Interviewing is one of the most powerful tools we have for answering research questions in the counselling and psychotherapy field—indeed, any field. ‘Interviewing’ is essentially just a technical term for ‘talking to someone’—and it’s a talking that can have a great deal of depth, nuance, and complexity. What better way of finding out the answers to your questions than by going straight to the source: people, real people, and asking them what they experience or what they think. It’s not the only way of answering research questions, and it’s not always the most suitable; but if you’ve got a question about how a particular group of people experience or perceive something, it’s one of the most effective and widely used tools.
Types of interviews
Interviews can be structured, semi-structured, or unstructured. That’s essentially the amount of a priori ordering you give to an interview: how much do you run it along pre-determined lines, with the same questions, asked in the same way, to each participant? Of course, there’s no right or wrong here—it all depends on what kind of questions you are asking. Structured interviews tend to be best for smaller q (i.e., more quantitative-like) research. For instance, if you wanted to compare how much gay and bisexual men found gay-affirmative therapies helpful, you might want to use a structured interview format so that you had clear, comparable answers to analyse. If, on the other hand, you are doing larger Q research (i.e., more in-depth, experiential, explorative) an unstructured format may be better so you can really go ‘with the flow’ in each interview and explore as deeply as possible their responses. In the middle is a semi-structured approach, which is often a good compromise, and means that you are asking every respondent similar questions, but you have flexibility to go into depth and follow up particular areas without worrying too much about ‘sticking to the script’.
Interviews can also be scheduled, unscheduled, or semi-scheduled, which refers to whether questions are asked in the same order or not. Scheduled tends to go with structured, and is more suited to research questions where its important to reduce biases. Unscheduled, by contrast, gives you the flexibility to go into depth in an interview, as and where the respondent seems to be opening areas up. Again, semi-scheduled can be a good compromise, meaning there’s a provisional order to your interview, but scope to go ‘off track’ if it feels like moving the order around will be productive.
As suggested above, a semi-structured, semi-scheduled approach is a good option—and commonly used—alongside a wide variety of qualitative counselling and psychotherapy research methods, including interpretative phenomenological analysis (IPA), grounded theory, and thematic analysis. However, if you are wanting to conduct a very large Q grounded theory or thematic analysis study, a wholly unstructured, unscheduled approach may be best.
Interview Schedule
However you do your interview, even with the most unstructured of approaches, it’s generally good to prepare a schedule or protocol of how the interview is going to go. This is absolutely essential if others are going to be doing the interview as well as you, so that they know what to do and your interview is consistent across participants.
In one of my first research studies, on therapists’ experiences of relational depth, I developed a very basic interview schedule, which consisted just a few questions and prompts:
Could you tell me about a time when you really felt that you met with a client at a level of relational depth?
What was that experience like?
What did you take from that experience?
What do you think your client took from that experience?
How did that experience come about?
What facilitates your ability to meet clients at a level of relational depth?
How important do you think that meeting clients at that level of depth is?
An example of a much more extensive interview schedule, for our study of clients’ experiences of working with preferences, can be found here. This was designed for use by myself and another interviewer; and with a team of very experienced international researchers discussing, drafting, and finalising it. As an interview schedule, however, the heart of it was still the questions to be asked. In this case, the main question was:
What, if any, preferences did you have when you came into therapy?
And then, for each preference, we asked:
Where do you think this preference might have come from?
How, if at all, did this preference change over the course of therapy?
How important was it to you that this preference was (a) elicited/discussed, and (b) accommodated in the therapy?
When writing a schedule, it may be useful to note follow-up prompts for different questions: potential responses you can use after the main question to help the participants go deeper into the exploration. For instance, for the question:
Where do you think this preference might have come from?
You might then prepare prompts like:
Was it related at all to your earlier life?
Have you had therapy before? Was it based on those experiences?
Are there any other origins you can think of?
On our schedule for the preference study, you’ll see that we also included:
Notes on the general approach/style to be taken in the interview
Materials to be taken to the interview and other general instructions
Approximate wording for the introduction to be given to participants
Additional questions
In this case, there’s also some guidance on questions that were asked outside of the scope of the main study.
