Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why some researchers think I’m wrong about social media and mental illness

jonathanhaidt.substack.com

Post reply on HN
Print view View original

471–480 of 580 posts

Re: Why some researchers think I’m wrong about social media and mental illness

#471

Post by caddemon »

standardUser wrote: post #205

> In social media research, we focus on “how much social media did a person consume?” and we plan our experiments accordingly...Most don’t even distinguish between platforms, as if Instagram, Facebook, Snapchat, and TikTok are just different kinds of liquor. This is the problem I have with studies of "social media". It's like saying "the internet" causes problem, or "computers". These are overly broad categories and…

When it comes to teens and their friend groups it's even more complicated than that. The same basic activity on the same platform could be worlds apart depending on the real world group dynamics.

Still, I think at a certain point (with 4 hours per day surpassing that point IMO) it's hard to argue the activity remains beneficial. I'm not sure how someone would have time to do an actual hobby and spend a literal 4 hours per day on hobby social media groups, unless they're so wealthy they don't need to work, or they're not taking care of their health e.g. sleeping too little.

So I don't think it would be a terrible public health thing to encourage people to set personal social media time limits (on average), even if it may be premature scientifically to say much specifically. Of course it's critical to not pretend that research is further along than it actually is in any sort of proactive public health PSA.

Re: Why some researchers think I’m wrong about social media and mental illness

#472

Post by GuB-42 »

Decabytes wrote: post #255

I wonder how much of the negative aspects of social media are due to the gamification of it as well as the algorithms that boost certain types of content. For example, if Twitter had no algorithm, didn't show you the number of views on your post, didn't show you the number of likes on your posts or retweets, and you just used it to post funny memes, interesting articles, and life updates to your followers if it would…

An IRL-only social media site could actually make the problem worse, especially for teen girls.

One common hypothesis for why social media are bad for mental health of teens is that there is no escape. All the terrible things kids do at school (like bullying) now happen 24/7, with little oversight. Forcing people to meet in real life to engage with each other can actually make the problem worse by making it harder for outcasts to find a more supporting network.

Re: Why some researchers think I’m wrong about social media and mental illness

#473

Post by random314 »

selimthegrim wrote: post #445
rayiner wrote: post #379

Earlier quoted context omitted.

Are you seriously comparing non-passing trans people to drag performers?

You do realize in the subcontinent that transpeople interact with kids all the time in the street? I remember meeting them in my family’s neighborhood in Karachi when I was 7 or 8. You might want to ask your dad about this if you don’t believe me.

[dead]

Re: Why some researchers think I’m wrong about social media and mental illness

#474

Post by Karellen »

lukeasrodgers wrote: post #401
Karellen wrote: post #260

Earlier quoted context omitted.

Yeah, that seems completely absurd. If the experiments are rigorous enough, and the p value is low enough, you should be able to trust the results, no matter how small the effect is. If, for example, research shows that social media use increases one's risk of depression by, say, between 32% and 34%, with p can only show a 1/3 rise in risk, rather than a 3x rise? That's not skepticism. That's dogma.

Yes but in very large datasets, isn't it the case that you end up finding correlations for many things at very low p values? That is (one of the reasons) why effect size matters. Still, the "odds ratios are below 3" requirement seems excessive.

If you're just looking for any correlations, then p-values aren't really applicable due to the sharpshooter fallacy.

https://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy

You need a specific hypothesis to test before you start examining (or better, collecting) your data for a p-value to make any sense.

Re: Why some researchers think I’m wrong about social media and mental illness

#475

Post by DoreenMichele »

burnished wrote: post #469
DoreenMichele wrote: post #455

Furthermore, when you zoom in on girls, the relationships are not small: girls who spend more than 4 hours a day on social media have two to three times the rate of depression as girls who spend an hour or less. The common refrain “correlation does not prove causation” is certainly relevant here, but I showed that when you bring in the three other kinds of studies, the case for causation gets quite strong. I know fro…

You really should have gotten through the rest of the article, it addresses everything you bring up and more.

