Issues covered include the importance of empathy and rapport, listening and questioning, restatement, clarification and persistence.
Researchers should be aware of the powerful influence of these factors on the responses of interviewees. They should take steps to ensure quality data is obtained by using appropriate interview techniques and suitable interviewers.
In the investigation, the intention was to obtain the participants' views on the phenomena under investigation. The interviews enabled us to gain explanations and information on material that is not directly accessible: perceptions, attitudes and values, matters which are difficult to obtain by alternative methods. The kind of interview process employed was described by Hitchcock and Hughes (1989, p. 83) as the "semi-structured interview",
which allows depth to be achieved by providing the opportunity on the part of the interviewer to probe and expand the interviewee's responses. ... Some kind of balance between the interviewer and the interviewee can develop which can provide room for negotiation, discussion, and expansion of the interviewee's responses.The advantage of the semi-structured interview is that the interviewer is in control of the process of obtaining information from the interviewee, but is free to follow new leads as they arise (Bernard, 1988). In the study (Partington, Waugh, & Forrest, submitted for publication), the researchers were concerned to obtain accounts of events and attitudes towards those events when students were sent out of the classroom by the teachers for misbehaviour. A common set of questions, based on a review of literature on discipline, provided a basic framework for examining the phenomena and the accompanying attitudes but, given the diverse nature of the different events, freedom to move beyond the basic set of questions was essential.
The interviews were conducted at the school where the events took place. Efforts were made to ensure the physical context was conducive to effective interviews (Burgess, 1988). The interviews were conducted either in the Aboriginal Education Specialist Teacher's office, which was familiar and supportive territory for the students, or in a small room in the library. Students were interviewed individually following disciplinary incidents. Interviews were recorded and transcribed. The specific nature of the interviews is described in more detail below.
A number of assumptions can be made about the semi-structured interview. First, the interviewee has information that the interviewer wants. The interviewer is seeking to place him or herself in the shoes of the interviewee - to interpret a situation from the viewpoint of the participant. At the commencement of the interview, the interviewer and the interviewee do not share an understanding of the phenomena under investigation and the interviewer's comprehension of, and attitude towards, the phenomena are possibly very different from those of the interviewee. It is likely, however, that the interviewer can only ever come to a partial understanding of the interviewee's viewpoint, partly because the interviewee will have complex and contradictory perspectives and partly because it is not possible to fully encompass the experience of another person.
The interviewer initially constructs a frame within which the interviewee responds. By the nature and direction of the questions the interviewer creates meaning from the interviewees responses. This is an essential component of an interview, as McKeown and Freebody (1988) have shown. Rather than being a one-way construction however, respondents are likely to seek information from the interviewer to discover the parameters of the interview, a process that continues throughout the interaction. Depending upon the breadth of understanding of the interviewer, the meaning that is created may be quite partial or strongly biased in certain areas. For full comprehension of meaning, a dialogue in which the interviewee feels free to construct context and events is necessary. In this situation the interviewer must be alert to possibilities and ask questions which point in directions but do not limit the nature of the answer. It should not become a case of the interviewer trying to prove his or her point by selecting limited evidence from the interviewee's reality.
As Fine (1994) pointed out, the interviewee (the Other) does not have one location, one perspective, one reality. There are multiple perspectives and multiple realities, and the positioning of the interviewee cannot be limited. As a consequence, the interviewer is limited by his or her experience and knowledge. If the interviewer excludes the possibility of diverse responses and seeks confirmation of previously held notions, knowledge will not be advanced. In researching children, this can have the consequence of constructing a partial or ideal view of the child. Through our questions we can create a framework that forces the child to respond in ways which result in a construction of our own making rather than representing the realities of the child. The view that is created may be conscious or unconscious, but the interviewer has the power to construct such a view and to exclude contradictory or alternative views from surfacing. The interviewer has to be sensitive to the cognitive and social framework within which he or she is working when interviewing children, for it may be too sophisticated for them to understand. Under this circumstance a child may feign understanding and provide answers that seem to satisfy the interviewer.
