5 ms·

The article makes a strong claim, based on evidence, that social media and not anything else is the real culprit. Why is it wrong? What evidence supports your a

by dash2 4y ago

The article makes a strong claim, based on evidence, that social media and not anything else is the real culprit. Why is it wrong? What evidence supports your alternative proposed causes?

WithinReason 4y ago
The article points out correlation, which would be also explained by GP's comment
- treis 4y ago
  It's not even really correlated. The timing is off. Teen Social Media usage exploded in the early to mid 2000s and all the bad trends started 2010 or later. Plus, it was a seismic event. We went from little social media in 2000 to ubiquity by 2010. We should have seen one big rise to a new steady state not a continually increasing problem.
  - looknee 4y ago
    Social media definitely picked up 2000-2010, but the current manifestation of it didn't really lock in until ~2010 with the advent of the iPhone, instagram, snapchat, and all the other social media mobile apps that compelled people to constantly be sucked into it throughout the day rather than just whenever they were on or near a PC or laptop.
  - dash2 4y ago
    The graph here shows social media exploding from 2008, not before. https://www.worldline.technology/post/over-2-5-billion-people-use-social-media-this-is-how-it-has-changed-the-world https://www.worldline.technology/post/over-2-5-billion-peopl...
    - zeroonetwothree 4y ago
      This is showing all users worldwide. Surely US teens were ahead of the curve.
  - zeroonetwothree 4y ago
    Exactly. By 2018 around 95% of teens used social media. We would thus expect the trends to have spiked more before then rather than after. Instead we see a linear trend.
s1artibartfast 4y ago
The author mentions in his testimony, that the correlation between social media use and mental illness may be around r=0.10. This means in general terms that it it explains about 10% of of the impact. This leaves a lot of room for additional factors, and factors with much greater impact.
- dash2 4y ago
  Actually it explains 1% of the impact, since the R-squared will be 0.01. But there's a huge amount of random noise in whether people get depressed, which we would always struggle to explain. A measure of the effect size is more interesting than a measure of R^2.
10amxn10 4y ago

by dash2 4y ago

The article makes a strong claim, based on evidence, that social media and not anything else is the real culprit. Why is it wrong? What evidence supports your alternative proposed causes?

WithinReason 4y ago

The article points out correlation, which would be also explained by GP's comment

treis 4y ago
It's not even really correlated. The timing is off. Teen Social Media usage exploded in the early to mid 2000s and all the bad trends started 2010 or later. Plus, it was a seismic event. We went from little social media in 2000 to ubiquity by 2010. We should have seen one big rise to a new steady state not a continually increasing problem.
- looknee 4y ago
  Social media definitely picked up 2000-2010, but the current manifestation of it didn't really lock in until ~2010 with the advent of the iPhone, instagram, snapchat, and all the other social media mobile apps that compelled people to constantly be sucked into it throughout the day rather than just whenever they were on or near a PC or laptop.
- dash2 4y ago
  The graph here shows social media exploding from 2008, not before. https://www.worldline.technology/post/over-2-5-billion-people-use-social-media-this-is-how-it-has-changed-the-world https://www.worldline.technology/post/over-2-5-billion-peopl...
  - zeroonetwothree 4y ago
    This is showing all users worldwide. Surely US teens were ahead of the curve.
- zeroonetwothree 4y ago
  Exactly. By 2018 around 95% of teens used social media. We would thus expect the trends to have spiked more before then rather than after. Instead we see a linear trend.

s1artibartfast 4y ago

The author mentions in his testimony, that the correlation between social media use and mental illness may be around r=0.10. This means in general terms that it it explains about 10% of of the impact. This leaves a lot of room for additional factors, and factors with much greater impact.

dash2 4y ago
Actually it explains 1% of the impact, since the R-squared will be 0.01. But there's a huge amount of random noise in whether people get depressed, which we would always struggle to explain. A measure of the effect size is more interesting than a measure of R^2.

10amxn10 4y ago