Doing Nightlife Fieldwork III: Interviewing

Entrance to the Larco Museum of Pre-Columbian History in Lima, Peru. Why should all these pictures be party pictures, anyway?

eep reading, folks! This is the last installment of the series on fieldwork methodology for ethnographic research in nightlife scenes—in other words, how to study a dance music scene without getting in the way. Last Thursday, I posted a brief list of problems that nightlife scenes pose to conventional modes of ethnographic research. Unsurprisingly, most of those issues had to do with the circumstances of EDM (electronic dance music) events themselves; and so, on the next day I wrote another article on “going out” for fieldwork, mostly detailing my own methods and giving a few bits of general advice. Todays post (and the final one in this series) covers a more conventional but no less important aspect of music ethnography: interviews! While the process of actually interviewing someone off-site (i.e., away from an EDM event) is pretty similar to other interviewing situations, there are some important things to keep in mind when trying to secure these interviews and ensuring confidentiality.

Interviewing Party People

When preparing to do fieldwork for the first time, it may be tempting to see interviews as the easiest part of the fieldwork process. Anyone can ask questions, right? You just need to write up a list of questions and then put them to a bunch of people and record their answers. But no, there are lots of things to think about and to make decisions about: how will you phrase your questions, in order to get the richest and most accurate/honest information? Considering that the interviewee’s time is limited, how many questions will you ask and which will take priority? How do you get interviewees in the first place? How will you protect the interviewee’s identity? Will you record the interview, take handwritten notes, or both? Where will you conduct the interview? Will you ask uncomfortable questions? How will you analyze the interview afterwards? And then, after dealing with all of those decisions, there’s still the tricky social performance of maintaining the flow of conversation.

Let’s begin with “recruiting” (somehow, I’m reminded of Anita Bryant and Harvey Milk: “We recruit!”): securing an interview can be tricky in nightlife scenes, because there are heightened concerns about confidentiality as well as the political intentions of the interviewer (see #1 and #2 from Thursday’s post). I had once said in my dissertation that nightclubs (and nightlife/leisure spaces in general) strive to maintain “an atmosphere of relaxed rules, suspended responsibilities, expanded possibilities, and pleasures unburdened by guilt or sanction” (p. 133). This means that things go on at nightlife events that can be transgressive, scandalous, embarrassing, compromising, and sometimes clearly illegal; the stakes of privacy and discretion are thus much higher than in everyday settings, and scene participants are justifiably concerned about what you’re going to do with their information and how you’ll protect it. I’ll get to issues of confidentiality later, but I’ll focus here on how I get interviews.

In such an informal and “underground” setting (i.e., where a nicely-written letter of introduction isn’t going to get you anywhere), I rely on what I call a “trust network” to support my requests for interviews. This means making social connections first—often through the introduction of a mutual contact, but sometimes directly—and taking time to develop that connection into a relationship of trust. Essentially, this is a form of Kontaktpflege (cultivation of contacts), which allows the contact-person to get to know you better and to develop a sense that you aren’t going to hurt them in some way. This can take weeks and months, and often involves repeatedly meeting for coffee/drinks, attending music events together, attending his/her events if s/he’s a performer or promoter, exchanging e-mails, maintaining contact through social networking sites, and revealing more about yourself and your own relationship to this particular nightlife scene. An important factor in my success gaining interviews, I think, is the fact that I have been active in EDM scenes since the mid-90s and both 1) know the cultural norms of the scene, and 2) have a strong commitment to it.  In any case, what I wanted to avoid was scaring people off by presenting myself as an academic “studying you people” and then requesting that this person open up about his/her life without knowing anything more about me. Ironically, some of my contacts have actually told me that I’m too cautious in this regard; but even if someone might consent to an interview right away, I think that the quality and depth of information that you get will be far richer if that person has time to know you and become comfortable with you.