At issue in the interview is the responsibility of the interviewer to clarify the factors influencing the interviewee. If the latter is in an oppressed condition, it is unethical for the interviewer to ignore that condition and leave him or her in it. The interviewer has an obligation to the interviewee to provide critical awareness through the research, thereby empowering the interviewee. This transformation should be an outcome of the research, and while it may not occur during the process of data gathering, it should be effected when feasible. For example, if a child complains of racist treatment from other students, it would be appropriate, at the end of the interview, for the interviewer to discuss with the child what he or she should do about it.
The interviewer should minimise interruptions when a participant is talking. Providing supportive nods, agreement and so on is more appropriate than excess verbalisation which may distract the respondent and lead him or her in unproductive directions. As Keats (1993) notes, it is worthwhile saying little when a participant responds as it may lead to statements that are choice morsels of information.
The empathy evident in the interviewer's supportive statement in this interview was typical of her approach to students. They responded openly and apparently quite honestly about the events that happened and their attitudes towards them. Without making it obvious, this interviewer gave subtle messages to the respondents that she supported them and was sympathetic to their views. Judgment was suspended when discussing misdemeanours so the respondents did not feel threatened in the interviews.
|I||You, when you needed to go to the toilet then ...|
|Harry||Yea, it was just between me and her.||No other students were involved.|
|I||Couldn't you have held on till the end of the period?||Normally, students were not permitted to go to the toilet during class time.|
|Harry||There was a half an hour till the end of the class.||It was too long to wait.|
|I||That is a long time when you have to go.||Empathy from the interviewer.|
|Harry||And I had already waited for ten minutes.||Respondent acknowledges the support by adding supportive evidence of his difficult situation.|
In this study, the two assistants who conducted most of the interviews were carefully prepared to conduct interviews. A third interviewer, however, used at short notice because the others were unavailable at the time, produced data which were quite different compared with the other interviewers' data. This was despite being verbally instructed on the steps in the interview process. This interviewer's participation may have been less successful because, although Indigenous, rather than being a parent of students at the school, he was a qualified teacher (but not teaching at the time). Lack of rapport was evident in the interviews he conducted, both in tone of voice, which unfortunately can't be reproduced here and also in the way the questions were asked. In particular, his responses to the students' statements were judgmental and he didn't use empathy in his interactions with them. This is clear in transcript II.
While gender may have been a factor in the contrast between this interviewer's results and the others, it is more likely that the discourse of the interview was the principal factor. The interviewer focussed on the student's behaviour rather than exploring the exact course of events. Had he been more alert, he would have found that this student did little to warrant being sent out, but subsequently escalated the misdemeanour by throwing the gumnuts. This is discussed in more detail later.
Keats (1993) noted that the interviewer should establish his or her role at the outset of the interview. Unfortunately, in the above case the interviewer continued to portray elements of the teacher role, clearly siding with the school administration by seeking to define the behaviour as wrong. When the relationship is an unequal one, as in the case of an adult interviewing a child, care must be taken to avoid undue influence and thereby contaminate the response of the child. The child may be seeking to identify the kind of response the interviewer appears to want and by failing to utilise appropriate empathic strategies the interviewer may obtain answers that are inaccurate. Alternatively, the child will give up and provide monosyllabic responses.
|I||Can you tell me what happened?|
|Colin||It happened on Thursday. I was muckin' round and that and then I got sent to another teacher, I was throwing honkey nuts and that and then I got sent to Mr Clarke and he put me in contract.|
|I||Okay, so you were just throwing things around were you?||Notice that the interviewer misses some vital pieces of the jigsaw. The implication is that by throwing things the child is infringing school rules and so has established his wrongdoing.|
|I||And what happened?|
|Colin||I was sent to Mr. Clarke.|
|I||So who caused it?||Seeking blame|
|I||So you caused it did you?||Confirming blame.|
|Colin||Muckin' around in class and talkin' and that.|
|I||What were you doing when you say you were mucking around? What sort of things were you doing in class?||Goes back to the core of the situation.|
|Colin||Talkin' and shoutin'.|
|I||Shouting too were you?||More evidence of wrongdoing.|
As stated above gender may have played a part in the quality of the interviews. It may not have been simply the status of the interviewer as teacher that limited the above exchange but also the fact that he was a male. In their review of interviewing as a method of data gathering, Fontana and Frey (1994) noted the influence of male power structures, an issue that certainly applied in the research school. The presence of female interviewers who were better able to empathise and identify with the respondents' backgrounds and experiences enabled greater involvement of the students in the interview.
