"Compared to students who reported sleeping eight hours at night, high school students who slept less than six hours were twice as likely to self-report using alcohol, tobacco, marijuana or other drugs, and driving after drinking alcohol. They were also nearly twice as likely to report carrying a weapon or being in a fight."
Have they established causation? Or could it be that lifestyle including alcohol and weapons also happens to involve less sleep?
Based on nothing but my own personal experience when I get tired I get cranky and more truthful, so the sole factor could be nothing but more people filling out the survey honestly.
YouTube, Facebook and others are basically social cancer.
Remember when they thought putting asbestos in everything and smoking everywhere were great ideas? This "social all the things" strategy reminds me a lot of 1950s America where plastic, fiberglass, and pesticides were going to cure all problems if only you used liberal amounts of it in everything.
"Compared to students who reported sleeping eight hours at night, high school students who slept less than six hours were twice as likely to self-report using alcohol, tobacco, marijuana or other drugs, and driving after drinking alcohol. They were also nearly twice as likely to report carrying a weapon or being in a fight. Researchers found the strongest associations were related to mood and self- harm. Those who slept less than six hours were more than three times as likely to consider or attempt suicide, and four times as likely to attempt suicide, resulting in treatment. Only 30 percent of the students in the study reported averaging more than eight hours of sleep on school nights."
Seems like a lot of data mining to create correlations (hence the use of the word "association" a lot). This kind of work can be good for problem finding and hypothesis generation. It's unfortunate that we live in a world where:
- news outlets publish research that is meant to stimulate discussion + hypothesis exploration
- readers are unable, undereducated, and / or disempowered to discern the level of rigorous science examination done
EDIT: The funny thing is that I predict that this HN thread will lead hundreds of folks to tell stories of confirmation bias.
Nearly every morning one of our local radio station gives "fun" "facts" and always start with "Scientist say " and end with something ridiculous like "dogs can't look up", or "purple is the best colour", or something equally bizarre, and they treat it like because "scientists" say it it has meaning or value, and then they both discuss it as a fact, and have people phone in and talk about it. We utterly fail at scientific education and communication in North America.
I’m not sure if it’s only NA though... just read the chapter 6 “Authority” of Cialdinis book “Influence”. People take stuff that authorities like scientists say as given. It’s human nature. The example of how this can be abused as outlined in the book are quite bizarre.
I wonder if this is CBC Radio 2 Drive. If not, there must be a daily syndicated list available to radio stations to have for banter between actual content.
>Those who slept less than six hours were more than three times as likely to consider or attempt suicide, and four times as likely to attempt suicide, resulting in treatment.
Considering that insomnia is a common symptom of depression, this really shouldn't be remotely surprising
Edit: I'd bet you'd find a similar correlation with students that sleep more than 12 hours a day too
Not to mention that depression often makes people feel physically tired all the time. I don't think someone with normal neurochemistry even could sleep 16 hours a day if they wanted to.
> EDIT: The funny thing is that I predict that this HN thread will lead hundreds of folks to tell stories of confirmation bias.
Good reply, but why rush to label potential replies as confirmation bias? That simply makes your reply more subjectively problematic. For all we know, the causation is really there. We just have no idea, for the exact reasons you've posted. It's ok to throw potential causes around.
The leverage point in a reply like yours is in getting people to interpret those replies with more clarity, which should not be confused with getting people to label a bunch of personal anecdotes as confirmation bias.
I might recommend using a word more specific than "problematic" if your goal is to communicate your criticism. It's about as specific as "bad", and seems to form an in-group out-group dynamic for readers who get to fill in the gaps.
I agree with your point, so I've read your comment as "it makes your reply more biased than necessary", whereas people who disagree with you might read it as "I don't like your reply because it hurts my feelings" or some other uncharitable interpretation.
Hope this came across as a friendly recommendation! Communicating over text has so many pitfalls.
> The funny thing is that I predict that this HN thread will lead hundreds of folks to tell stories of confirmation bias.
Confirmation bias looks like on-topic content, and often spurs copycat posts that look like engagement. It's one of the faster ways of growing the number next to your user name in the upper right, which causes a dopamine release in most users.
Another way of wording the finding is "students who engage in activities that adults don't approve of found to also go to sleep at a time adults don't approve of."
