4 ms·

I agree with the authors premise and argument but the data visualizations are not well done, and I'm confused about the conclusions he draws from them. For sta

I agree with the authors premise and argument but the data visualizations are not well done, and I'm confused about the conclusions he draws from them. For starters, it's unclear why he draws a line at 2012, when in many of the graphs the slope gets steeper at 2010 instead, and in fact that's where the percentage increase calculations start from as well. That vertical line at 2012 is misleading and confusing. Also, in the US Teens, Suicides (Ages 10-14), the uptick clearly started in 2007. I don't disagree that social media is bad and I've seen firsthand people who become mentally ill because of the standards and expectations supplied by a constant stream of highlight rules that make you question your own worth. But I think that the data provided here is not strong enough to come to that conclusion.

williamcotton 4y ago
I think if you do a moving average that the trend would adhere more to 2012 being the inflection point. I'm basing that on just noticing that in US Teens, Suicides (Ages 10-14) chart the there was a decent amount of fluctuation in the leading years. As for why these trends exist, one of the interesting things is that these are present in all of the anglosphere. Can we think of another explanation? What else has changed on a global scale in the last couple of decades?
zeroonetwothree 4y ago
For boys suicides 10-14 it sure looks like it started in 2007, yeah. For girls the 1 year increase from 2012 to 2013 is kind of suspicious, so I could see an argument for 2007 as well. That does seem to decrease the strength of the argument for social media specifically since I don't believe it was all that widespread until the 2010s. I think the point is that you can't just draw a single year as the change, which obviously makes sense with any social trends. It's not like people wake up Jan 1 and suddenly they have depression.

williamcotton 4y ago

I think if you do a moving average that the trend would adhere more to 2012 being the inflection point. I'm basing that on just noticing that in US Teens, Suicides (Ages 10-14) chart the there was a decent amount of fluctuation in the leading years. As for why these trends exist, one of the interesting things is that these are present in all of the anglosphere. Can we think of another explanation? What else has changed on a global scale in the last couple of decades?

zeroonetwothree 4y ago

For boys suicides 10-14 it sure looks like it started in 2007, yeah. For girls the 1 year increase from 2012 to 2013 is kind of suspicious, so I could see an argument for 2007 as well. That does seem to decrease the strength of the argument for social media specifically since I don't believe it was all that widespread until the 2010s. I think the point is that you can't just draw a single year as the change, which obviously makes sense with any social trends. It's not like people wake up Jan 1 and suddenly they have depression.