Speaker 1: In this video we're going to look at semi-structured interviews, a very popular method for qualitative research. We're going to look at some of the reasons why it's so popular and some tips to help you make sure that they go smoothly. So why are semi-structured interviews so common in qualitative research? Well I think the first reason is familiarity. I mean an interview is basically a conversation and that's something that we're very familiar with doing. Participants will have conversations with people all the time and researchers have conversations with people all the time. So it's not something that's a completely alien or bizarre way to collect data from someone, just having a natural chat with them. The other thing that's very important about semi-structured interviews is their flexibility and that's one of their key strengths. They're flexible because they allow you to slightly change the wording and the order of questions that you ask to make sure that you get the most detail and the most information from each person. And they allow you to ask things in a different way for different people and that gives you a lot of extra ability to get detail, you can get clarification or information and you can ask extra questions about interesting things that come up that you might not have anticipated in the first place and that's one of the key strengths of qualitative methodologies in general. There are also some disadvantages to qualitative interviews. So because you're not asking the questions in exactly the same way, it's not completely replicable between your different respondents. You may not have covered the same questions with all the people and you may have asked them in different ways, so slightly different wording. In quantitative research that would mean that you couldn't really compare across the different participants. In qualitative approaches we're taking a slightly more anti-positivist stance, drawing from feminist methodologies, we're saying there isn't pure objectivity, there isn't pure truth, so what we're looking for is really to get a sense across the participants of the things that are most important to them. And so that's not really as big an issue for this kind of approach, but it does mean that there's a lot more onus on the researcher to be a good interviewer and to ask the right questions in the right way. If you had a structured interview guide and you're asking the same questions, it doesn't require very much cognitive ability really to kind of juggle the different things that we need to deal with in the interview, so we'll see a bit later that can be quite complicated. Semi-structured interviews do have a structure, that's why they're called semi-structured, and they're usually based around creating a question guide or an interview guide, and that's based on the questions that you want to have answered by the participants and those questions should help you answer your research question. Now most of the time you won't go directly to your participants and ask them your research question, you won't say what factors affected the drop in newspaper readership in the UK in the 1990s. You won't ask such a direct question to your participants, you'll be pulling out a lot more detail. So what you really need to do for your semi-structured interview guide is to break your research question down into, it usually ends up about 20 to 30 questions which help you kind of tease out the issues that you're interested in investigating. And then that guide will be in front of you and that will be a list of a kind of a memoir of the different questions that you could ask, but again you'll be a little bit flexible with them. You may change the wording of them a little bit, you may change the order in which you ask them, and that helps you to really get the sense that every participant is different, they'll have different experiences, different ways of describing their lived experience, and so you'll be able to pull that out in a much more natural style if you've got that semi-structured interview guide memorized. And that's one of the very important things you need to do is to memorize your interview guide. Literally know it backwards and forwards, so that if somebody suddenly starts saying something as a result of one question that naturally leads you on to the fifth question, then you can jump to the fifth question and come back and ask the second and third and fourth a little bit later on. But you do need to make sure that you're ticking them off and covering as many of those different questions as you can at the time, and that's one of the key skills that you need to learn as a qualitative interviewer. You need to be able to listen and also plan what you're going to do, what the next question is going to be. Some of the other things that you should bear in mind when planning your interview guide are to make sure that you start with some easy questions, not kind of just how's the weather kind of questions, but interviews should start with questions which are on the topic but are fairly simple for respondents to answer. Like a very basic question about their opinions or their experience or when they first came across a certain thing or something like that. And that helps ease people into the conversation and helps them loosen up a bit for more complicated questions that you might ask later. Now, the next thing you want to do is ask stupid questions. It often feels like as the researcher and the interviewer, you want to go in and know everything there is about your expert topic area, so you don't have to ask people to define basic terms and things like that. But actually, that might be something that you want to do because it's not how you understand a particular participant's life experience, it's how they understand it. So even if they have a concept such as engagement or happiness, I mean, what does that mean to them in this particular sense? And it's also true when you're working with very technical things. If you're working with an organization and they have departmental meetings, you could say, well, what's a departmental meeting? It's very interesting to hear what that person thinks that terminology means. It might not be what you assumed that they thought it meant. And so that's very interesting too. So sometimes it feels a bit stupid asking those questions, but those can often be actually the most rich questions to get answered. When you've developed your semi-structured interview guide, it's a good idea to test it and get some feedback. So you may try it with some of the participants, just try it with one or two first and see how it goes. But it's also worth testing with just friends or colleagues just to see whether the questions make any sense to anyone else, and also to practice your interview skills a little bit. How good are you at jumping ahead to different questions? How good are you at listening and nodding and things like that? So do practice and test before you start doing your real interview questions. The location where you do your semi-structured interviews is also very important. You need to find a place that's appropriate for the researcher and the participant, a safe location, a neutral location ideally, or somewhere that's kind of reflecting what you want to capture about them. So you may interview someone who you're looking at workplace scenarios in their workplace, and that might be the most appropriate place to do it. Somebody's talking about something very personal, maybe they'd rather do that in a private setting at home. Well, you need to be a little bit sure that that's going to be a safe place for both of you. It can