|I||And what were you supposed to be doing at the time that you got ..?||Finding out what the teacher expected of students at the time.|
|Norman||Looking at the TV, watching this video.|
|I||What sort of video was it?|
|I||Tarzan. You weren't too interested?||Makes a prior judgment about the respondent's interest. Should have asked, "What did you think of it?"|
|Norman||It's good really.|
|I||You didn't really want to watch it.||Not listening.|
|Norman||Yes, I wanted to watch it.|
A problem with inattentiveness or pre judgment is that it can lead the respondent to believe that the interviewer is either not interested in what is being said or does not care for the respondent's point of view. In either case, the respondent may decide to terminate the interview, withhold valuable data or fabricate responses.
|I||Can you tell us like why you get on with, okay with the teacher?|
|Doug||This is when I'm not sitting near my mates because she knows I do my work with no-one near me, to distract me but when there's someone sitting next to me you start to talk and she gets mad and if we keep on going she gets mad.|
|I||So that when ever you're sitting with someone you tend to talk. And get drama for it? When you're on your own you don't talk so much? What do you think the teacher should have done rather than send you out of the class? Do you think she should have done something else or do you think that being sent out of the class was fair?||Restatement. Note the implied non-verbal responses from the interviewee here. Also note the use of vernacular - "drama". Also the interviewer has extrapolated from the interviewee's response. This is acceptable. It shows to the interviewee that the interviewer is not stupid. [Earlier in the transcript, the student had indicated that she believed that she shouldn't have been sent from class]|
|Doug||It wasn't that serious. I don't know why she made us go out of the classroom because of the talk, she started it, she was saying oh why not come to school, and I was just answering back to her tellin' her we have to, and she goes, Oh, bein' smart now and she sent us out.||The student confides significant information to the interviewer. This is a significant statement by the student because this material is potentially very damaging to him if it got back to the teacher. It is clear he trusts the interviewer.|
The respondent in this interview is much more forthcoming with views than the respondent with the teacher-interviewer. The absence of judgment, the willingness to clearly set the parameters of the interview and the evident early rapport all contribute to an effective interview.
Restatement allows the interviewer to demonstrate that he or she has listened to the answer. In the above interview the restatement of the respondent's answer provided a base from which to launch further questions which clarified the response. Also note the use of vernacular ('drama')which would cement the relationship with the student. Much of the vernacular used by the Aboriginal interviewers was drawn from Aboriginal English and so the problems of lack of inclusion identified by Hosie (1986) would not occur. Rather than being seen by the students as outsiders trying to act as members of a student group, the adults would be seen as bona fide members of a non-school group. Being Indigenous, the interviewers would understand the experiences of the students in a way that non-Indigenous people could only guess at (Partington, Godfrey, Harslett, Harrison, & Richer, 1998). As a consequence they are able to interpret responses in ways that are consistent with the meanings expressed by the students.
Notice also how the interviewer extrapolated a judgment from the response: being sent out of the room was too harsh a penalty for the offence committed (from data obtained in the interview with the teacher, it was clearly the teacher who was at fault). As a consequence, the interviewer asked what should have been done instead. This approach would have shown the student that the interviewer was alert to the situation and aware of appropriate action: it would have been clear to him that the interviewer was not biased in support of the school.
|I||You gave me the steps which led up to this teacher sending you out. After you got sent out you said that you were sent to another teacher. And that you were throwing honkey nuts at each other on the way.||Note the importance of recapitulation. By restating the evidence, the interviewer clarifies the information. In this case, she had it wrong.|
|Colin||No. I was throwing honkey nuts at these boys so they could put it in the bin because they were picking up honkey nuts.||The interviewee corrects her.|
|I||Was this before you went to the ..?|
|Colin||This was on our way to [the other teacher].||More clarification.|
Clarification obviously was valuable here because it cleared up a misunderstanding and provided valuable data that contributed to the research. Seeking clarification does not have the same import for the participant as failing to listen: when you can itemise the steps correctly, the former process shows the participant that you have been listening carefully. Even if you get it wrong, the ability to construct a sequence of events from the interview shows you have been listening, although not necessarily comprehending.