Generally, there’s no harm having as much detail as possible in your schedule, even if it’s about engaging with your participants in an entirely unstructured and unscheduled way. Write it down—it’s a great reminder for you about what to do, something you can draw on when you write up your Methods section and, if you’re writing a thesis, a great resource to put in your Appendices: showing your examiners that you were really clear, consistent, and methodical in what you did.
Piloting
Before finalising your schedule, and definitely before running your interviews, try the interview out on someone—ideally a few people. And who better to start with than yourself. As well as giving you a real sense of what it’s like being interviewed with this schedule (and, perhaps, what’s working and what needs changing), it can help you unpack your biases and expectations with respect to the research question (again, something you may then want to write up and include as an appendix). Even if you aren’t part of the target group for the research study, you can role play it. Then try it out as an interviewer—perhaps with friends, colleagues, or peers—to get a sense of how it feels running the interview (bear in mind you may need ethical approval before doing this). If they are not part of your target group, again, they can always role play, though you will probably want to also pilot the interview with someone who is part of your target group before moving in to the main phase of interviewing. And remember, when you write up, you can describe this piloting process, showing your care and methodicalness in developing and testing out your interview schedule.
In fact, I generally encouraged my research supervisees to transcribe their pilot interviews, and even do some initial coding or analysis on them. That’s not suitable to all methods (particularly if you’re keen to go into each subsequent interview with an entirely open and ‘unbiased’ mindset), but it’s a great way of checking whether you’re data is making sense: that is, whether you’re actually getting meaningful answers to the questions that you’re asking. At the very least, ask yourself: ‘What have I learnt from this/these interviews (including the one with yourself) to the question(s) that I am asking?’ If the answer is ‘nothing’ or ‘I’m really not sure’, you may want to revisit your interview schedule or the focus for your project. Ideally, you’re coming away thinking, ‘Wow, those were some great initial answers to my questions… I’m hungry to learn more.’
Can you include data from your pilot interviews in your main analysis? Ideally, with most methods, not—formally, it’s an activity where you are preparing for the actual method you are going to use. But, as above, it’s great to write up your pilot interviews in your Methods section, and particularly what you have learnt from them. But some approaches, particularly larger Q ones (where there’s less focus on consistency or unbiasedness across interviews), may see less problems folding in your pilot interviews into your main data. If you’ve got really rich and meaningful data from them, why not use it?
Dialogue
Except for very structured, ultra-small q interviews, one of my main pointers for conducting interviews in counselling and psychotherapy research would be to see them as dialogues—with the aim of finding answers to the question(s) you are asking. Too often, particularly with trainees, there is a tendency to conduct interviews a bit like TV’s Question Time: with lots of questions, one after the other—and the interviewer’s nose in their interview schedule—without enough time, space, or attention to really explore the participants’ experiencing or perceptions in depth. I’m sure this often comes out of the interviewer’s anxiety: a worry about what to say next, and a concern to keep things going. But an overly-scheduled interview, with lots of quick fire questions, generally doesn’t allow the richness and depth of a participant’s experiences to emerge. People aren’t always able to answer well the first time around—to say what they really mean—so the interviewer needs empathic reflections, follow-up questions, prompts, paraphrases, summaries (many of the skills you may have learnt as part of a counselling training) to help the participant unpack what has really gone on for them. Below is an example of such dialogue from our study of clients’ experiences of preference work. The participant has already said that they came into therapy with a preference for ‘structure’ but, as an interviewer, I’m using reflections and further questions to try and understand more about what that means.
Interviewer: What would you say your main preferences were when you came into the therapy? You’ve already said about– there was one about structure, there– say a bit more about that.
Participant: ….So like you say, structure was important to me. I didn’t– I’ve never done anything like this before, so it’s completely new to me. I didn’t know what I should be talking about, or how to achieve those goals. So it was more kind of guidance, and what can we do to– that’s going to [pause] […] less anxious, um, I don’t know. I don’t know. I’m completely lost about how to get to those goals. So my structure would be what– what can we do to bring about those changes, how do we work towards that, what– um, so that at the end of the session, you know, I’d think– or I’d come back the next week and say, ‘Have you felt less anxious?’ and my initial reaction was that, ‘Well no, I haven’t felt less anxious, because why would I, nothing’s changed. I’ve done nothing to change it. I don’t know what to do to change it. Um… I’ve had no guidance’, so, yes, my- my answer to the question would be the same, it’s […] unless I’m doing something different.