often seem very tempting to do research in a cafe environment. It's a neutral location, everybody can access it, lots of conversations are going on there. But there's kind of two things that you need to bear in mind there. The first of which is there may be a lot of extra distractions going on there. And the other thing is there may be a lot of external noise. Now, you probably won't notice that too much just when you're sitting across someone because our ears are very good at kind of picking up audio and conversations on a one-to-one basis. But there will be a lot of things that get picked up by a dictaphone if you're using one of those to record the conversation that might make it very difficult to transcribe or hear what's going on later. So if you're working in a cafe, make sure you can, you'll have this kind of level of noise, make sure you can minimize the music that might be going on, and also any background conversation that might be loud. So scope out your cafe first, make sure that it's not somewhere that plays live music, go there at a quiet time, and then you'll get a good audio out of that as well. Now, I already mentioned very briefly recording the conversation. I would recommend that you get a proper dictaphone to record your qualitative semi-structured interviews. It may be tempting to just use your phone or a camera that you have lying around or something like that. Generally, they won't provide very good quality audio and they'll be very bad at picking up all the external noise as well. Even a very cheap dictaphone, you don't need to spend more than about £20, $30 on something appropriate for this, will give much better audio and will give you a lot more flexibility when you come to transcribe the audio later on if you need to do that. So when you're using a dictaphone, there are a couple of things which you will learn very quickly. First of which is always make sure you have backup batteries, so the batteries don't run out on the time. Make sure before you turn it on that there's plenty of recording space left on it, and that goes for any recording device that you're going. You may even want to have a spare way to record the conversation even if that is the phone or another dictaphone or something like that. You also want to make sure that the participant's not too nervous or made uncomfortable by having a dictaphone pointing at them. It's okay to point it a little bit away from them so it's not staring up in their face. Some dictaphones will also have a red recording light on them, like this one does, and I've often put a piece of tape or even a piece of blue tack over the recording light because that red light staring in the face really makes people seize up. Another trick to help people ease into it is don't turn on the dictaphone and then immediately start asking questions. Turn it on in the middle of your warm-up when you're still asking how people are going and then there won't be that kind of sudden start to the interview process. You also need to obviously make sure that they're comfortable with the dictaphone and they want to be recorded. So I would recommend taking some notes during the interview even if you're recording it fully with a dictaphone. You never know when that might cut out or you might lose the interview entirely and then you've got at least something to go back on. But it's also really good as an aide-memoire of what you've covered already in the interview, what you still need to cover before the person goes, and it can also be used to write kind of extra notes like, you know, when someone was particularly concerned or didn't seem very clear on the answer or some things that might not come across on the tape later. The other good reason to do notes is that it can be a little less intimidating for the participant. If you're just listening intensely to them and just keeping eye contact all the time it can be a little bit of an intense experience, but just looking down every now and then taking some notes not only can help them feel a little bit more relaxed but also give them the impression that you are listening and the notes you're writing are something significant, so they're saying something that's useful. When the interview is finished, do a little warm down as well as a warm-up. Say thank you, a little bit of chat about traffic. Don't be too preoccupied with with turning off the dictaphone at a particular moment, again make it kind of fluid and easy. And then when the person's gone just take five or ten minutes to sit there if you're able and take a few more notes. Even if you're not using a formal research journal as part of your project, I always recommend that you write a little bit of notes on the context, some of the key things which came up, things that surprised you. It could be very useful later as a bit of context when you come to analyse that particular interview. I'd also recommend that you say thank you to the person more formally, send them a thank-you card and keep them updated with the research. Share any findings which are based on their interview. It really helps them feel that they were part of something important and that their time, you know, usually an hour or so that they've given to your project was worthwhile and did something useful. And the most important thing is as soon as you get back home, get to a computer, back up that audio recording. Copy it on as many different places as you can, just to make sure that you don't lose it and it's everywhere that you need it to be. You don't want the embarrassment of going back to that person and saying, I'm sorry, we did all that intensive work and I lost the audio recording, so we need to do the interview over again. It does happen, so don't get yourself up about it, but just try and make sure that there are a few opportunities for that kind of problem as possible. The next thing to think about is transcription. I would recommend most people probably do want to transcribe their semi-structured interviews. Although it's great to have the audio close to hand, the text is much quicker to read through and it also helps you be able to kind of skip particular parts and find particular keywords and particular sections that people are talking about. So whether you do that yourself or you use a professional automatic transcription service, it's up to you. And there's some tips in our blog post on doing that. But you can also use qualitative analysis software like Quarkos to manage your participant database. Most people will have a journal or an Excel spreadsheet or something like that with all the information about the participants, their name, location, phone numbers, how to contact them, whether you've got consent forms. Well, you can actually keep all of that data as source properties in Quarkos and you can use that to make sure that you know exactly everything about the participants and even find people you need to contact for follow-ups or you don't have the consent forms back for and things like that. So I hope this is useful. There's a lot more information on our blog, quite a few posts which are relevant to this topic. So we've got one on 10 tips for doing semi-structured interviews. We've got several on automatically transcribing audio data and also one for doing it yourself. There's also some tips and tricks on getting good audio and managing interviews and focus groups. So Quarkos aims to make qualitative analysis like this very easy, very simple. It's very inexpensive, very easy to learn and there's a complete free trial which you can download and try from our website. So do go and try Quarkos today and good luck with your semi-structured interviews.