Talking to Humans
An introduction to approaches for interviewing users and the resulting consideration of research ethics. Lecture slides will be made available on the day of the lecture (June 15).
This week we want to fill in more of our gaps in experience to ensure we understand the humans in the context of this problem.
Why We Interview
We can use interviews to understand individuals, their experiences and behaviours in context. Be careful to keep in mind:
- We are not our users.
- We are not persuading our users.
- We are guests.
(Before You Wreck Yourself)
As an interviewer you want to avoid bringing your own biases and opinions, as this can influence how the interviewee responds.
Before starting any research, consider what assumptions do you have about the brief, your sources and related topics?
A Structured Approach
When interviewing, there are a couple of approaches we can take:
- Structured — survey-like approach, exact same questions and order to each person.
- Un-structured — radio interview like approach, spontaneous or semi-spontaneous questionioning.
A Semi-Structured Approach
For gathering someone's experience we recommend a semi-structured approach to interviewing. This means you will plan materials but allow for flexibility in having the interviewee guide the questions.
Building a Script
As part of our semi-structured approach, you will need a script. Included in this script should be:
- Asking for permission
- Setting expectations
- The questions
Asking for Permission
It is important that your interviewee understand the why and what you are doing. This ensures that they can safely agree to participate in what you are doing.
Before you ask interview questions, you must make sure they understand:
- How long it will take
- That they can leave at any time
- That their information is private
- What you will do with what they share
- Why you are doing this
Ethical Data Options
Given that cloud services often cannot guarantee the physical location of your data will stay in Canada, I would recommend working with SFU-based tools such as Canvas Groups or SFU Vault to store research data.
When exploring someone's experience, we likely want to ask open-ended questions as much as possible. For example:
Open-ended: "Can you tell me about your experience as a parent?"
Close-ended: "What three things do you love most about your experience as a parent?"
An approach to building questions is to consider what do you want to find out about. From there we can check our biases and try to build out a question that could give us the information we are looking for.
Another approach we can take to building questions is taking our assumptions and turning them into questions.
Review your questions:
- How open or closed in the question?
- How might the individual not answer the question?
- Is there any apparent bias in the question?
Sometimes words are not enough to effectively get the response you want.
Keep in mind that prototypes, props, materials, and the ability to generate a response with more than words can be useful.
There will be more than one of you on a team. As a result, it is important that everyone has a clear role during the interview:
- Lead interviewer
- Audio/video capture
Just like with the interview questions, it is important that notes you take check your biases. You should aim to capture unbiased explanations of the interviewee as much as possible.
Internal distortion happens in interviews when there are problems with:
- The interviewer (and their technique)
- The clarity of questions
- General ill-preparedness
These are things that you can control.
External distortion happens in interviews when there are problems with:
- The interviewee (and interacting with them)
These are things that you do not control as strongly.
Let's practice some 'live' setup of the space. I'll talk through some 'digital' considerations.
- Setting up the space
- Body language
Preparing for remote
There are a couple other items we recommend considering when preparing for remote interviews:
- Rehearse with a teammate or classmate beforehand.
- Pilot questions with a teammate or classmate. Have them explain how they think about answering your questions (i.e. think-aloud).
- Over-prepare rather than under-prepare.
Conducting remote research
For conducting the remote research, we recommend:
- Ask permission to record the interview. If given, record the interview.
- Make sure note-taker introduces themselves and then mutes their audio and video for the interview.
- If testing a prototype with tasks, make sure the participant has some time to get use to using the prototype before giving the tasks.
How to work with and make sense of research results we will talk about in the coming weeks.
To help offset some of the in-lab critiques, we are going to start using the second half of lecture for weekly crits starting June 15.
These crits will not be public and will require sign-up in advance. Attending the lecture crit time means you will be free from the lab crit time. There are five slots available per lab (D101, D102, D103).
You can start signing up today by messaging me over the course chat or via email.
Next week's lecture
Teeny Tiny Actions and Patterns
Exploration of more interaction and interface patterns as well as the introduction of microinteractions. Pre-recorded lectures and slides will typically become available the day of the lecture.