Better qualitative analysis
Doing more with your research data
Analyzing qualitative data is one activity that sets good designers apart from great designers.
It can be challenging to pinpoint what defines excellent analysis. Sometimes it can feel like somewhat of a “black box” and literature on the topic is scarce. We hope this summary might inspire you to do more with your qualitative analysis.
What does this mean to me?
It is impossible to separate yourself from your data and your analysis. Acknowledge this and ask these questions to understand better how you relate to your material:
- How am I, as a designer reacting to this phenomenon or situation, and what does that tell me?
- What stakes (interests, benefits, risks) do I have in the study’s outcomes?
- Whose side am I on? (This might feel scary to admit, but it is not possible to be 100% unbiased).
Everything is data
Include more than what was explicitly said during the interview in the analysis:
- What did the participant do during the interview? How do you, as a designer, interpret this?
- Did the participant smile?
- Did the participant look away?
What is this a case of?
One very effective way of creating better analysis is to make an effort to generalize concepts:
- Ask yourself how you can generalize what you have observed.
- Explicitly ask yourself — what is this a case of?
- Are there similar things I have heard/observed that are applied differently?
- Describe what you have heard/observed without any of the specific markers associated with the case (names, titles, organizations, setting)
Find the anomaly
Look out for data that doesn’t fit your preconceptions. When clustering, we look for familiarities in data. We need to be careful to find patterns that lay outside of what we expect:
- What is unexpected?
- What contradicts your clusters? (comment on it)
- Is there a difference between participants’ self-image and actual behavior?
Don’t forget about the gestalt
Gestalt — an organized whole that is perceived as more than the sum of its parts. The risk of clustering is that it makes the full picture difficult to overview:
- Take a step back and breathe.
- Try always to remember that there is a bigger story behind the clusters.
- Make notes on the broader context and insert them into the analysis.
Read between the lines
What is not said? You can find this out in two ways:
- Ask during the interaction.
- Interpret the silence yourself.
Write more
One often under-utilized way of deepening analysis is formulating and refining the description of research findings:
- Create a working title for your project that describes what you have learned so far. Update the title when you know more.
- Consider the difference between writing “the participants claimed that…” and “the participants said that…”
Read more
The more general knowledge we as designers have of the world, the easier we connect what we learn in a research project to other things that have happened in history, technology, and society. We can then use that understanding to understand our research in new ways.
Be curious!
The Course
This material is made as a companion poster for my Skillshare course on the same topic; please check it out if you want more context. It is currently featured on the front page of Skillshares UX/UI section, or you can reach it through this link:
(subscription required after trial)
(if you leave a review that really helps me out)
The podcast
The course, is based on an episode of the Podcast, Designing the Robot Revolution that I co-host with David Griffith Jones.
The episode can be found on any podcast app.
Apple podcast:
Spotify:
