6 ms·

So 1 in 3 felt happier, 1 in 3 felt more depressed, and the remaining 1 in 3 felt no major change. Sounds pretty much like a normal (gaussian) distribution, or

So 1 in 3 felt happier, 1 in 3 felt more depressed, and the remaining 1 in 3 felt no major change. Sounds pretty much like a normal (gaussian) distribution, or basically a non-story. Guessing if they measured satisfaction on a scale, the majority were clustered near the "no major change" camp, as with any gaussian curve. As we already know, it turns out humans can adapt to things. By phrasing the headline result as "1 in 3 felt happier," it creates a narrative where there likely isn't one. You imagine the 33% were this solid block of people feeling much happier. But, re: the gaussian distribution, its likely very few were at the extremes, and most were clustered towards the middle. Textbook clickbait. It invites you to react with your emotional, anecdotal feelings about what this means to you, as if there was any interesting finding there.

huijzer 5y ago
> So 1 in 3 felt happier, 1 in 3 felt more depressed, and the remaining 1 in 3 felt no major change. > > Sounds pretty much like a normal (gaussian) distribution, or basically a non-story. For a Gaussian, I would have expected 1 felt worse, 6 felt the same and 1 felt better. The current numbers sound more like an uniform distribution.
lemmsjid 5y ago
The article contextualizes itself at the beginning. A large number of people believe the effect of the lockdowns on children was not a normal distribution but was skewed negative. So a normal distribution is news. This does not discount your argument that emphasizing the positive side of the distribution is clickbait, but if it is positioning itself as emphasizing that positive side to a population who thinks it is all negative, then I think it’s fine. For me textbook clickbait is an article that does not substantiate the promise of the headline, or one that comically exaggerates the excitement of the information in the article. There are people who believe the lockdowns were wholly an atrocity on children, and this article substantiates a counterbalance without going all the way (after all, the headline itself tells you that 2/3 children were unaffected or negatively affected).
- bavell 5y ago
  > The highest proportions of students who reported improved mental wellbeing were among those who were in school every day (39%) and most days (35%), while the highest proportion of students who reported worse wellbeing were those who attended just once or twice (39%). Seems that it absolutely skews negative for the kids who had to deal with more lockdowns, i.e. not going to school with their peers. > So a normal distribution is news. I don't see anywhere in the article where they claim it's a normal distribution? OP was just speculating IIUC.
  - _Nat_ 5y ago
    > Seems that it absolutely skews negative for the kids who had to deal with more lockdowns, i.e. not going to school with their peers. That's incorrect. If you look at Figure-1 from [the article](https://link.springer.com/article/10.1007/s00787-021-01934-z https://link.springer.com/article/10.1007/s00787-021-01934-z ), it appears that students were happier when they attended either "Not at all" or "Every day". It's the middle-ground options that averaged out to show lower-satisfaction. Or to frame it differently, students who never went to school in-person reported a net-positive average: 33.5% better vs. 32.7% worse (vs. 33.8% the same). That doesn't support the notion that students didn't like staying home. That said, I'd warn against over-interpreting such small differences. --- Separately, if we did want to speculate based on the data: we might guess that it was about consistency. Students appeared to be happiest when they attended not-all-all or every-day, with lower improvements in middle-ground options. Comparing the two extremes, there is a preference for every-day over not-at-all (stressing that we're talking about smaller figures here), but even that might be explainable in terms of consistency: because, before the pandemic, students did attend every-day -- so not-at-all was a major shift in lifestyle for those students, i.e. a major inconsistency. Hypothetically, if students always attended online, but then some had to start attending in-person every-day, we might see the reverse bias, as then it'd be the students who'd have to attend having the major shift in lifestyle. --- All that said, there appear to be more underlying factors. For example, if we look at Table 2, it appears that there was a more considerable gender-gap: * males were happier (2400 vs. 1836); * females weren't (3185 vs. 3861). That's a large enough relative-disparity as to suggest that it may be a mistake to interpret the data in a gender-blind way; the little on-average differences seem overshadowed by the gender-differences.
civilized 5y ago
Come on. Would you really have predicted, before a massive two-year pandemic, that it would have a mean-zero random impact on people's well-being? Let's keep ourselves honest here.
- pembrook 5y ago
  I thought it was a well recognized fact that humans adapt--I remember this even from my old high school psych class. There's been a thousand studies about this (becoming a refugee, poverty vs. being rich, etc etc). For most people, the only thing that has the power to permanently change your mental state is the death of someone very close to you (a spouse, child, etc) or forced exile from an identity forming social tie (eg. divorce, getting fired)
  - pawelmurias 5y ago
    It's not a commonly held belief. Why would anyone want to be rich or afraid of being poor of a refuge if you could adapt to anything?
    - lillecarl 5y ago
      Because adapting sucks, but we adapt well to many things. The fear of change can often be worse than change itself.
marcosdumay 5y ago
> or basically a non-story Are you kidding? Of all the possible consequences from completely changing the daily life of people "no overall change on happiness" was the least plausible one.
zwieback 5y ago
There's been rising awareness that researchers refrain from publishing neutral or "boring" results, this is a real problem because it skews the expectations not only of researchers but also the general public. So, for me this is not a non-story and, in fact, somewhat unexpected. I'm hoping it'll trigger more in-depth research into what made those 33% happier. Some of the reasons are obvious but I suspect there's more nuance.
kolbe 5y ago
It’s also a survey, which is largely useless in the discovery of real information. Whereas in my county, ER visits for teen attempted suicides are up over 100% from pre-Covid. That seems more like actual evidence that something is wrong here
rmbyrro 5y ago
As any newssian distributuon would predict: journalists fabricating stories to sell clicks.
bjornsing 5y ago
> Sounds pretty much like a normal (gaussian) distribution, or basically a non-story. Depends on your perspective I think. Where I live (Sweden) the general sentiment was that young people would kill themselves in droves if there was a lockdown. So here this should be news, I’d say.
manigandham 5y ago
This isn’t serious research. It was a survey of students, from June/July 2020. Barely a few months of lockdown by then.
causi 5y ago
A negative event doesn't cause a normal distribution like that. When a hurricane floods a city a third of the citizens don't feel happy about it. The story is the fact there is a normal distribution, thus prompting an examination of why a massive restriction of movement wasn't a purely negative event.
theptip 5y ago
> basically a non-story I don't think this is the right take. I'm going to venture that pretty much everyone thinks it was worse than this, i.e. that the more young people feel depressed than happier in COVID-time. So "it's actually on average no effect" very much is a "story", it would be something that would cause most people to substantially update their models of the world if it's sound research. Put differently, "1 in 3 young people felt depressed" isn't surprising at all. "1 in 3 young people felt happier" is surprising to me. So leading with the surprising bit makes sense for a lede. All of this is meta-level, I have no opinion on the object-level validity of the research.
yodsanklai 5y ago
> Sounds pretty much like a normal (gaussian) distribution, or basically a non-story If your interpretation is correct, this means the center of the distribution is "neutral", but it could very well not be the case.
Terry_Roll 5y ago
>Textbook clickbait. LOL I'd say your post is!
nemetroid 5y ago
> Guessing if they measured satisfaction on a scale, the majority were clustered near the "no major change" camp, as with any gaussian curve. They did measure satisfaction on a scale. From page four of the report: > Students responded on a sliding scale of 0–100 with five labelled categories, ‘Much worse’, ‘Slightly worse’, ‘The same’, ‘Slightly better’, and ‘Much better’, anchored at scores 0, 25, 50, 75, and 100, respectively. We simplified the scale in this analysis by transforming it into groups of those doing ‘worse’ (scores 0–37.5 on the sliding scale), ‘the same’ (scores 37.6–62.5), or ‘better’ (scores 62.6–100) (N.B. categories are not evenly divided due to how the scale is presented in the survey; please see https://osf.io/fpbt3/ https://osf.io/fpbt3/ for an example of the survey scales). Fitting a normal distribution with 1/3 of the volume in the 37.6–62.5 range ends up with 1/5 of respondents above the "Slightly better" mark (i.e. above 75).
_Nat_ 5y ago
The article calls out exactly one quote: > The common narrative that the pandemic has had overwhelmingly negative effects on the lives of children and young people might not tell the full story [-Emma Soneson] They appear to be arguing against that narrative rather than for the opposite.
karaterobot 5y ago
I mean, I wouldn't expect the distribution to be normal. It would also be interesting if the headline was "1 in 3 young people say they don't mind getting run over by a truck". You'd think things which are generally viewed as negative would tend towards a biased distribution, so if that's not the case I'm interested.
jollybean 5y ago
"or basically a non-story." - that it's a 'Gaussian distribution' does not mean it's a 'non story'.a

huijzer 5y ago

> So 1 in 3 felt happier, 1 in 3 felt more depressed, and the remaining 1 in 3 felt no major change. > > Sounds pretty much like a normal (gaussian) distribution, or basically a non-story. For a Gaussian, I would have expected 1 felt worse, 6 felt the same and 1 felt better. The current numbers sound more like an uniform distribution.

lemmsjid 5y ago

The article contextualizes itself at the beginning. A large number of people believe the effect of the lockdowns on children was not a normal distribution but was skewed negative. So a normal distribution is news. This does not discount your argument that emphasizing the positive side of the distribution is clickbait, but if it is positioning itself as emphasizing that positive side to a population who thinks it is all negative, then I think it’s fine. For me textbook clickbait is an article that does not substantiate the promise of the headline, or one that comically exaggerates the excitement of the information in the article. There are people who believe the lockdowns were wholly an atrocity on children, and this article substantiates a counterbalance without going all the way (after all, the headline itself tells you that 2/3 children were unaffected or negatively affected).

bavell 5y ago
> The highest proportions of students who reported improved mental wellbeing were among those who were in school every day (39%) and most days (35%), while the highest proportion of students who reported worse wellbeing were those who attended just once or twice (39%). Seems that it absolutely skews negative for the kids who had to deal with more lockdowns, i.e. not going to school with their peers. > So a normal distribution is news. I don't see anywhere in the article where they claim it's a normal distribution? OP was just speculating IIUC.
- _Nat_ 5y ago
  > Seems that it absolutely skews negative for the kids who had to deal with more lockdowns, i.e. not going to school with their peers. That's incorrect. If you look at Figure-1 from [the article](https://link.springer.com/article/10.1007/s00787-021-01934-z https://link.springer.com/article/10.1007/s00787-021-01934-z ), it appears that students were happier when they attended either "Not at all" or "Every day". It's the middle-ground options that averaged out to show lower-satisfaction. Or to frame it differently, students who never went to school in-person reported a net-positive average: 33.5% better vs. 32.7% worse (vs. 33.8% the same). That doesn't support the notion that students didn't like staying home. That said, I'd warn against over-interpreting such small differences. --- Separately, if we did want to speculate based on the data: we might guess that it was about consistency. Students appeared to be happiest when they attended not-all-all or every-day, with lower improvements in middle-ground options. Comparing the two extremes, there is a preference for every-day over not-at-all (stressing that we're talking about smaller figures here), but even that might be explainable in terms of consistency: because, before the pandemic, students did attend every-day -- so not-at-all was a major shift in lifestyle for those students, i.e. a major inconsistency. Hypothetically, if students always attended online, but then some had to start attending in-person every-day, we might see the reverse bias, as then it'd be the students who'd have to attend having the major shift in lifestyle. --- All that said, there appear to be more underlying factors. For example, if we look at Table 2, it appears that there was a more considerable gender-gap: * males were happier (2400 vs. 1836); * females weren't (3185 vs. 3861). That's a large enough relative-disparity as to suggest that it may be a mistake to interpret the data in a gender-blind way; the little on-average differences seem overshadowed by the gender-differences.

civilized 5y ago

Come on. Would you really have predicted, before a massive two-year pandemic, that it would have a mean-zero random impact on people's well-being? Let's keep ourselves honest here.

pembrook 5y ago
I thought it was a well recognized fact that humans adapt--I remember this even from my old high school psych class. There's been a thousand studies about this (becoming a refugee, poverty vs. being rich, etc etc). For most people, the only thing that has the power to permanently change your mental state is the death of someone very close to you (a spouse, child, etc) or forced exile from an identity forming social tie (eg. divorce, getting fired)
- pawelmurias 5y ago
  It's not a commonly held belief. Why would anyone want to be rich or afraid of being poor of a refuge if you could adapt to anything?
  - lillecarl 5y ago
    Because adapting sucks, but we adapt well to many things. The fear of change can often be worse than change itself.

