Research interview can be bigger beast than people expect. Let's delve into the subject once more. How do you get the most truthful answers, what can you expect from your interviewees and what are the most common mistakes beginning interviewers make?
Research interviews can be difficult and require lots of adapting. Leading a discussion is helpful but it also is not everything. There are rules to the questions asked, guidelines to follow and mistakes that can be made.
Getting the truth
Asking "Do you exercise?" will give you very biased answer. People tend to say yes. What are you asking though? What is the definition of exercising? Is once a week enough? How about once a month? Or is four times a week the lower limit? This question is wrong for two reasons: It is too generic and it allows biases.
The generic part will be addressed later. But lets look into the biases. People like to believe that they are better than they are in reality. That is simply how we are as human beings. You need to be counting with this. Therefore, lets try to ask said questions differently, for example: "How many times have you exercised this week?"
Asking about past gives you more precise answer than asking about future. Asking about specific and not very far past gives you even more precise answer. This is because asking about certain time interval they can remember does not require them to fill in the blanks and guess what interval you are referring to.
If people lie to the second version of the question, there is not much you can do about that, but at least they won't be biased and pretend they are hardworking gymnasts if the last time they went to gym was 3 weeks ago. And yes, that still counts as exercising.
Interview is a puzzle
You want answers to specific questions but during an interview you need to be careful about asking too generic questions. Just as it could cause bias described in previous section, you can also get very incomplete answers.
Example: "How was the workflow of you work this week?". Questions asking about specific past doesn't give the interviewee lots of possibilities for biased answers, or does it? Even here they can say various things. I find it better to ask questions surrounding specific problem and then connect all the answers like a puzzle.
"What was the biggest workflow problem you faced this week? How long did this situation lasted? Was there other similar or completely opposite situation?" These questions are lot more specific and from answers you can put together a picture of what was truly happening with the workflow that week.
I will also state that there are times when specific questions are not so good. Before you delve into deep discussion you first need to cover the foundations, the basics of the problem. If you start asking specific questions right away, you might miss the real issue because it simply never comes up.
It is completely possible that even though workflow in the example above is viewed as problematic, people might actually be experiencing way bigger issues with client communication. And solving that would have way bigger impact than solving workflow. Therefore asking only about workflow will make you blind to the reality of the situation.
Alternate more generic questions to get the overview of the situation and then start asking more specific questions to get to the bottom of the situation.
Lets imagine you are in situation where... It doesn't matter what continues, question like this should raise a red flag for you. The moment you are force or even let people imagine something you are once again getting into the biased waters.
People are very bad at predicting future. Really bad. Its just another part of us we have to account for. The moment you are asking if they used a tool in certain situation in the future or if they would like to get certain information, you are pretty much gambling. They can answer anything and you won't really know if it would be truth if such situation ever arisen.
Example: When you are hiring new manager and you ask her how she would react if she needed to kick someone, her response can be different than what she would really react like in such a situation. The rush of emotion suddenly changes her view in the heat of moment. The reality and the predictions are always different.
Users are not designers
Design is more than just creativity. It requires mind set, skill set and some other sets. On the serious note, designing is not as easy as people consider it to be.
Letting people design something has two problems:
- They only know their side of the view. They know their problems, not the problems other people are facing.
- They will implement solutions that help them and only them. As limited as their view is, they limit the ideas even more by making them worthwhile only for limited audience.
Be careful about asking people to design something. You can ask if you deem it important or useful, but do not treat it as must have features or ideas.
Blue sky question
On the other hand, there is practice of allowing blue sky thinking. Therefore creating ideas not limited by current possibilities or beliefs.
I prefer to do this at the very end when we already discussed everything there is. I lay out a situation in which the person has infinite amount of money, possibilities and no one to limit them. I ask what is that they would do? Of course I guide the question to the topic we are talking about the whole time.
This might result in people trying to design solutions, but if you pay attention and read between the lines, you will notice what is bothering them. It can confirm all you learned beforehand from that person. Alternatively it can tell you what they think is bothering them, what they feel to be the problem and you can compare this to what truly seems to be the problem.
Psychology is for experts
Some people in UX branch of study have IT backgrounds, some have psychology backgrounds, some have artistic background, etc. Therefore there are different skills you yourself have and you can use.
Understanding body language can be useful. It could also be downright necessary if you want to be the biggest expert on the scene. But for the starters it is not required. Its nice to have, but it ends there for beginners.
If you are really not satisfied with your ability to comprehend body language, try to get someone who is little better at it. I will note here that women are in average better at reading and understanding body language than men.
Lastly, taking notes. It is difficult to listen, speak and also take notes. For that reason, it is suggested to always come in pair. One person is the main facilitator, who is leading the conversation, while the other person is taking notes and occasionally entering the discussion.
Having too many people there can be intimidating, so pay attention to that aspect. Also make sure that if you are not alone, you are not undermining each other with your colleagues. You need some synergy to pull this off.
Most common mistakes
Beginners tend to guide the interview in a direction they would like it to go. People tend to come up with some ideas for solution before they know problem and then they try to fit the problem the solution instead of the other way around. It is one of the biggest mistakes since it undermines the whole point of doing research interviews.
Arguing, persuading, being judgmental, etc. Also not good. You are there to listen and learn. Not to judge. Even if you do not agree with something, try to keep your cool and maintain smile or at least poker face.
Asking people about their ideas, their solutions. Not good since their view is limited and they may very well lack the skills necessary to be a designer.
Incorrect balance of generic and specific questions. There are times people come so focused on one thing, they tunnel vision. And there are times people can't get deep into the issue and leave with unsatisfactory information.
Forcing interviewees to imagine and asking them unclear questions. Unclear question and pretending or imagining something that has not happened or how it could happen in the future results in very imprecise answer.
Jumping from topic to topic. There are times you are forced to do this, but it should never be overdone otherwise it will create chaos in conversation.