Dissertation Diary #8: Developing an Interview Guide
The definition and design of an interview guide for qualitative data collection
Interviewing as a research methodology is distinct from other types of interviews, but a few things are consistent across fields. For example: whether you’re a researcher, a journalist, or a detective, you’ll want to avoid the Chris Farley-esque questions and be confident in and reflexive about your positionality before going into the interview (rather than figuring yourself out in the moment).
But researchers must follow the additional recommendation to let theory guide their design of a good research question. This driving question will direct the development of interview questions whose answers, in the aggregate, will offer an answer to that theoretical puzzle.
So how does one go about creating such an interview? Generally in one of two ways: the river and channel or the tree and branch1.
River and Channel Approach
This frame for interviews is more of an unstructured, natural conversation with a singular focus in mind. You have one topic (the river) that has many possible themes to explore (the various channels that build from the river), but you wish to explore one theme in depth, following that theme wherever it takes you, even at the expense of looking into other themes (i.e. you travel down the channel of interest and flow with it, no matter what). This approach is most worthwhile when you can’t fully understand the research context without knowing what this one vital concept or theme means. So you have one driving question and then follow that question up with many additional detailed questions that are crafted as the conversation flows.
Tree and Branch Approach
If you are more interested in how the many parts (branches) that create and connect to the whole (tree), then the tree and branch model would be a more appropriate frame for your interviews. Here you have a topic that you want to explore piece by piece, using several main questions to structure the flow of the interview, with follow-up questions and opportunities for further probes for each section. This approach is perhaps the most common in sociological studies, referred to as “semi-structured interviews.”
Think of semi-structured interviews as a guided conversation. The interviews typically all start the same, but they each may take their own path through the interview protocol’s set of questions; the goal is that all main questions will be asked, but in a way that builds a research partnership to produce rich data with a specific participant. It shouldn’t feel like a survey; for the participant, it should feel like chatting about the topic at hand.
My dissertation is using the tree and branch model to design semi-structured interviews with teachers, staff, and school leaders at geographically diverse neighborhood public schools. I won’t share my exact interview protocol yet, as I am still in data collection, but I’ll walk through the general outline of the conversation.
Opening Questions
You want to build rapport with the participant, so the first few questions you ask should work to warm you both up — get the participant comfortable with talking and give you an initial sense of the participant. I start by asking the participant to tell me about themselves and see where that takes us. Typically they’ll mention things about where they’re from, how long they’ve been teaching, etc. and then I’ll follow-up to get more details about how long and where exactly they’ve lived in the city, what their path was to teaching, what other previous teaching experiences they had, etc. — this is a friendly, “normal conversation” start to our interview.
Branch 1: Experiences at X School
It’s usually a pretty natural segue from the background questions to learning why and how the participant ended up at their current school and get more details about their specific role and responsibilities. I also ask participants to walk me through a typical day and share a bit about their students.
Branch 2: Neighborhood & School Community
Talking about their students helps us jump to a new branch that focuses more on the neighborhood(s) aspect of their work. Here we talk about their school as a school that serves students from many different neighborhoods, both from the perspective of the school’s neighborhood (e.g. how does the school’s neighborhood impact your work?) and the students’ home neighborhoods (e.g. do students’ home neighborhoods factor into their daily and long-term decision-making for their work?). This branch gets to the heart of the study’s primary research question.
Branch 3: School Choice
This branch asks educators to opine about the idea of school choice broadly, then to consider the actual policies and practices of the school district and how, if at all, they feel the impact of school choice in their day-to-day work.
Wrap-Up and Demographics
This final section tries to tie a bow around the interview and bring it to a close. I ask participants if there is anything else they’d like to share or that they think I should have asked. I then ask them a series of demographic questions. I explain that these are purely for my contextualization of the data and for better understanding patterns across the data; the interviews will all remain anonymized and confidentiality prioritized in any public-facing outputs from the study.
Given the kinds of research topics and questions you’re interested in exploring, what would be the more appropriate approach: the river and channel, or the tree and branch? Share in the comments!
Rubin, Herbert J., and Irene S. Rubin. 1995. Qualitative Interviewing. SAGE Publications.