Well, there's pretty good controlled experimental evidence that lack of sleep causes impulsivity. But obviously causality works the other way too and this study doesn't help us know how much of these effects would go away if, say, high school start times were pushed back and how much would persist which is what we're really interested in.
It could be cortisol. I've been researching the effects of this hormone and it's kind of crazy. It's the hormone that regulates stress, fat deposits, memory and a lot of other things.
I would think that fighting, alochol and other drugs are being consumed by 6 hour sleepers during the two hours that the 8 hour sleepers are... asleep.
There is also the possibility that a third variable contributes to both. Like maybe a stressful family life or living in a city (louder at night, more access to drugs) causes teens to be more likely to use drugs and to sleep less.
The confusion of causation and correlation is often presented as a bug in the human operating system. It’s definitely a mistake here and you are right to point it out. But I sometimes wonder if the bug in the human cognitive process is (not mistaking causation vs correlation) but actually that we attribute greater weight to conclusions presented as science, even when they are bad science. We - rightly - trust the scientific process, more than for example testimony or hearsay. Disreputable scientists hack this heuristic to pass bad science as fact.
Hmmm. As a former grades 8-12 educator, if I were to observe 'risky teen behavior' I'd first want to talk to their parents and to ascertain their (compulsory) school environment.
Adolescents have enough to worry about (which they're good at), and their need for a caring support structure is a paramount concern. Experimentation is natural and healthy for them, within limits.
Many of these risky behaviors are likely to take place at night. Students who are out partying and drinking are likely to be doing this at night, which will cut down into their sleep time.
The implication of this study is that the lack of sleep is causing the risky behaviors. However, saying that risky behavior caused less sleep could also be as reasonable an implication.
I wish high school teachers would take note of this. My kid goes to an accelerated program high school in Toronto and she is so swamped with homework and tests, she hardly gets any sleep on weeknights and on some weekends too.
Take this study with a grain of salt. It only shows that teens who engage in risky behavior are more likely to get less sleep, not that less sleep leads to risky behavior in teens.
We know in small scale sleep lab studies that low sleep leads to changes in emotional processing (one of the functions of REM sleep is to tune your emotional regulation), so that at least offers a plausible mechanism of action. As far as risky behavior though, you're right that causation isn't yet firmly established here.
You can take them out of the accelerated/gifted program and put them back on the "normal" university track in high school.
The big risk is that it will cause your kid to get alienated socially among both the "normal" and the "gifted" kids, so make sure that your kid is involved in the decision-making process.
If your kid has an IEP, you can also talk to the teacher to get accommodations. For example, maybe your kid could quietly work on their homework during their school day, instead of being forced to pay attention to the teacher's lecture.
You can also just tell the kid that they can decide how much homework they want to complete. In high school and university, any un-completed work just gets recorded as a zero; if the kid is already getting 80s and 90s then a few missed or half-complete assignment questions shouldn't materially affect their final grade (provided they understand enough of the main concepts to complete major tests/projects/exams). Having a strategy for deciding what work is safest to drop/defer when you're overwhelmed is a useful skill both in upper education and in the workplace.
I wonder if similar experiments have been done in the military, where you have a good number of young people that basically sleep as much as you let them sleep, and that during the day are subject to (extreme) fatigue/training (the amount of which is anyway under your control), including some (allowed/part of the training) fights.
The military has a pretty good idea just how long they can keep people awake, and what the best combinations of stimulants are to keep them in warfighting condition. The Wehrmacht in WW2 used an awful lot of amphetamines, and dexedrine has been in common usage for a long time as well.
The US Air Force still uses 'go pills' (methamphetamines) during operations to keep people awake. I would expect them to prefer something like provigil, but that just might be a matter of time. It causes problems, though, when an airman hopped up on meth kills a bunch of ally troops as happened several years back with some Canadian forces.
We could be moving from 1% to 2% here. Technically true, but statistically meaningless. So while we could say that a teen who engages in risky behavior is less likely to have adequate sleep, we can't say that a teen that has inadequate sleep is likely to engage in risky behavior. And if we're looking to identify teens likely to engage in risky behavior, it's useless.
