How can we accurately research child personas?

Alongside our concepts for the design sprint, our team is working on researching some initial user personas which we can use to refine concepts and better understand the contexts and scenarios in which our final concept might be used.

An interesting issue that has arisen while I've attempted to do this has been how the research process might differ when the user in question is a child and not an adult. Our team has decided to focus on children who are between the ages of 4-8, and at this age it is hard to determine what a child’s needs, wants and motivations first hand through interviews or other participatory techniques. Antle (2008:156) notes that field work in this respect usually relies on interpreting information gleaned from second-hand sources such as the adult care-givers and teachers of young children.

An additional problem is that we are not trained in child development, and even if we were to use first-hand observation techniques (e.g. through watching YouTube videos), it would be difficult to interpret a young child’s behaviours and verbal expressions accurately.

Finally, there is the fact that we are under particular time and monetary constraints. It would be great to employ a variety of different techniques to gain a more first-hand understanding of child users (e.g. getting them to draw responses instead of interviewing them), but to do so in the limited time that we have would be near impossible. 

How then can we create an accurate child-persona given these constraints?

This has been something that I have been thinking about this week. So far I have been relying on using a variety of sources in conjunction with each other (triangulation): scholarly articles on designing for children, existing literature reviews on cognitive and physical development in early childhood, YouTube videos, as well as real-life publications and websites intended to aid teachers and the existing school curriculum.

While it all sounds very rough at the moment, as we continue to refine and test out concepts, I’m hoping that the user personas I am researching at the moment start to flesh out a bit more.