|I||So do you get sent out from Social Studies, um before or?|
|Colin||Ah, yeh, I've been sent out lots of times before.||The question elicited the anticipated response.|
|I||So do you think you were, you were deliberately, did you deliberately go in there to disrupt the class or yell at the teacher, or?||Interviewer tries to pose the question in a non-judgmental way.|
|Colin||When I, when I come back inside, um, when I come back inside I got straight on with my work.||The student doesn't answer the question.|
|I||And you really think that you were the, well you did wrong by yelling, that you were the right person that was sent out.||Interviewer has another go at getting the information by rephrasing the question.|
|Colin||Um.||The student doesn't answer the question.|
|I||But there were other kids involved.||Interviewer elaborates on the question.|
|Colin||Yeh, there were um, probably at the time I was the right person to be sent out but um, when the teacher saw em, um, they should of got sent out too.||The student now answers the question as to whether he was responsible. The additional questions clarify the question. He sees it as a case of him or others now.|
The interviewer has to simultaneously reflect on the information being provided in order to relate it to prior information; plan the next question; decide whether to pose the question or make provision for the respondent to answer it in his or her own time. Most importantly, the respondent has to be convinced that the interviewer is not an adversary but is at the very least impartial. Using a variety of skills, the interviewer can develop a positive relationship with the respondent and obtain quality information. Failure to attend to these basic principles of interviewing will diminish the quantity of data obtained. In contrast to interviews conducted by good interviewers, poorly conducted interviews will be characterised by a lot of interviewer talk and little respondent talk.
Fontana and Frey (1994) refer to interviewing as 'the art of science'. There is certainly much of art in the process of obtaining sound data through the interview. Many of the factors that contribute to the quality interview, however, can be developed through careful attention to a range of skills such as careful listening and responding as well as skill in the development of rapport and empathy.
Burgess, Robert G. (1988). Conversations with a purpose: The ethnographic interview in educational research. In Robert G. Burgess (Ed), Studies in Qualitative Methodology Vol. 1: Conducting Qualitative Research. Greenwich Connecticut: JAI Press.
Fine, Michelle. (1994). Working with hyphens: Reinventing self and other in qualitative research. In Norman K. Denzin and Yvonna S. Lincoln (Eds), Handbook of Qualitative Research. Thousand Oaks: Sage.
Fontana, Andrea & Frey, James H. (1994). Interviewing: The art of science. In Norman K. Denzin & Yvonna S. Lincoln (Eds), Handbook of Qualitative Research. Thousand Oaks: Sage.
Hitchcock, Graham & Hughes, David (1989). Research and the Teacher: A Qualitative Introduction to School-based Research. London: Routledge.
Hosie, Peter (1986). Some theoretical and methodological issues to consider when using interviews for naturalistic research. Australian Journal of Education, 30, 200-211.
Keats, Daphne. (1993). Skilled Interviewing. (3rd ed). Hawthorn, Vic: ACER.
McKeown, Gerry & Freebody, Peter. (1988). Aboriginal children's oral discourse: A comparison with school texts and tests. In Graham Davidson (Ed), Ethnicity and Cognitive Assessment: Australian Perspectives. Darwin: Institute of Technology.
Moyser, George. (1988). Non-standardized interviewing in elite research. In Robert G. Burgess (Ed), Studies in Qualitative Methodology Vol. 1: Conducting Qualitative Research. Greenwich, Connecticut: JAI Press.
Partington, G., Godfrey, J., Harslett, M., Harrison, B., & Richer, K. (1998, November). Aboriginal and Islander Education Workers (AIEW) perceptions of schooling. Paper presented at the Australian Association for Research in Education, Adelaide.
Partington, G., Waugh, R., & Forrest, S. (1995). Interpretations of classroom discipline practices by teachers and Indigenous secondary school students. Submitted for publication.
|Authors: Gary Partington is an Associate Professor in Kurongkurl Katitjin, the School of Indigenous Australian Studies at Edith Cowan University where he coordinates research.|
Address: Edith Cowan University, 2 Bradford St, Mt Lawley WA 6050 Australia
Please cite as: Partington, G. (2001). Qualitative research interviews: Identifying problems in technique. Issues In Educational Research, 11(2), 32-44. http://www.iier.org.au/iier11/partington.html