Interviewer: So you really wanted to be… guided, and have something to do, say in between sessions?
Participant: Yeah, that would have been nice. Yeah. You know, I’ve never done this before, so I didn’t have any expectations. It was just, ‘Let’s see where it goes.’
Interviewer: Yeah, but you wanted– What you’re saying– Did you have that, like when you started therapy, were you thinking, ‘I’d really like some structure,’ or was it something that kind of emerged as the therapy continued?
So we don’t just move on from ‘a preference for structure’ once the participant has said it. Rather, we circle around it, try and go deeper, explore what the participant really means by it and how it links to other responses. It’s a process, all the time, of deepening understanding—again, just as you might do in therapy if you were trying to help a client make sense of particular experiences.
You can see that, in this last question, I’m trying to understand in more depth how this preference emerged. What’s going on in my head is, ‘I really want to understand what the participant wanted in therapy and why?’ That, those core research questions, are my focus, and I’m trying to engage with the participant in a range of different ways so that we can answer those questions. So, at those points, I’m not really thinking about my interview schedule. It’s something I’ll come back to, and I’ll try and make sure I’ve asked all the questions I’ve wanted to; but, at those times, it’s the overall objective that is the orienting point to the interview. I genuinely, deeply, want to discover—with the participant—what the answers are to my research questions. Indeed, I sometimes think that the best interviews are when the interviewer has just one main question, and riffs around it, with the interviewee, in a variety of ways to answer it as best they can.
Interpret along the way
What you can see in the interview above, and my second key pointer for most kinds of medium-to-large Q interviewing, is to try and make sense of what the participant is saying as you go along. Far better than trying to interpret what the participant was saying once you’ve got your transcript, do it as you go along with the participant. Because you’ve got them there to help you, and guide you, and tell you what it is that they actually mean. For instance, in the example above, they’d said that they wanted ‘structure’, but what did this actually mean, and were they saying they wanted guidance or something else? I wasn’t sure, and I asked.
Lots of times in interviews, you’ll be sitting there thinking, ‘I’m not really sure what this person is trying to tell me.’ That’s nothing to be ashamed of—it’s a really valuable indicator to you that you need to be asking more. You can even say to a participant, gently and kindly, something like, ‘Sorry, I didn’t quite get that, can you explain it to me?’ or ‘I think what you’re saying is …, Is that right?’ If you leave an interview really confused by what a participant has told you, chances are you’ll struggle to make much sense of it once it’s transcribed. And, of course, participants won’t always be able to make sense of things or tell you what they really mean. But to the extent that participants can explain and describe their experiences in depth, use the interview—itself—as a way of unpacking all that richness and meaning. In my own experience, I’ve been amazed by how much research participants can actually make sense of their own ‘material’, themselves—given time, space, and a dialogical sounding board to help articulate and explicate on what they are saying. And that way, when you leave the interview, a lot of the hard work of the analysis has already been done.
Just related to this, a point my colleague Gina Di Malta raised, is that it’s important to try and stick to the participant’s own words as much as possible, rather than introducing your own terms and ideas (again, a bit like therapy). You want to interpret, but you want that interpretation to be one of ‘unpacking’ the participant’s meanings, not introducing terms or conceptualisations from outside. So a question like:
You said you wanted ‘structure’, can you say more about what that means to you?
Is probably better than one like:
By structure, do you mean having an agenda for the sessions?
