[flagged]
> very very dishonest
It isn't dishonest. You know what's dishonest? Calling it "5 minutes of Excel". Jonathan Haidt isn't in the business of selling substack. He's a professor at NYU.
I think the most basic thing is correlating with wealth, or other aspects of a livid environment. “Urban life engenders both early smartphone acquisition and more mental issues (at least self reporting)” is one hypothesis. Another is “social network usage correlates to more mental issues, and having smartphones gives more access to this” (which is relevant! But should be checked. Is it the phone? Is it social networks?)
The point being that you are trying to establish a causal narrative, and right now there’s not much to chew on.
If we want to help children we should dig deeper to understand actual causes instead of simply reporting the horizontal lines we found.
But it's good to view social sciences sceptically. Their theories don't offer much in terms of truth. It's best if they keep their theoretical judgement to themselves, and see where the research takes them. Psychological and sociological theories are not a good basis for policy.
The data don't lie, though. Now, there may be issues with the way it was gathered and whathaveyounot, but Haidt isn't exactly the first to observe a large increase in mental health issues. As a matter of fact, the papers have been mentioning it for years by now, as have the health organizations. The correlation from the article can almost not be denied. It's like climate change: you can try to find an alternative explanation (which will bring you into the exact same problems the psychological research you criticize is facing), but until then, we must be cautious.
The fact that you don't like messenger is simply not enough to ignore the message.
I can't say for certain but I imagine that voluntarily visiting a website to complete a mental health survey might appeal more to those with mental health issues thus skewing the data. Also you may have some people that visit to purposely give false data. Very much not unheard of in internet polls. I have no real idea how much either of these have affected the data but it makes me very suspect of the data. If you can't trust the data you can't trust any studies based upon it.
I am appalled at the willingness to make huge logical leaps on the basis of emotions or morality in this thread.
We can however easily imagine what's happening in this particular case, and the data support it. The damages are enormous. Like climate change, continuing is not an option, nor is waiting until we've got conclusive evidence, because that will take ages.
Making claims based on scientific method (not even theory or findings, because we have them, and they support the tenor of the article) is not helpful.
I have yet to see credible data regarding causation and I prefer that large societal decisions not be made based upon imagination. Comic books, pinball, rock n roll, violent video games at some point were all claimed to be destroying our youth and there was "data" presented to back it up.
Teen drug, alcohol, nicotine and sexual activity are significantly down since 2009. Maybe social media is causing that too and on the whole is a benefit. Or maybe teens are more depressed because they aren't getting high, drinking, smoking or screwing enough.
> Like climate change
Climate change has actual quality data and science behind it, so it is nothing alike.
I don't disagree, but
> I prefer that large societal decisions not be made based upon imagination
with that attitude, no decision can ever be taken. Social sciences can't produce the evidence you need. It's correlations all the way.
But also:
* Not acting is also taking a decision, because social media isn't going to stop any time soon. How's there evidence for just letting things continue?
* How is stopping children (until, say, 18) a large societal decision? That's just minor. There isn't even a risk involved. Children didn't have social media until 10-15 years ago at all. Did that stump their progress or something?
Because it is like that, I would say both of these conclusions is not scientifically acceptable: the fact that some people believe that "obviously it's not the case" is enough to demonstrate that it is in reality not obvious.
It does not mean it is impossible in social science to define causation: it is done in a lot of place, where there is a majority consensus on what is "obvious". The fact that this happened does not mean that you can conclude causation on everything you want.
But isn't this equally admitting smartphones are the issue? Like whatever deeper cause you're talking about existed before the smartphone. Like imagine you're a farmer and you grow crops that need a lot of water and climate change causes water to dry up in your area. What you're saying is basically "climate change didn't cause your crops to fail because you weren't growing drought resistant crops. Climate change is just a confounding effect on a deeper issue."
I think what makes the discussion about smartphones unique vs. other social topics is that the adoption of the smartphone and social media was so swift that it's very easy to study mental well-being before the smartphone and after the smartphone without worrying about other changes in lifestyle.
Not really: a confounding effect means that there is a cause X that leads to two things: using more smartphone, and depression. If you ban the smartphone, you don't stop the cause X, and you will still have depression.
> "climate change didn't cause your crops to fail because you weren't growing drought resistant crops. Climate change is just a confounding effect on a deeper issue."
That is not a confounding effect. Confounding is: X -> A + X -> B, and if you break the link between X and A, X -> B is unaffected. Your example is X -> A -> B, with X = climate change, A = drought, B = non-drought resistant plant die.
A confounding effect with climate change would be: imagine you're a farmer and you grow crops that need a lot of water and you use the electricity from the hydroelectric central 100 miles away. What you are saying when you say that smartphone are the issue is like saying "the crops are dying and the central cannot produce electricity anymore, so if I use photovoltaic panel, the crops will not die anymore".
("crop dying" = depression, "producing electricity" = not using smartphone)
> I think what makes the discussion about smartphones unique vs. other social topics is that the adoption of the smartphone and social media was so swift that it's very easy to study mental well-being before the smartphone and after the smartphone without worrying about other changes in lifestyle.
That does not make sense: if you are arguing that smartphones increase global depression, then you are saying that smartphones induced a change in lifestyle. If smartphone can do it swiftly, then other things can do it too.
For instance, people are saying “people who get phones earlier may have less involved parents, and less involved parents lead to worse mental health outcomes.” Sure, this is possible, but there have always been poor parents, yet these changes to mental health have only started with adoption of the smartphone.
> if you are arguing that smartphones increase global depression, then you are saying that smartphones induced a change in lifestyle. If smartphone can do it swiftly, then other things can do it too.
Yes, that’s exactly my point! It theoretically could be something else, but what else has so drastically changed in the last 15 years. We see plenty of evidence of how dramatic smartphone adoption has been. What else has experienced such a fundamental in adoption?
And now we’re back to Occam’s razor. Sure, the mental health crisis could be caused by some unseen force that has profoundly impacted life yet has so far eluded observation, or it could be smartphones, whose profound impact on lifestyle is easily observable.
They started with adoption of the smartphone, which also corresponds with a specific and brand new phase in globalisation, in climate change crisis, in global loss of trust in the media, in disillusion in the capacity of changing elements that matters, in faster access of overwhelming communication, a more open communication about personal issues (gender, ...), ...
It's incorrect to say that "it has to be smartphones because smartphones are the only thing that appeared", it's just really obtuse, and also a bit of Dunning-Kruger.
> but what else has so drastically changed in the last 15 years. We see plenty of evidence of how dramatic smartphone adoption has been. What else has experienced such a fundamental in adoption?
Smartphone adoption was a dramatic adoption that is visible for the layman, but in sociology, it's a very unimpressive thing. There are TONS of technology that were introduced that fast, from domestic electricity usage to internet, rap music or video games. But layman people are victim of Dunning-Kruger effect: smartphone is very visible to them, so they don't realise to which extent smartphone is in reality important or not.
And again, if I have to bet, I would say it is rather social media that is the cause, and if you ban smartphones, people will just use laptop and the problem will be the same.
This is insufficiently discussed, which invites the reader to dismiss the premise. This is bad, because it might also be true. In essence, highlighting it's own criticism a bit more strongly would have gotten the point across better.