> Everything in here runs completely contradictory to the body of science from the past decade.

I'd say it doesn't so much contradict the existing science as it suggests the possibility that the population-level correlations we see are not explained by direct effects of social media use on individuals.

Just as an illustrative example, suppose Alice has an higher-than-average need for in-person social interaction in order to remain mentally healthy, and so has decided not to use social media. If the people around Alice start spending more than some critical percentage of time on social media instead of in-person socializing, then her mental health might decline even though she doesn't use social media.

Yup, anecdotal evidence: I'm pretty much Alice. I avoid social media, but I can't avoid its effects - since everyone else doesn't, what social media does also indirectly affects me. It definitely affects me negatively how people around me have goldfish attention spans, it definitely affects me when they try to drag me into nonsensical trends, etc...

Great example! Its hard to think of a topic more complex and obtruse than social interactions. There are going to be no straightforward answers here because social media is a conduit. Its like arguing whether home telephones are bad; its not really the phone’s fault if some arsehole keeps calling you or if it helps you stay in touch with loved ones - its merely the tech facilitating social interaction.

I can't find any study looking for a direct link between social media use in children and depression. Most articles I can find can only find correlations, but there's no indicator of if depressed kids start to use social media more or if kids using social media become depressed.

I have found articles linking social media to worse sleep quality (and sleep quality and depression are well known to influence each other) but still no direct link.

Do you happen to know what papers prove the link between depression and social media use in children? The theory makes sense but I can't find studies that prove more than mete correlation.

Biased sample size as well. Most of the individuals in this study made well over 100k. An important factor to note: "Children spend most of their waking hours in day care or in school; therefore it is important to study how aspects of these environments can affect children’s development and functioning. "

This is not the average case scenario as the majority of children on social media and the longest amount of screen use are of children who are often not in after school programs or daycares. The reality is that this study is not a good measure of if social media causes depression or not.