User testing with children

Trine Falbe looks at the challenges of involving children in the design process

If there is one thing I’ve realised over the past years working with kids and digital things in combination, it’s that kids are the most unpredictable user group I’ve ever come across. They are goofy, playful and have a very short attention span. Their intuition drives them, and they often shift focus. These things mean you need to take a specific approach when testing with this user group.

I’ve found it incredibly useful to adopt an ethnographic approach. Put simply, ethnographic research means observing people in a natural setting and describing what they do. This can result in many different types of data — for instance user portraits and patterns of behaviour.

Ethnographic research works particularly well with kids because they have such different mindsets to adults. Their playfulness is hard to capture with more structured research methods. If there is one thing to keep in mind when planning to involve kids it is this: you cannot plan much, and the plans you make will most likely fail.

Fly on the wall

This means you have to do a few things differently. For one, loosen the structure. Secondly, use a video camera to record your observations. Kids will jump from here to there in a split second, so it is impossible to document your data at the level of detail you need if you don’t use a video camera.

I recently did a study with a bunch of kids to map their practice patterns when using the web through computers. Ultimately, the aim was to create a set of design principles for making web stuff for kids.

My observations took place in an after school care group and in a class with a third grade — both settings were natural and on-site, which is crucial. I planned by simply writing down a few interview questions (“Do you use a computer? For what? Do you use a tablet? For what?”). Other than that, my goal was to simply video record the kids while they were using the computers.

Because there is no way you can get a group of kids to do a structured walkthrough for you, instead of asking them to “show me what you do” (a common ethnographic observation approach), I acted like a fly on the wall.

I went home with 60 minutes of raw footage, which gave me incredible insight on how the kids actually used computers and the challenges it represented. In never would have gained such insights if I had simply interviewed the kids or done usability test-type sessions with them.

Thinking aloud

Another common way of involving users in the design process is by doing usability testing ‘think aloud’ sessions. Again, doing this with kids is a different world. I learnt this the hard way: the first time I did a usability test with two kids, my plans simply failed because the kids were really bad at thinking aloud. They were also bad at following instruction, so they simply didn’t work with the tasks I had set. I didn’t video record the sessions, which turned out to be a bad idea, too.

I learnt that you have to let the kids play with the interface and probe them along the way if you want them to test something specific (e.g. “Try to use X feature on this page”). I also realised it was a good idea to have them test the interface twice, mainly because they are so fast and spontaneous in the way they interact. Testing twice gave me a chance to observe in more detail.

My point is that when you work with kids, you can’t plan too much; you have to follow their lead and see where it takes you. And you have to video record your research so you have material to analyse afterwards. Forget about keeping a tight structure — jump on the train and see where it leads you. And don’t forget your video camera.

Denmark-based Trine has been working with the web since 2001. She is a senior lecturer and UX consultant, and has spoken at various conferences in Europe


This article originally appeared in issue 272 of net magazine.