Back
to materials page . The advice given here
is based on advice we have received from departmental colleagues (in particular
David Barton, Greg Myers and Jay Banerjee), reading the items listed in
the references list at the bottom of this page and discussions with various
MA students who helped Mick in his pilot investigation.
We suggest that you prepare carefully for your interviews, reading as much
as you can about interviews beforehand and planning carefully when, where
and how you aim to conduct them. After reading the advice below and any
other material and advice you manage to get hold of, it is a good idea to
do one or more pilot interviews, and then review them, to make sure that
you forestall as many problems as possible in your main study. The advice
below may seem ridiculously obvious. But in our experience it is very easy
to miss obvious things (for example Mick forgot to switch the tape recorder
on in the first pilot interview he conducted!)
Ethical issues. Note that there are ethical issues
involved in collecting interview data from students. You need to consider
them carefully in advance and make sure that you and your students are happy
with what you are doing. You should ask their permission for interviews
in advance, and also make it clear that you intend to tape-record the interview
and make notes, so that you can analyse the data produced. It is helpful
to anonymise the collection of data as much as you can (see 7 below) to
protect individual students and their feelings. Having interviewers who
are not the students' teachers also makes a big difference to anonymity.
You should tell the students in general terms why you are conducting your
research (to investigate attitudes to web-based and traditional learning
modes in stylistics). But try to keep the description as non-specific as
you can, so as not to alarm them or push them towards producing particular
answers to your questions later on, and so invalidating your research.
Lancaster University have an online Ethics
Resource Centre which has further details relating to ethical issues
and guidelines for researchers.
What kind of interview? First decide on the kind of interview you intend to conduct. Of the three kinds of interview referred to in the literature (structured, semi-structured and unstructured), we have found the semi-structured interview most useful for our needs. The structured interview is the kind of interview you are sometimes asked to do in the street by market researchers. The questions are all worked out exactly in advance and everyone is asked exactly the same questions in order to get reliable statistical information. They often involve the interviewer in ticking of lots of boxes, and are quite like a verbal questionnaire. This kind of interview is not flexible enough to reveal the kinds of thing we are trying to find out, which involves exploring in some detail student reactions to the educational experience they are having. On the other hand, the unstructured interview, which is a kind of free-flowing conversation in which the interviewee has quite a lot of control over what is discussed does not make it easy to compare the responses of different students to the same issues. The semi-structured interview involves planning in advance the areas you want to cover and the general questions you want to ask, but then being reasonably relaxed about the order topics are covered, the amount of depth individual interviewees feel the need to go into on particular areas and exactly what questions you ask (but see 9 below on questions). The semi-structured approach allows the uncovering of in-depth responses while making sure that similar things are covered in equivalent interviews, and, unlike the structured interview, also allows for the possibility of exploring, as they arise, new and important issues you had not thought to ask about in advance.
Choosing a room and a time for the interview to take place. You need somewhere to conduct the interview which will be quiet, where the interviewee will feel relaxed and where you will not be disturbed. Choose somewhere as neutral as possible. Don't invade the private space of the interviewee. Try to find somewhere the right size and with the right degree of formality/informality. Make sure that the telephone and other possible distracters are disconnected before the interview takes place, and that you have put a 'do not disturb' notice on the door. Clearly you need to arrange a time convenient both to you and your interviewee. The interviews we have conducted have typically lasted 30-45 minutes, and it is helpful to indicate to students how long they are likely to be involved. However, you should allow more time in case the interview runs over, and you will also need some time in advance to prepare the room, the recording equipment etc. before the student arrives and some after the interviewee has left, too. We recommend that you allow at least an hour for the interview itself, plus 15 minutes beforehand to set everything up, and 10 minutes afterwards to clear up, make sure you have labelled recordings and so on.
Who should conduct the interview? You may have no choice but to do the interviews yourself. But, if possible, it is a good idea to have an interviewer who is not teaching the students, if only for reasons of anonymity (see 7 below). This will put interviewees under less pressure. Your MA students (who will need to be trained by you in interviewing techniques, however) are a good possibility, if you have them. They might then be able to use the material they gather towards a project or dissertation, in which case they can probably be persuaded to transcribe for you (and themselves) the data they collect. Even if you have to conduct some interviews yourself, it is a good idea to use additional interviewers if you can.
Putting interviewees at their ease. You will get better results if your interviewees are relaxed. This is best achieved by (a) consulting them in advance, (b) arranging a relaxed and non-threatening place in which to hold the interview (see 3 above), (d) trying as much as is possible to reassure them and make the data collection anonymous (see 6 below), (e) creating an informal atmosphere for the discussion itself and (f) trying to make the questions you ask non-threatening (see X below).
Student anonymity. Students are more likely to be honest and forthright if they feel that their comments are anonymous. In the Lancaster pilot investigation, we asked each student to invent a private nickname for themselves at the beginning of the course, which they each wrote on the questionnaires they filled in, and by which they could be identified in the interviews and discussion groups. This allowed us the possibility of comparing the same student's questionnaires and interview comments without revealing the identity of the student. Having interviews conducted by interviewers who are not the students' teachers also clearly aids anonymity.
