I’ve been lucky enough to work on a lot of kids games over the years, from Wonderbook, Invizimals and PlayStation Vita Pets in my time at PlayStation, to many more games working with popular childrens IP since (uncredited now that I’m external!).

When moderating these sessions, it becomes very clear straight away that playtesting games for kids requires a specialist moderation style and skill-set, to avoid wasted sessions and failing to answer your research objectives.

To explore this topic further, I invited Walker Tate to come and give a talk for the Games User Research Fundamentals course earlier this year. Walker has been a user researcher on kids-focused titles such as LEGO Star Wars: The Skywalker Saga, Disney Pixar Cars 3 and Sesame Street: Once Upon a Monster. He shared some best practices based on his experience on how to make the most from playtests with kids. 

In this article, I’ve picked out some of the lessons Walker covered, and some of the ways you can apply this to your own studies to get high quality data from your playtests with children. 

Focus on behavioral data

With any participant, behavioral data is a lot more reliable than opinion data. When we see a usability issue, (such as the player missed the objective) we objectively know that a defect exists in the intended experience and it’s safe to take action to resolve it. Opinions are mor nuanced – ‘some players didnt like it’ will be just one of many opinions to balance, and requires more context and thought on how to decide what to do about it.

This effect is magnified 10x when running a test with kids. Most children will be less confident than an adult participant, which will impact the opinion data they give – they will be less likely to tell you things they don’t like, and much more likely to say what they think you want to hear.  They may also find it difficult to understand the questions being asked, especially when topics get more abstract. They will tell you they love the game, and struggle to separate ‘the game’ from ‘the experience of doing playtesting’.

When planning your study with kids, look for behavioral ways of testing your objectives, and focus on ‘what do players do’, rather than ‘what do players think’, to draw more confident (and representative) conclusions from your study.

Kids haven’t learned UI conventions

There are many conventions we expect gamers to have learned. How to use dual-stick controls. That yellow paint highlights where to go. That red barrels explode when shot. Many conventions become a common language within games, and we no longer think about actively teaching them to players.

Many child participants won’t have had the same conventions or experience – and a lot of early gaming is performed on touch-screens for younger children. This means that the conventions they have learned will be different from ours as developers, and we don’t share a common language for ‘how to communicate information’.

This has two impacts on us as researchers. First we need to broaden our research objectives, and make sure that we’re consciously looking for confusion about the UI or assumed gaming knowledge.

The second impact is that this reinforces how critical it is to test kids games with real players regularly. The user research mantra ‘you are not your players’ is especially true with this audience, as our understanding of how games work is so far apart from kids – testing becomes essential. 

Scheduling logistics

It’s obvious when you think about it – but children aren’t available during a normal 9 – 5 workday. Outside of school holidays, you’re stuck testing with kids in evenings and weekends (and children go to bed early).

This means that studies can take longer than anticipated – in order to see enough participants, you’ll be stuck with 2-3 hours of playtime on weekdays, and the weekends, making it take longer to see an appropriate number of participants.

When scheduling your playtests with kids, make sure your conscious of when they will be available, and plan your schedule accordingly, to make sure that participants can actually turn up!

Children tire easily

Children will struggle to give their extended attention to a playtest, which will impact the length of a session (and how much content we can cover). 

I’ve observed that there is a hype curve for children taking part in tests – an initial burst of excitement about ‘the idea of testing a game’, followed by disappointment when it’s not the same as playing their favorite game at home, and then a gradual improvement as they get invested in the game they are playing. 

Keeping sessions short (no more than 2 hours), having a break in the middle, and creating opportunities for people to use the bathroom (as they might not feel brave enough to ask), will help maintain comfort levels and engagement. This will help you make the most from the time you spend with players.

Children are very diverse

Within even very small age ranges, there can be a big difference in children’s cognitive ability and emotional intelligence, which will have a direct impact on their behavior and how they approach games – creating additional usability issues and opinions.

This is important to recognise when recruiting participants (especially as some studies imply that gender has an impact on cognitive development too). Consider being very specific in your recruitment criteria – going beyond just age, but also looking at more objective measures of defining and screening your target audience (perhaps on reading age, or other measures of development).

Surveys will be difficult for children to answer

Surveys can be very abstract – requiring you to convert a sentiment (such as ‘how I feel about this game?’) to a number isn’t a natural process and is a learned skill. I’ve observed that children often struggle with giving ratings, or differentiating between different aspects of a game (e.g. asking a rating for ‘what the game looked like’ and ‘whether i liked what you do in the game’). They will instead give every question the same rating.

Walker also highlighted how inconsistent reading skills can be in this age range, making surveys increasingly difficult for some participants.

To get high quality data, avoid using surveys with children, and instead focus on a qualitative discussion to understand what aspects of the game are resonating with players. 

To encourage natural qualitative data, without having to rely on asking too many direct questions, consider recruiting children in friendship pairs. Their natural conversation will reveal a huge amount about their current understanding of how the game works, what’s confusing to them, and what aspects are resonating with them.

Consent, Safeguarding and Behavioural Management

Research with children introduces additional safeguarding requirements, and we need to be careful to ensure we’re running ethical research. The exact rules for consent will differ based on the country you are in, but a good principle is to recruit parent + children together, and require the parents to sit in the session (with some magazines to keep them occupied). 

Having parents present not only alleviates some safeguarding concerns (because a parent or guardian is always with them), but can also help with behavior management if the participants get bored and act out. 

Testing your own games with children

If you’re running your own playtests, these tips will help you make the most from your time with participants. You can see Walker’s full talk (and learn the entire end-to-end games user research process) from the games user research fundamentals course – you can join the waitlist here. 

If you’re building a game and need your game tested, I’ve been running playtests with children for over a decade – you can see more about my games user research studies, and book an intro call here.