Research

How to write an interview discussion guide

Caretaker: Kayleigh


Everyone has their own style – some people write out every single question in the exact phrasing it will be asked, some (more experienced) people just write a list of bullet points….

We believe the more complete version is better. It means that anyone can pick up your guide and run the group in (almost) the same way, ensuring consistency across researchers and internal validity in responses

  1. Break your subject into a series of discrete, single minded topic areas
  2. 8 “topics” / sections is ideal for a 2 hour group (definitely no more than 12)
  3. Think about most natural flow through topics and order topics in that narrative
  4. As a rule of thumb, start easy and get more complicated as you move through, but beware that people will get tired / bored towards the end so don’t put your most crucial topics too late in the group
  5. Think about how long each topic area will take to explore ahead of time and mark timings on the guide to help you stay on track in the session
  6. Think about your audience when designing questions to ask. Think about the language they would use; no jargon or marketing terms, think about their intellectual level and vocabulary (e.g. kids vs. low SEG vs. high net worths), ensure phrasing of questions is non-threatening
  7. Ask good questions: what, where, why, who, when, how, which.  Avoid questions that can be answered by “yes” or “no”
  8. Don't ask biased or leading questions: “How do you feel about X” not “Do you think X is bad?”. Read your guide once its written to identify biased questions – it’s easy to miss them!
  9. Always begin with a vaguely relevant ice breaker question to make people feel comfortable and get the group off to a good start (cf. forming and storming in group theory)…..  “Tell me about your dream home?”
  10. Try and come at each topic area from a couple of different directions; question around the same topic in different ways to help uncover new insight without it feeling repetitive. Think about how you might be able rephrase prompts and probes to help coach answers from participants ahead of time – especially if you have an audience that is “different” to you (e.g. low SEG, not techy). 

Here is an example guide template.