What do you make of this?
If you start at 2009, suicidal thoughts reported among teens have gone up by more than half. But if you look at the entire data series, suicidal thoughts declined by half during the '90s and aughts and have now merely reverted to their historical average.
It's all especially hard to figure out when you add this:
The number of teens who seriously attempted suicide has been roughly flat for 30 years. There's a bit of a dip in the aughts, but that's about it. Even the COVID pandemic appears to have had only a small effect.
Finally, the actual number of suicides is also fairly stable. It spiked in 2014-17 but has been declining ever since. COVID had no effect at all.
I'm not sure what to think of this, but it's the kind of thing that makes me a little skeptical of the whole teen mental health crisis narrative. If, over the past decade, teens say they think about suicide more but they don't actually attempt suicide more, what's the explanation? A genuine increase in depression? A culture that encourage teens to talk about depression more? Just a bit of random movement in a data series that goes up and down over time? I don't know.
Well, they sure hear about it more. It stands to reason it would prompt a thought about it.
No they don't. We cant even say the words on most social media without it bring deleted
this chart maps the collapse of grunge & rise of emorap.
it's the cobain-xxxtentacion curve.
1999 is the inflection down when limp bizkit played woodstock iii & 2011 is the inflection up when odd future played sxsw.
+1
Oh dear lord. I think it was '97 when I finally stopped following pop music on accounta it all starting to sound the same. So limp bizkit is my extreme outer range. At la la palooka this year I did not recognize a single band.
I think the most popular think I listen to these days is Priscilla Stewart's version of Mecca Flat Blues Hey, I've always been midlist ... and I apologize for nothing 🙂
Kevin, do any of Jonathan Haidt's charts move you?
Not saying their should or shouldn't. Just curious.
Just rename this blog "Kevin Drum tells you why nothing matters, with charts" already.
As a parent of a kid who was hospitalized after attempting suicide, a coach of young folks, and not just reading numbers I can tell you that not only is there a much higher rate of depression and suicide but a shitty system for dealing with it.
As we all know, anecdotes are not statistics.
Wait a minute. Eyeballing it, the second chart starts at what looks to be 0.008 and peaks at 0.012. That’s a bump of 50% in actual suicides. No sh*t people are upset. Wrong chart, Kevin. This one actual hides the information.
Is it time to move on from this blog 🫤
With so few suicides, that's not alot. A minor shift in methods (more guns, for instance) would create a difference in result.
Could also be minor changes in attention to the problem and finding and collecting the information.
Honestly speaking, I feel like there has been more result-oriented analyses like this one in the past few years. More bad trendline fits (inflation!), more high-leverage outliers, more deployment of y-axes that start at zero or not as befits the argument, more arguments dependent on carefully selected time-series endpoints.
it does seem that way
It gets better.
Again, you'e averaging good places with bad.
IDK, looks like suicide ideation, attempted suicide, and successful suicides have all gone up since hitting bottom in 2008-2009, and the flattening in successful suicides is because there's been an increase in kids getting treatment.
Data visualization would be cleaner if instead of %, you did X per 1000.
This is disingenuous for three reasons:
First, the 1990s were the last teen suicide peak.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1586156/
Arguing that teen suicide today is comparable to teen suicide in the 1990s is like arguing that murder rates today are like murder rates in the 1990s. If you find yourself having to make that argument, you should just concede the point.
Second, suicide is an act of impulse. As you have repeatedly noted, teens appear far less impulsive today. They commit less crime, are less likely to get pregnant, are less likely to try illegal drugs, etc. You have attributed this comparative restraint to a decrease in lead poisoning.
If suicide is diverging from these other favorable trends, that should be concerning.
Third, suicide is an act of opportunity. Suicide attempts that involve guns are typically successful. Suicide attempts that do not involve guns are rarely successful. But the number of households with a gun has dropped dramatically between the 1990s and today. So fewer teens have access to guns. If, in spite of that, they are killing themselves at comparable rates, that should be concerning.
Yeah, I noticed the 90s thing as well. I'm surprised Kevin presented these stats without comment, given his best-known theme.
The charts certainly seem to show that something is going on over the last 15-20 years. This change would appear very dramatic if the charts showed % changes
The argument that the changes over the last 15-20 years are not meaningful because its still not as bad as a few data points from 30 years ago doesnt make a lot of sense when its used to discount what is happening right now.
It is possible to have 2 different time periods of crisis. The current one doesnt have to be the all time worst situation ever for it to still be a crisis.
I have personal acquaintance with some of the data points involved here. My son started thinking about suicide before he was a teen, attempted it while a teen, at 19, and succeeded when no longer a teen, at 20.
In the eternal tragedy vs. statistic war on this blog, maybe this is a battle you shouldn't have chosen to fight.
Just guessing - the Santa Ana strip mall photo is not from a posh part of town. Actually I didn't know Santa Ana even had such an far unposh area.
Well my guess would be if the trend is down for 15+ years and suddenly it begins to increase giving back most of the gains that is ACTUALLY REALLY BAD YOU CHILDLESS CAT LADY.