The original research I present in iGen finds that teens who spend more time on screens are less happy and more depressed (in a large, nationally representative sample of U.S. teens). For example, 8th graders who spend 10 or more hours a week on social media sites are 56% more likely to be unhappy than those who spend less time. The link holds when gender, race, and socioeconomic status is taken into account. But those analyses are correlational, so it is possible that unhappy or depressed teens spend more time on screens.
Several longitudinal studies find that digital media use comes first, and then depression or unhappiness, suggesting at least some of the causation goes from digital media use to depression/unhappiness. Three true experiments, which can show causation, come to a similar conclusion: College students who limited their social media use (vs. those who did not) were less depressed and lonely after three weeks; adults who gave up Facebook for a week (vs. not) were less depressed and happier; and those who web-surfed were happier than those who looked at their friends’ Facebook pages. Jon Haidt and I are reviewing the evidence (correlational, longitudinal, and experimental) for links between digital media use and mental health in this Google doc.
Also: Depression causing social media use doesn’t explain why depression would increase so suddenly after 2011-12. In that model, something else would have to cause teen depression to rise so sharply, which would then lead to more smartphone and social media use. It seems much more likely that smartphone and social media use increased, and depression and unhappiness followed.