What I do is similar to a well-known sociological method called “snowball sampling,” which involves asking each of your interviewees to recommend other potential interviewees for your project. This method has the benefit of using pre-existing networks of trust (i.e., those connections between your interviewees) to vouch for your trustworthiness and to find people who might not otherwise want to be found. It’s a common technique in drug-use research, for example. But the method has its drawbacks, most of which have to do with the risk of skewed results. Essentially, developing your study group through snowball sampling means that you risk doing an ethnography of a particular social network, rather than the scene in general; the status of your group as a “representative sample” of the wider scene is put into question, and you risk mistaking the quirks of this social group as the characteristic traits of an entire population. On the one hand, there’s nothing inherently wrong with developing a richly-detailed account of an interrelated group of people; on the other hand, it limits the truth-claims you can make about the larger world of nightlife. There’s no simple solution to this issue, but I work to mitigate this methodological weakness by working hard to develop my trust network beyond the network of those people I already know. This means going out and meeting strangers, being outgoing, engaging in small talk, becoming a “familiar face” at important venues and events, and trying to cultivate friendly relations. Since my dissertation was on stranger-intimacy, this sort of approach was especially important to me at the time. And, one of the advantages of the generally heightened conviviality of nightlife scenes is that people are more open to strangers than they would be in everyday, normal life.

So, how do you handle privacy and confidentiality? You can find a lot of general advice on this through anthropology and sociology portals (textbooks, professional society handbooks, etc), but here’s what I do. I have a verbal consent “script” that I usually send to my interviewee in PDF form a few days before the interview, asking her/him to read it and let me know if there are any questions. This script describes the following: myself, my project, how I intend to conduct the interview, what the interviewee will be asked to do/answer during the interview, what I’ll do with the information I collect, and his/her rights to refuse or withdraw from the study (if you want to see a sample, email me). At the time of the interview, I go over these details verbally, and then turn on my dictaphone and make sure that I get the following questions and answers on the recording: 1) May I interview you? 2) May I record this interview? 3) May I cite this interview in my research?

After the interview, I store the recording of the interview only on two hard drives, my laptop’s and an external backup drive, both of which are encrypted. I assign a pseudonym to the interviewee by default—they sometimes prefer their real names or their performer/DJ/producer moniker—and then ensure that all files, notes, and transcripts bear only the pseudonym and never the person’s real name. The only place where I connect a person’s real name to her/his pseudonym is in my head; I leave no physical trace of the connection. When I cite interviewees in a piece of writing, I send them a draft of the essay/chapter, indicating the pages where they appear, and give them a reasonable amount of time (usually, 1 month / essay or chapter) to look it over and reply to me; they can: approve the citations as-is, contest any misquotations or misinterpretations, expand upon or emend their statements, or withdraw themselves from publication. It’s also usually at this point that interviewees can request to be cited with their real names, if they prefer. This final stage gives a lot of power to the interviewees to control their own portrayal; this is important for my project in order to maintain a sense of trust, but I’ll admit that it’s not always feasible for everyone. If you’re working with people that you’re not likely to see or contact again, you may have to just get permission to cite and leave it at that.

Of course, if you’re doing your work with the support of a university or any other academy, you’ll most likely have to go through an IRB (Internal Review Board) process, where you submit all of your fieldwork “protocols” for approval. Unfortunately, this process is often directly imported from the IRB process for medical studies/drug trials, which means that you end up having to answer questions that don’t seem relevant and to re-think your project in awkward terms. All of this will also shape your methods somewhat, for better or for worse. Nonetheless, the process is valuable for forcing you to think clearly about your methods before going out and making a mess in the field.

As far as how I conduct the interview itself, it’s actually pretty straightforward. I usually let the interviewee choose the location and time of the interview—although I have learned to insist that the location be relatively quiet. For my first round of interviews for my doctoral research in Paris, I didn’t think about this and all of my Parisian friends said, “Let’s meet in a café!” This made perfect cultural sense for Paris, where people treat their local café like their living room, but it made the interviews 10 times harder to transcribe and analyze afterwards. So, make sure you agree on a place where you can get a decent recording. I never had a way of paying people for their time, so I usually invite them to my place and feed them lunch or afternoon tea/coffee with pastries. Even small gestures like that go a long way towards showing that you appreciate your interviewees’ time.