marcosdumay 5y ago

> or basically a non-story Are you kidding? Of all the possible consequences from completely changing the daily life of people "no overall change on happiness" was the least plausible one.

zwieback 5y ago

There's been rising awareness that researchers refrain from publishing neutral or "boring" results, this is a real problem because it skews the expectations not only of researchers but also the general public. So, for me this is not a non-story and, in fact, somewhat unexpected. I'm hoping it'll trigger more in-depth research into what made those 33% happier. Some of the reasons are obvious but I suspect there's more nuance.

kolbe 5y ago

It’s also a survey, which is largely useless in the discovery of real information. Whereas in my county, ER visits for teen attempted suicides are up over 100% from pre-Covid. That seems more like actual evidence that something is wrong here

rmbyrro 5y ago

As any newssian distributuon would predict: journalists fabricating stories to sell clicks.

bjornsing 5y ago

> Sounds pretty much like a normal (gaussian) distribution, or basically a non-story. Depends on your perspective I think. Where I live (Sweden) the general sentiment was that young people would kill themselves in droves if there was a lockdown. So here this should be news, I’d say.

manigandham 5y ago

This isn’t serious research. It was a survey of students, from June/July 2020. Barely a few months of lockdown by then.

causi 5y ago

A negative event doesn't cause a normal distribution like that. When a hurricane floods a city a third of the citizens don't feel happy about it. The story is the fact there is a normal distribution, thus prompting an examination of why a massive restriction of movement wasn't a purely negative event.

theptip 5y ago

> basically a non-story I don't think this is the right take. I'm going to venture that pretty much everyone thinks it was worse than this, i.e. that the more young people feel depressed than happier in COVID-time. So "it's actually on average no effect" very much is a "story", it would be something that would cause most people to substantially update their models of the world if it's sound research. Put differently, "1 in 3 young people felt depressed" isn't surprising at all. "1 in 3 young people felt happier" is surprising to me. So leading with the surprising bit makes sense for a lede. All of this is meta-level, I have no opinion on the object-level validity of the research.

yodsanklai 5y ago

> Sounds pretty much like a normal (gaussian) distribution, or basically a non-story If your interpretation is correct, this means the center of the distribution is "neutral", but it could very well not be the case.

Terry_Roll 5y ago

>Textbook clickbait. LOL I'd say your post is!

nemetroid 5y ago

> Guessing if they measured satisfaction on a scale, the majority were clustered near the "no major change" camp, as with any gaussian curve. They did measure satisfaction on a scale. From page four of the report: > Students responded on a sliding scale of 0–100 with five labelled categories, ‘Much worse’, ‘Slightly worse’, ‘The same’, ‘Slightly better’, and ‘Much better’, anchored at scores 0, 25, 50, 75, and 100, respectively. We simplified the scale in this analysis by transforming it into groups of those doing ‘worse’ (scores 0–37.5 on the sliding scale), ‘the same’ (scores 37.6–62.5), or ‘better’ (scores 62.6–100) (N.B. categories are not evenly divided due to how the scale is presented in the survey; please see https://osf.io/fpbt3/ https://osf.io/fpbt3/ for an example of the survey scales). Fitting a normal distribution with 1/3 of the volume in the 37.6–62.5 range ends up with 1/5 of respondents above the "Slightly better" mark (i.e. above 75).

_Nat_ 5y ago

The article calls out exactly one quote: > The common narrative that the pandemic has had overwhelmingly negative effects on the lives of children and young people might not tell the full story [-Emma Soneson] They appear to be arguing against that narrative rather than for the opposite.

karaterobot 5y ago

I mean, I wouldn't expect the distribution to be normal. It would also be interesting if the headline was "1 in 3 young people say they don't mind getting run over by a truck". You'd think things which are generally viewed as negative would tend towards a biased distribution, so if that's not the case I'm interested.

jollybean 5y ago

"or basically a non-story." - that it's a 'Gaussian distribution' does not mean it's a 'non story'.a