Right. As their own statistics show, the average age in the groups from least to most use increases from 13.8 to 15.8, and there's a big separation by grade level in terms of use that supports that observation, too. That's a pretty big gap in terms of developmental status and, bluntly, levels of teenage angst, so perhaps while there is clear
correlation here, there is not
causation. I think this is the clincher - "Conversely, studies conducted among university students found no such relationship [between mental health and SNS use]." What's the main difference? Lack of the steep developmental range seen in early teens. They also openly state this:
"The cross-sectional nature of the data precludes evaluation
of temporality and causality of the observed relationship
between use of SNSs and mental health problems."
So, when the press run with "Social media causes brain damage in children, think of the children!", remember what the paper actually says.
Personally, I think from anecdotal experience that heavy SN use fucks you up no end, but these findings don't prove this.