I've skimmed his argument and it boils down to "I have a hypothesis explaining this phenomenon and you don't."

It doesn't address anything I've asserted. I can think of several possibilities off the cuff but that's not likely to go good places.

Everything I know about human psychology says "Contrary to popular opinion, family life is much more formative and influential for legal minors than their peer group."

Edit:

I will add that where their peer group is a very strong influence it's because the family has already failed them.

Re: Why some researchers think I’m wrong about social media and mental illness

#476

Post by kodah »

1827162 wrote: post #94

I think it's the total visibility that's causing mental illness. There is no space where you can be with your friends, screw up, and not have it become known to everyone in your peer group. There's little space to make mistakes. No space to be yourself. You are under the watchful gaze of your friends, i.e. peer surveillance, nearly all of the time. Which induces stifling conformity. Everything's under constant scruti…

> So your "friends" in essence, force you to use these platforms. If you don't you will be treated as a social outcast, a pariah. It makes you wonder what these "friends" really are in the first place.

I think this is truthy, but I abandoned social media a long time ago and my experience is different. At first, I lost touch with a lot of people and my world became small. After some time people found me (I still run a blog, so I'm discoverable) and now folks will call and visit with me. I get a lot more authentic conversations with folks because they know I'm not going to put them on blast. I still have my judgements and my opinions like anyone, but those, I think, are easier for people to deal with.

Re: Why some researchers think I’m wrong about social media and mental illness

#477

Post by whats_a_quasar »

caddemon wrote: post #462
amelius wrote: post #46

Earlier quoted context omitted.

It's still just guessing. Can we please go from "plausible" to "we studied this"?

It's pretty hard to study beyond doubt though. You're not going to get much parental consent (or probably even institutional approval) to run a study where you require the experimental group of teenagers to spend 3+ hours a day on social media. And if the experimental group limits screentime, well I think parents of teens know how hard that can be to actually enforce across devices if the kid really wants to get on s…

It has been studied, with ~46 studies using three sorts of approaches. Haidt's writeup is below. His conclusion is that the available evidence shows that social media causes mental health issues, not the other way around. Though you can look through the studies yourself if you'd like.

https://jonathanhaidt.substack.com/i/104255435/question-a-do...

Re: Why some researchers think I’m wrong about social media and mental illness

#478

Post by whats_a_quasar »

raxxorraxor wrote: post #16

> girls who spend more than 4 hours a day on social media have two to three times the rate of depression as girls who spend an hour or less. Not that it proves anything, but the converse argument, that depressed girls spend more time on social media or online in general, should also be examined. If the data could support this thesis to a similar degree, there is no conclusive statement about social media really induc…

It has been studied, with ~46 studies using three sorts of approaches. Haidt's writeup on causation. His conclusion is that the available evidence shows that social media causes mental health issues, not the other way around. Though you can look through the studies yourself if you'd like.

https://jonathanhaidt.substack.com/i/104255435/question-a-do...

Re: Why some researchers think I’m wrong about social media and mental illness

#479

Post by caddemon »

Deegy wrote: post #239
standardUser wrote: post #205

> In social media research, we focus on “how much social media did a person consume?” and we plan our experiments accordingly...Most don’t even distinguish between platforms, as if Instagram, Facebook, Snapchat, and TikTok are just different kinds of liquor. This is the problem I have with studies of "social media". It's like saying "the internet" causes problem, or "computers". These are overly broad categories and…

This is a great point and it's where I struggle to moderate my own social media consumption. Is aimlessly scrolling my twitter feed a net benefit to my overall mood and well being? I'm 95% sure it's not. Is spending 20-30 minutes browsing hacker news every morning a net benefit to my overall mood and well being? I'm not sure.

Yeah it's hard to know precisely, it's probably not a big negative either though - focusing on cutting down the aimless scrolling is probably a better priority than figuring out the effects of 30 minutes of HN (easier said than done ofc).
Post reply on HN