Misleading Context (Facebook Expansion Study)
Contrary to assertions in the Social Media and Mental Health paper, mood disorders among young adults did not start rising in mid-2000s.
We saw in Facebook Expansion: Invisible Impacts? that effects of the social media explosion circa 2004 fail to be noticeable in 2000-2008 depression trends per the National College Health Assessment (NCHA), the very survey used by the authors of Social Media and Mental Health (SMMH) to conclude that the expansion of Facebook in 2004-05 had almost immediate (next semester) effects on the mental health of college students.
We also saw that the SMMH authors failed to provide any NCHA trend data, leaving readers without sufficient context.
The authors did, however, assert context despite their omission of relevant trends information — and in the process they severely misled their readers.
The mid-2000s
Given that the main results asserted by Social Media and Mental Health are the immediate effects of Facebook expansion 2004-05 on mood disorders among college students, the obvious context of this study should be mental health developments of young adults during the mid-2000s.
And that is the context the authors indeed seem to be providing, at the beginning of their Social Media and Mental Health paper, when they state:
As social media started gaining popularity in the mid-2000s, the mental health of adolescents and young adults in the United States began to worsen (Patel et al. 2007; Twenge et al. 2019).
This notion is reinforced when the authors later declare that our study specifically targets the population (young adults) that experienced the most severe deterioration in mental health in recent decades and studies it around the time in which those mental health trends began to worsen — the time here being 2004-06 [emphasis mine].
The authors then assert that their findings are in line with the hypothesis that social media has a negative impact on mental health and played a role in the increase in mental illness among adolescents and young adults over the past two decades.
This all of course directly implies that the major significance of the the Social Media and Mental Health study is that it helps to explain why mood disorders among young adults began to worsen in the mid-2000s.
Mood Disorders Rise Among Young Adults
The problem is that no evidence I know of supports the notion that mood disorders among young adults began to rise in the mid-2000s — and all the evidence I know of points to early-to-mid 2010s as the beginning of the rise.
Wait, readers may object that the SMMH authors included two references, Patel et al. 2007 and Twenge et al. 2019, and so must have proven their assertion.
No.
Patel 2007 has nothing at all on mental health trends in this millennium and Twenge 2019 includes only post-2008 trends for young adults.
Furthermore, the mood disorders data in Twenge 2019 point to early-to-mid 2010s as the start of the rise, just as all the other data I’ve ever seen.
Indeed even the very first figure in the online Appendix of the SMMH paper illustrates that young adult disorders started to rise substantially only in mid-2010s:
The Social Media Theory Paradox
I previously pointed out that one of the fundamental problems with blaming the doubling of adolescent mood disorders on social media is that this rise started long after the social media explosion among teens 2003-2006 (see A Preview of Main Concerns).
Indeed major proponents of the social media theory, psychologists Jean Twenge and Jon Haidt, have repeatedly endorsed 2012 as the approximate start of the mental health decline among adolescents (see The Teen Mental Illness Epidemic Began Around 2012).
And yet we now have three economists (the authors of Social Media and Mental Health) telling us that mood disorders began to rise in mid-2000s, nearly a decade before what the psychologists have been telling us — and what the actual data is telling us.
Such misinformation makes the problem of the long delay between social media explosion and mood disorders rise go away for these three economists — but not for science.
Conclusion
The authors repeatedly misinform readers by declaring the time of the Facebook expansion — the mid-2000s — to coincide with the beginning of the mental health declines among young adults. This misinformation provides a false context for the study results, implying they help to explain developments that in reality occurred much later.