Recording the data. Clearly you need a good tape recorder, and if possible a separate (and preferably small) microphone. This means, apart from anything else, that you can put the microphone on the table, but not the recorder, thus reducing the 'being recorded' feeling. In our experience, students soon forget about the microphone and recorder if you reduce its presence in this way and keep the conversation relaxed, particularly at the beginning. Make sure you have an adequate power source, both for the recorder and the microphone (back-up batteries are a good idea!), and test them out beforehand. Don't forget to switch on both the microphone and the record button on the recorder before you start recording. And don't forget to switch them both off afterwards - otherwise you may find that you microphone battery, in particular, fails in the next interview! The interviewer could, in theory, take 'field notes' instead of recording the interview. But this tends to distract the interviewer from planning the structure of the interview as it proceeds and from thinking carefully 'on-line' about the form of the questions to be asked. In any case, the field notes will not be as detailed and reliable as a transcript of a recording.
Planning the form of an interview. You should consider
whether or not it would be helpful to have a computer linked to the website
available, so that you can discuss with the interviewee issues which arise
in relation to particular web pages (we did this in our pilot, but, as it
turned out, it wasn't used an awful lot). You also need to think about the
overall structure of the interview, and you will find it useful to have
a sheet prepared in advance to refer to with your outline structure and
areas to question (possibly with some example questions). This will help
you not to miss things and to have enough time to think about the formulation
of questions as you go along. Remember, though, that you do not have to
cover the various areas and questions in the exact order on your plan. Be
flexible, letting the interviewee take the lead as much as possible, but
use the plan to help you to make sure you do cover all the areas you need
to, and to help you bring the interviewee back to the purpose of the interview
if he or she strays off into other, non-relevant areas.
It is important that you remember to say at the beginning of the interview
the date of the interview, the name of the interviewer and the nickname
of the student, so that you have them recorded for reference. Try to start
with some reasonably general, open-ended and non-threatening questions in
order to settle the student down. You can move into more specific questions
once you feel that the student has relaxed.
Questions. It is important that your questions do
not push the student towards giving the answers you prefer. So don't ask
"leading questions". For example, in a mid-course interview, where
the students have finished the poetry section of the course and have now
done part of the prose section, the question 'Has your enjoyment of the
course increased now we have finished the poetry section?' leads the interviewee
to say 'yes'. It suggests that the interviewer wants to be told that the
student was enjoying the poetry section of the course, but not as much as
he/she is enjoying other prose section. And there is plenty of evidence
in Psychology to suggest that interviewees tell interviewers what they think
they want to hear. So 'yes' become an almost guaranteed answer to the above
question. We often use noun phrases and embedded clauses in questions to
signal what we want to hear. Tag questions are another way in which we signal
what we want to hear (cf. 'You liked that, didn't you?'), and there are
many other ways too. Try as much as you can to guard against 'leading' your
interviewee in this sort of way. The above question is also a 'closed' question,
inviting a yes/no answer. It would be better to ask a series of open questions,
helping the student to give fuller, more thought-out answers, and taking
one aspect at a time, e.g.:
(i) 'What, if anything, did you enjoy in the poetry section of the course?'
(ii) 'What if anything, did you dislike in the poetry section of the course?'
(iii) 'Now that you have moved onto the prose section, what, if anything
do you like about that section?'
(iv) What if anything do you dislike about the prose section so far?
And even then, it will be important to follow up on the student response
to each of the questions, to make sure that important things are not missed.
We have provided some information regarding the kinds of structure and questions
we have found useful in interviews at the beginning, in the middle an at
the end of the course. it is important that you only treat these as a rough
guide though. It is important that you get students to talk about the particular
factors which are important for them in their context, whatever they might
be. See interview questions for
further details.
Interview sets. Depending on the size and composition of your course, it might be worth structuring your interviews (and focus groups) into sets. Let's pretend, for example, that you have both native speakers and non-native speakers, or both young students and mature students in your course. If there are enough of each group, it might be worth collecting interview and focus group material from each subset at each stage, so that you can compare the responses of the subgroups to see if they differ in any significant and interesting ways.
There are large numbers of books and articles about qualitative research methods, including interviewing. We have found the following useful:
Becker, H. S. (1970) 'Interviewing medical students'. In W. J. Fistead (ed.), Qualitative Methodology, Chicago: Markham, pp. 103-6.
Burns, R. B. (2000) 'Unstructured interviewing'. In R. B. Burns, Introduction to Research Methods (4th edition), London: Sage, Ch. 24, pp. 424-41.
Burroughs, G. E. R. (1971) 'Data collection - The interview'. In G. E. R. Burroughs, Design and Analysis in Educational Research, Birmingham: School of Education, University of Birmingham, 99-101.
Cannell, C. F and R. L. Kahn (1953) 'The collection of data by interviewing'. In L. Festinger and D. Katz (eds.) Research Methods in the Behavioural Sciences, New York: Holt, Rinehart and Winston, pp. 327-80.
Cohen, L and L. Manion (1994) 'The interview'. In L. Cohen and L. Manion, Research Methods in Education (4th edition), London: Routledge, Ch 13, pp. 271-98.
Denscombe, M. (1998) 'Interviews'. In M. Denscombe, The Good Research Guide, Buckingham: Open University Press, Ch. 7, pp. 110-38.
Fontana, A. and Frey, J. H. (1998) 'Interviewing: The art of science'. In N. K. Denzin and Y. S. Lincoln (eds.), Collecting and Interpreting Qualitative Materials, Thousand Oaks, Calif.: Sage, pp. 47-78.
Oppenheim, A. N. (1992) 'The exploratory interview'. In A. N. Oppenheim, Questionnaire
Design, Interviewing and Attitude Measurement (2nd edition), London: Pinter.