Writing successful research questions #57

lindsayboylan · 2020-06-24T13:42:56Z

General Guidelines

Avoid leading questions and try to talk as little as possible, if you can. Let the participant lead and fill any gaps in speaking.
Try not to pre-fill your questions with answers. If they need prompting, that's okay, but avoid blurting out "for example....xyz"
Ask for specific, past examples. Users are inherently bad at predicting future behavior. They self-report how they want to see themselves or want to act rather than what they'd actually do.
As an interviewer, you should ask open-ended questions that allow users to explore options and question their own responses to really get into the 'why'. Practice the 5 Whys technique.
Use your script as a guide, not a list of bullet points you absolutely must hit. There are no extra points for asking every question! Make your participant as comfortable as possible. If you need to skip or re-word a question, that's okay.
Both your participant AND you need to be comfortable. You can end the session at any time if you deem necessary.

I find it helpful to give broader scenarios rather than specific questions. This keeps you from writing leading questions that turn into "ok do this, then this, then this" and lets us see how someone would actually navigate your prototype.
Obviously, not all prototypes lend to this strategy, so be sure to write your goals first and then I find it easiest to write the questions & construct the prototype at the same time.
Again, open-ended, non-leading questions are best. A good rule of thumb is if you're looking for more, your question shouldn't be able to be answered with a "yes" or "no." However, sometimes you just need to know!

lindsayboylan added INCOMPLETE Resource and removed INCOMPLETE labels Jun 24, 2020