I record the entire interview with a dictaphone (currently the Olympus VN-6500PC, which has served me well despite only recording in the most inconvenient of sound formats: WMV), and I bring along a small notebook. My “interview guide” (i.e., list of questions, topics) is usually written on the first few pages of the notebook for my reference. I take skeletal notes during the interview, but mostly these notes are just prompts to remind me to pay attention to particular moments in the recording. I try to keep note-taking to a minimum, because it breaks eye-contact with the interviewee, interrupts the flow of interaction, and invokes body-language that tends to shut down conversation. An interview usually lasts about 1 hour for me these days, although my dissertation interviews often took 1.5-2 hours, given the more abstract and vague nature of what we were talking about then. It depends a lot on how many anecdotes you are requesting from the interviewee.

Post-interview analysis is also important to think about. When I first started doing interviews for my dissertation, I transcribed every interview all the way, from beginning to end. For English-language interviews, 1 minute of interview took 6 minutes to transcribe. For French-language interviews recorded in noisy cafés, it took as much as 10 minutes to transcribe a single minute of recording. It was awful, and I have no idea what I was thinking. Still, I am reluctant to send interviews to professional transcription services; not only does that cost a lot of money and create more opportunities for a breach of confidentiality, but you also miss out on the opportunity to listen closely to the recording yourself. As shitty as transcription is, you notice a lot more about the fine details of the interview when you do it. My solution now is to do a sort “annotated listening” of an interview with occasional transcribed bits. I open up Scrivener (which is amazing), create a new blank document, write down the date and location of my interview, copy out the notes that I took during the interview, and then flesh them out with a sort of running commentary as I listen to the interview again. Most of the commentary is just me summarizing what is being said in point-form and adding the occasional time-stamp to show where in the recording we switch to a new or important topic. If I come across something that seems eminently quotable or important, I stop and transcribe the relevant sentences. Otherwise, I just summarize the content and highlight text as necessary, so that I know where to go for a proper transcription when I’m actually writing my essay/article/chapter/whatever. It still takes about 2 hours to analyze a 1-hour interview, but that’s much faster and it generates a more useful, readable set of research notes.

Tips & Notes

• Types of Questions: There are many kinds of questions you can ask, and not all of them are helpful or appropriate to your research goals. Closed yes/no questions are generally a bad idea, but crafting good open-ended questions require some forethought and practice. Your question might: ask for examples; ask for stories; ask for clarification (“What did you mean by…?”); ask for expansion (“Say more about…”); ask about hypothetical situations (“What if…?”); ask for filtering (“Which of these…?” “What’s important about…?”); ask for opinons (“What do you think about…?”), and so on. For my own work in nightlife scenes, I’ve often found that asking for stories about relevant topics (e.g., your entry into the EDM scene, your most memorable time partying, an example of stranger-intimacy—or un-intimacy, etc.) provides a great way to accumulate a rich archive of examples while also giving you ideas of where to take the conversation next. Further reading: Terry J. Fadem’s The Art of Asking and MacKay & Weinstein’s Asking Questions.
• Anything Else? One of the things I’ve learned to ask at the end of every interview is some form of, “Anything Else?” e.g., What did I forget to ask you about? What is important about this topic that we haven’t covered already? What should other people know about this topic?
• Interview Structure: Structurally speaking, you can describe your interview as structured, semi-structured, and unstructured. Structured interviews involve having a full set of questions that you ask in a particular order, avoiding deviations or tangents. Semi-structured interviews have a few prompting questions and a set of topics to cover, but approaches these issues in any order and follows tangents if they seem interesting. Unstructured interviews involve sitting down, maybe setting a topic, and otherwise following the interviewee wherever s/he takes things. All of these approaches have advantages and disadvantages—most of which you can figure out for yourself, I imagine—but I tend to prefer semi-structured interviews, as they allow me to create some points of comparison between interviews while leaving openings for things I hadn’t thought to ask about.
• Enjoy the Silence: (Sorry, I couldn’t resist a Depeche Mode reference) Silences can be uncomfortable, especially if you’re nervous about your interviewing skills and interpret silence to be a failure on your part. Sometimes, however, an interviewee is collecting his/her thoughts or struggling to find the right way to express something. Learn to recognize these moments and resist the urge to fill in the gaps or finish her/his sentences.
• Don’t Manipulate: A skilled interviewer will know how to put the interviewee at ease and guide the flow of conversation such that s/he provides clear, insightful, and useful information; but don’t try to trick the interviewee into saying something s/he doesn’t want to disclose, or pressure him/her to say what you want to hear, or misrepresent your intentions.