being drawn or correlations established 

Unstructured interviews take longer to carry out and this usually limits the size of the research sample. Because it is harder to obtain a representative sample from a small number of respondents, this may limit the ability of the researcher to make generalisations

When answering questions about sensitive issues, some people may prefer to fill in an anonymous postal questionnaire rather than answer probing questions asked face-to-face

When you point out a disadvantage, you can briefly attach a strength to it as evaluation, e.g. ‘Although people might prefer the anonymity of questionnaires, unstructured interviews build trust and allow more in-depth questioning 

Interviewers need to be trained in sensitivity, the purpose of the research, how much to explain to interviewees etc. This adds to the cost, as does the fact that it takes longer to carry them out and to process the data they create. As a result, unstructured interviews are more costly than structured interviews

The interviewee may wander off into all kinds of areas that are irrelevant to the research. Unless the interviewer is prepared to redirect the respondent- with a possible loss of validity0 a lot of time may be wasted and irrelevant data collected

These are usually largely unstructured and involve interviewing a group of people together. This can help jog memories, stimulate answers and suggest lines of enquiry, but there is a danger that individuals will offer conformist answers rather than say what they really think   

Interpretivists have a strong preference for unstructured interviews, particularly when researching sensitive issues. however, unstructured interviews have some difficulties in achieving the degree of validity they seek

Structured and unstructured interviewing techniques do not have to be seen as irreconcilable opposites. They can be used in a complementary way in the form of semi-structured interviews to produce both quantitative and qualitative data

• A structured approach with standardised questions can be used to gather basic factual information about age, occupation, income, etc

• An unstructured approach with open-ended questions can be used when the researcher wants to explore interviewees’ meanings and attitudes in depth