Given the power differential in the interviewing situation, a ‘yes’ to the second question really isn’t going to tell you much. Did they really want to have an agenda for the sessions, or were they just ‘deferring’ to you to be polite?
keep to task
Key pointer #3: you’re not the participants’ therapist, and you need to come away from the research interview with answers to the research questions that you’re asking (at least, in the ideal). Something I see a lot in supervisees’ interviewing, particularly when they’re training in a person-centred or relational approach, is that they tend to fall in to a non-directive, therapeutic mode of being: following the participant in whatever the participant wants to talk about. Of course, it’s important to be sensitive, empathic, and non-judgmental when you’re interviewing—and not overly-authoritarian—but participants may well come into the interview with you with things that they want to talk about or tell you about, and what they want to say may not relate in any particular way to what your research questions are asking. A participant, for instance, might want to tell you about difficulties in their marriage when you’re actually focusing on their experiences of a particular service; or they might want to tell you about their experiences of a service when your research project is actually focused on their childhood experiences. Participants have every right to want to tell you what they want to tell you, but it’s both ethically and methodologically important that you help steer the conversation to what you’re research is about. Otherwise, you can end up leaving the interview without any useable material and, more importantly, your participants can go away feeling confused and exposed by a process that was different to what they were invited in to.
So even though participants may take the interview off in all kind of directions (and that’s their right to do so), your role is to gently, sensitively, and compassionately bring them back to the main points of discussion. That doesn’t need to be rude or abrupt: if, for instance, a participant is really going off track, you can gently ask them how it relates to your key research questions, or thank them for their response and go on to the next question. And, in any interview, there’s bound to be material that is less relevant to the researcher’s question(s). In a sense, the issue is less about there being lots of ‘irrelevant’ material, and more about their being insufficient relevant material to be able to answer your research question. Having said that, transcribing and then trying to analyse page after page of material that, actually, isn’t much use to your research can be quite a heart-sink. Again, that’s why it’s really useful to do a few pilot interviews and a bit of an analysis, to make sure that the way your interview is set up is getting you the answers you want.
Keeping an interview on track means that, at times, you may need to be quite directive. For therapists schooled in the idea that directivity is just one step away from satanic worship, adopting a more proactive and leading stance can feel challenging: the natural tendency is just to go with the client/participant and not interrupt, structure, or lead. And, as above, there’s a lot of relational counselling skills that can be really valuable in interviewing work. But interviews do need to be more than just an unstructured, unfocused chat: again, not just for you, but out of an ethical duty to your participants to stick to ‘what you said on the tin’—they didn’t sign up for therapy, so you can start providing it for them.
Below is an excerpt from an interview, from the same study on client preferences, which demonstrates how an interviewer can gently and sensitively focus the interview process. In this case, the participant has been talking about a preference for a therapist who was challenging-but-supportive, but then goes on to talk about a preference for a female therapist. The interviewer positively affirms the expression of the latter preference, but then invites the participant to say more about the original preference so that they can explore it in more detail and systematically:
Interviewer: I guess there are lots of different preferences back on who your therapist is, and what the…
Participant: Oh yes.
Interviewer: …treatment is and…
Participant: Yes, I did prefer women because I just thought I would be more relaxed with that.
Interviewer: That's a good point. That's a really good example of preference, but what you said before, I'll just come back to that because that was really interesting as well. You were saying that you kind of have a preference, so someone who was a bit challenging but also quite supportive. How would you phrase that one?
Fear
Let’s be honest, interviews can be pretty scary to do especially if, like me, you’re natural tendency is towards being quite shy and unconfident. You’re interviewing someone who, pretty much by definition, knows more about something than you, and you may be interviewing people in positions of authority or power. You want to look good: you don’t want to babble away, lose your recording device, or have participants leave the interview thinking, ‘What kind of moron was that!’ Recognising any fears, of course, is the first step in managing them, and talking them through with peers or supervisors can be a really valuable next step. That can help normalise any anxieties, and also help you think about strategies for dealing with them. Ideally, you’re going into the interview feeling pretty calm, so that you can really listen to your participants. You want to have your head in your participants’ worlds—not in your own anxieties. And you also need a degree of confidence to keep your participant to task.
Having said all that, it’s also worth remembering the Yerkes–Dodson law: a bit of arousal isn’t a bad thing, and can help you be focused, attentive, and organised. It also conveys to your participant that you’re taking them and their time seriously: too laid back and your participant might feel a bit taken for granted. And, whatever else, please don’t fall asleep while doing an interview!