And there are other things to consider as well. The study was done with 67,615 participants. 70% who are sleep-deprived compared to 30% who are "well-rested". The population of "sleep-deprived" is much, much larger.
Also, we're looking at self-reporting. A bit unreliable in the best conditions.
And we're drawing a casual link here that may not actually exist. They both may be symptoms of a third problem. For instance, students in impoverished conditions may have more risky behavior regardless of sleep schedules. And if that was the real reason, forcing more sleep on people is hardly going to fix the problem.
I would expect the confounding variable in this study to be peer pressure. If your peers can pressure you into drinking alcohol or using marijuana, they probably peer pressure you into staying up late -- to do the aforementioned activities.
And the consequences of insufficient sleep on adult behavior? Child behavior? Elderly behavior? Perhaps this is a human trait, and not something isolated to the ever-examined and ever more heavily regulated adolescent?
Not that any of this is news. That adolescents need more sleep than any other age group, and that they suffer tremendously from being deprived it by adults, has been known for an extremely long time. When I was in high school in the 90s, there were dozens of studies about it even then. But adults don't care. The most common resistance to the notion of starting high school later is that it would interfere with football practice, making it a total non-starter. That football destroys brains, teaches mob mentality, and other such things doesn't matter. It is, to our society, far too important to let those other matters interfere.
> And the consequences of insufficient sleep on ...
Ahh.. the Copenhagen Interpretation of studies.
This particular study is about teenagers - perhaps because 'risky teen behaviour' is something you're more likely to find in teenagers, or, less churlishly, because risky behaviour in general is something you find much more often in teenagers.
Also compare age-segmented stats for depression, suicides, violence, etc - and they're an obviously higher at-risk demographic.
> Not that any of this is news. That adolescents need more sleep than any other age group, and that they suffer tremendously from being deprived it by adults, has been known for an extremely long time.
This is not true.
Outside of newborns / toddlers, pretty much everyone needs about 8 hours of sleep. Changes as you get (much) older make it more difficult to get the full 8 hours, and adolescents' circadian rhythms means that to the uninformed it seems like they want more / less / longer / lazier etc.
They don't - all the current research indicates they need ~8 hours too, but they don't naturally want to fall asleep until 2-3 hours after the average 25+yo.
> But adults don't care.
This is a breathtaking thing to say.
> The most common resistance to the notion of starting high school later is that it would interfere with football practice, making it a total non-starter.
I'd suggest this courageous hypothesis could at best apply to something less than 5% of the world.
66 comments
[ 2.6 ms ] story [ 138 ms ] threadHave they established causation? Or could it be that lifestyle including alcohol and weapons also happens to involve less sleep?
I'm only half joking.
Remember when they thought putting asbestos in everything and smoking everywhere were great ideas? This "social all the things" strategy reminds me a lot of 1950s America where plastic, fiberglass, and pesticides were going to cure all problems if only you used liberal amounts of it in everything.
Seems like a lot of data mining to create correlations (hence the use of the word "association" a lot). This kind of work can be good for problem finding and hypothesis generation. It's unfortunate that we live in a world where:
- news outlets publish research that is meant to stimulate discussion + hypothesis exploration
- readers are unable, undereducated, and / or disempowered to discern the level of rigorous science examination done
EDIT: The funny thing is that I predict that this HN thread will lead hundreds of folks to tell stories of confirmation bias.
Considering that insomnia is a common symptom of depression, this really shouldn't be remotely surprising
Edit: I'd bet you'd find a similar correlation with students that sleep more than 12 hours a day too
The inability to _want_ to do anything or _care_ often isn't a conscious choice.
Good reply, but why rush to label potential replies as confirmation bias? That simply makes your reply more subjectively problematic. For all we know, the causation is really there. We just have no idea, for the exact reasons you've posted. It's ok to throw potential causes around.
The leverage point in a reply like yours is in getting people to interpret those replies with more clarity, which should not be confused with getting people to label a bunch of personal anecdotes as confirmation bias.
I might recommend using a word more specific than "problematic" if your goal is to communicate your criticism. It's about as specific as "bad", and seems to form an in-group out-group dynamic for readers who get to fill in the gaps.
