I wasn’t aware that there is a media campaign to discredit this finding. Teen suicides exploded in growth YoY the moment smartphones became cheap enough around 2009.
That isn't an accurate picture you paint, at least in the UK: https://www.ons.gov.uk/visualisations/dvc2132/heatmap/datado...
Teen suicides first peaked in 1998 at 6.1 per 100,000. The numbers are noisy but it looks like it declines to a lower level of 3-4 in the 2005-2013 period (with an all time low of 3.1 in 2010 but still low at 3.6 at 2012), then starts to increase again to the 4-5 range, before covid gives it a giant spike to 6.5.
Teen suicides in the 90s were higher than the 21st century at between 5.0 and 6.1.
Social media and cellphones, if they are a factor, aren't dominant enough to be distinguished from the noise.
The parallel being drawn here is that social media platforms, being presented with the consequences of their products ie: teen mental health declines with use, are complicit in those consequences at some point in the value chain, either by marketing, product design, accessibility, etc. If the assumptions hold, then the conclusions are sound and valid.
There's two ways to disagree with this. Either the assumptions are not valid, or the argument is not sound.
The effects are most obvious with teens but I see this stuff as bad all around. Social media algorithms have really driven today's insane political polarization, and there's lots of examples of adults having their brains sucked out by social media rabbit holes.
Here's one pile of examples:
https://www.reddit.com/r/QAnonCasualties/
It seems pretty obvious to me at this point that if you connect a bunch of people together and then artificially prioritize content to maximize engagement, the result is unbelievably toxic.
This phenomenon isn't fundamentally new. Media has understood "if it bleeds it leads" for hundreds of years. What I think is new is how tight the optimization loop is, how personalized it is, and the way crowdsourcing the inputs leads to a firehose of content that isn't even attempting to be accurate or sane. It's a machine that automatically curates randomly sourced content for maximum inflammatory response and maximum addictiveness.
Here's another pile of examples unrelated to conspiracy theories:
https://www.reddit.com/r/ElsaGate/
This is what you get when you run the recommendation algorithm on random crowdsourced kids' content.
So is it really related to smart phones? What was causing the decrease?
0: https://www.childtrends.org/publications/teen-suicide-databa...