My issue is the quality of the research. IMHO not by fault of the researcher, but because of the subject itself.

The first result that isn't a guideline or call to action when I search "preschool children screen time research"

https://www.sciencedirect.com/science/article/pii/S074937972...

On the recruitment:

> Note: Eligible families were recruited primarily through paid social media advertisement (83%) as well as other online sources (6%) and word-of-mouth/other (11%).

So most of this experiment was done with families who click ads on Facebook/Instagram. The results are relevant for that specific public, but tougher to expand to other families and environments.

That's one specific study, but I really read a lot, and this kind of bias is common. Perhaps because they need families that are stressed about screen time in the first place ?