I agree with your point, so I've read your comment as "it makes your reply more biased than necessary", whereas people who disagree with you might read it as "I don't like your reply because it hurts my feelings" or some other uncharitable interpretation.
Hope this came across as a friendly recommendation! Communicating over text has so many pitfalls.
Confirmation bias looks like on-topic content, and often spurs copycat posts that look like engagement. It's one of the faster ways of growing the number next to your user name in the upper right, which causes a dopamine release in most users.
Although I somehow doubt the parents' issue is their kids go to bed promptly at eleven but then insist on getting up at five in the morning.
And suicide, for instance, is a bit more than just "activities that adults don't approve of". It's self-evidently harmful.
Adolescents have enough to worry about (which they're good at), and their need for a caring support structure is a paramount concern. Experimentation is natural and healthy for them, within limits.
The implication of this study is that the lack of sleep is causing the risky behaviors. However, saying that risky behavior caused less sleep could also be as reasonable an implication.
https://www.psychologytoday.com/us/blog/dream-catcher/201112...
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3163498/
You can take them out of the accelerated/gifted program and put them back on the "normal" university track in high school.
The big risk is that it will cause your kid to get alienated socially among both the "normal" and the "gifted" kids, so make sure that your kid is involved in the decision-making process.
If your kid has an IEP, you can also talk to the teacher to get accommodations. For example, maybe your kid could quietly work on their homework during their school day, instead of being forced to pay attention to the teacher's lecture.
You can also just tell the kid that they can decide how much homework they want to complete. In high school and university, any un-completed work just gets recorded as a zero; if the kid is already getting 80s and 90s then a few missed or half-complete assignment questions shouldn't materially affect their final grade (provided they understand enough of the main concepts to complete major tests/projects/exams). Having a strategy for deciding what work is safest to drop/defer when you're overwhelmed is a useful skill both in upper education and in the workplace.
Source: Went to a high school in York Region.
We could be moving from 1% to 2% here. Technically true, but statistically meaningless. So while we could say that a teen who engages in risky behavior is less likely to have adequate sleep, we can't say that a teen that has inadequate sleep is likely to engage in risky behavior. And if we're looking to identify teens likely to engage in risky behavior, it's useless.
And there are other things to consider as well. The study was done with 67,615 participants. 70% who are sleep-deprived compared to 30% who are "well-rested". The population of "sleep-deprived" is much, much larger.
Also, we're looking at self-reporting. A bit unreliable in the best conditions.
And we're drawing a casual link here that may not actually exist. They both may be symptoms of a third problem. For instance, students in impoverished conditions may have more risky behavior regardless of sleep schedules. And if that was the real reason, forcing more sleep on people is hardly going to fix the problem.
Not that any of this is news. That adolescents need more sleep than any other age group, and that they suffer tremendously from being deprived it by adults, has been known for an extremely long time. When I was in high school in the 90s, there were dozens of studies about it even then. But adults don't care. The most common resistance to the notion of starting high school later is that it would interfere with football practice, making it a total non-starter. That football destroys brains, teaches mob mentality, and other such things doesn't matter. It is, to our society, far too important to let those other matters interfere.
Ahh.. the Copenhagen Interpretation of studies.
This particular study is about teenagers - perhaps because 'risky teen behaviour' is something you're more likely to find in teenagers, or, less churlishly, because risky behaviour in general is something you find much more often in teenagers.
Also compare age-segmented stats for depression, suicides, violence, etc - and they're an obviously higher at-risk demographic.
> Not that any of this is news. That adolescents need more sleep than any other age group, and that they suffer tremendously from being deprived it by adults, has been known for an extremely long time.
This is not true.
Outside of newborns / toddlers, pretty much everyone needs about 8 hours of sleep. Changes as you get (much) older make it more difficult to get the full 8 hours, and adolescents' circadian rhythms means that to the uninformed it seems like they want more / less / longer / lazier etc.
They don't - all the current research indicates they need ~8 hours too, but they don't naturally want to fall asleep until 2-3 hours after the average 25+yo.
> But adults don't care.
This is a breathtaking thing to say.
> The most common resistance to the notion of starting high school later is that it would interfere with football practice, making it a total non-starter.
I'd suggest this courageous hypothesis could at best apply to something less than 5% of the world.