Stonewalled
What if you’re interviewing someone who, however, hard you try, doesn’t seem to want to give you much, or just doesn’t seem to be able to give in-depth responses? You’re getting short, one sentence answers; you probe, prompt, try to ‘unpack’ what they’re saying, but it’s just more one sentence responses, ‘I don’t knows’, or even silences. Or perhaps your participant has something that they’re so keen to tell you that you just can’t keep them to task: nothing they are saying—however hard you try—is related to your research questions.
Of course, ethics comes first. To repeat, your participant has the right to come into the interview and do, or not do, what they want (within reason), so if it’s clear that they don’t want to say more or go into more depth, that has to be respected. More than that, it’s important that the participant comes away from the interview feeling that their contribution has been valued, and not that they’ve let you down in some way.
In my own experience and that of my research supervisees, it’s very much the norm that, say out of eight interviews, at least two or three might be more limited in content. So if you’re finding that you’re not getting much from a particular interview—and you’ve tried, as much as would be reasonable and respectful, to invite more, without success—then it’s generally best just to accept that that is how that interview is going to be. In a sense, the quality and depth of a research project is not defined by how many ‘bad’ interviews you have but how many ‘good’ ones there are. And one option, if you are finding that a number of your interviews are quite limited, is simply to do more. Having said that, even where participants give relatively limited responses, the material may still be very helpful: giving an indication of the range of different responses that may be out there. Also, if you’re doing a smaller q study, with some quantification of responses, more limited interviews can still help you develop a sense of how many people experience or see things in a particular way.
Of course, if it’s the case that a large proportion of your respondents can’t give you in-depth responses to your interview questions, then that probably says something important about the way your interview is structured, or what you are trying to do. Again, that’s why it’s so important to pilot your interview schedule and to see whether people—including yourself—can give rich, deep, and interesting answers to the questions you are asking.
Ethics
A few further words on the ethical side of the research interviewing process.
Ethics, as always needs to come first. Interviews can be extremely rewarding for participants, but there are also some risks to be mindful of, particularly if you are exploring highly sensitive topics (for instance, clients’ experiences of childhood difficulties). Some pointers:
Make explicit in the information sheet what the research will involve, including potential risks, so that prospective participants know what they are signing up for.
Empathy and a non-judgmental acceptance should be de rigueur throughout the interviewing process: to repeat, you want your participants to leave feeling that their experiences and perceptions have been heard and are of value.
Pay close attention to maintaining your participants’ confidentiality, particularly if materials are going to be made publicly available (for instance, on a thesis University depository or in a journal). Also, be extremely careful with audio files—it’s a lot easier for participants to be identified if their voices, as well as the content of what they say, is disclosed.
Hope for the best but plan for the worst. What happens, for instance, if your participant gets extremely distressed during your interview and says that they are going to kill themselves? Make sure you have strategies in place to respond to such emergencies (for instance, making it clear in your information sheet that you would tell their GP if there was serious risk of harm, and providing links to support services like the Samaritans).
Be careful (as above) not to slip into a therapist role. Some of your interview may be therapeutic for the participant, but the focus should remain on answering your research questions rather than healing for the participant, per se.
Read up on it
Finally, as with any method, it’s important to explore the contemporary literature on interviewing, to get a deeper sense of how it can be done, the issues that can emerge, and different tips and strategies for optimising benefit. My own personal bible on interviewing, for many years, was Steinar Kvale’s (1996) InterViews, now in its third edition as Brinkmann and Kvale’s (2014) InterViews: Learning the craft of qualitative research interviewing. The book gives a brilliantly humanistic, relational guide to the interviewing process. Sage also have a series of videos on in-depth interviewing that you may be able to access.
Acknowledgement
Thanks to Gina di Malta, and trainees on the PsychD Counselling Psychology Programme at the University of Roehampton for suggestions and advice.
Photo by Christina@wocintechchat.com on Unsplash
DISCLAIMER
The information, materials, opinions or other content (collectively Content) contained in this blog have been prepared for general information purposes. Whilst I’ve endeavoured to ensure the Content is current and accurate, the Content in this blog is not intended to constitute professional advice and should not be relied on or treated as a substitute for specific advice relevant to particular circumstances. That means that I am not responsible for, nor will be liable for any losses incurred as a result of anyone relying on the Content contained in this blog, on this website, or any external internet sites referenced in or linked in this blog.