The most important part of user-centered design is the interview. To create effective user interfaces, it's absolutely essential to get honest, detailed feedback from users. The biggest pitfall, in my experience, is the leading question. Here are tips on how to avoid leading questions and my strategy for effective UI/UX interviews.
What is a Leading Question?
A leading question is one that prompts or encourages a particular answer. For example:
Wouldn’t your life be simpler if you had XXX product?
This question leads by implying that there is a "correct" answer. The positive clause at the beginning of the question ("Wouldn't your life be better if...") implies that the person should agree in order to get the "correct" answer.
The same can be said for negatively-led questions:
Don’t you hate when XXX happens?
Isn’t it harder that you don’t have XXX in your life?
The structure of these questions changes the focus from the intended subject to whether or not the interviewee agrees with the interviewer. Most people avoid conflict, so interviewees will tend to agree with the question (answering "yes").
Another type of leading question involves using a leading statement before the question:
Bright colors are important. What do you think is the weakest part of XXX?
With this structure, the interviewer has declared the focal point before asking the question. Most people will be led by the preceding question and restrict their feedback to issues related to color.
What is a Non-Leading Question?
Non-leading questions are open-ended questions that eliminate our bias as interviewers. They are designed to allow the freedom to answer in any way. The most effective non-leading questions lead to full-blown discussions and can often reveal insights and ideas that the interviewer may never have considered. Thus, our goal is to avoid leading questions.
All of the leading questions above can be rewritten in non-leading forms:
How would your life be if you had XXX product?
How do you feel when XXX happens?
How has your life been affected by not having XXX?
What do you think is the weakest part of XXX?
In the non-leading forms of these questions, the person isn't forced to agree or disagree with the interviewer, nor are they influenced by the interviewer's suggestions.
Emotion-Based Questions
People are more likely to start a conversation when there is an emotional attachment to the topic, because feelings are abstract and personal and, therefore, take more time to convey. As such, the best interview questions are somewhat abstract and emotion-based. If we use this insight and avoid leading questions, we can create effective conversation-starters:
When you are doing XXX, does anything frustrate you?
What do you enjoy the most about XXX?
Strictly speaking, the above questions aren't entirely non-leading (the first one suggests that there might be something frustrating about XXX, while the second nudges towards something positive), but they are open-ended within those boundaries.
A Strategy for Effective User Interviews
With emotion-based, non-leading questions as our starting point, we can create a strategy for effective user interviews:
1. Start with an emotion-based, non-leading question
First, we avoid leading questions and start off with an emotion-based question as described above:
When you are doing XXX, does anything frustrate you?
2. Ask for an example or anecdote
After the person's initial answer, immediately follow-up by asking them to provide an example or anecdote to further refine their answer.
Can you describe a specific example when you were frustrated by that?
This will provide you with a real world example - in their terms - that can be used as a point of reference into your research. It also pushes the discussion across the threshold from short answer to conversation, which is important in getting deeper, more honest feedback.
3. Ask the person to quantify their feedback
Next, ask the person to quantify their answer.
One a scale from 1 to 10, with 1 being very frustrated and 10 being very happy, how would you rate your experience with XXX?
Using numerical values in your interview process allows you to start assigning quantitative measures to your data gathering.
4. Dig Deeper or Start Again
At this point, you can either repeat the strategy with a new question, or use the numeric answer as a starting point for another question:
Given that your frustration in this case was 3/10, have you encountered something else with XXX that you found even more frustrating?
Ultimately, I liken this process to bringing an interviewee to a maze you have constructed. You know at least one way out, but your goal is to avoid leading questions so that you're not showing them how to solve it. The value comes when they show you a new way out of the maze. If the maze (the questions) are structured well, then you should be able to follow behind, observing their progress and learning from the path they take to reach their destination.