If this is just a correlation this would have to be a correlation at the population level. Countries where kids start school later on (e.g. 7 years old) have significantly lower rates of myopia than countries that start school early on in a child’s development (e.g. 3 years old). It’s still possible that this is a correlation, but the correlation would have to be capturing something deeper than just an individual kids screen time. Granted, this correlation would still need to account for differences between individual kids, but it would also need to account for differences between kids at a population level. It’s hard to see what could be causing this correlation though. So maybe there’s something there we’re just not seeing, but at a certain point though the idea that there is a causal relationship starts to seem like the most plausible explanation for explaining this data