4 ms·

There is a Cochrane review [1] on the health effects of reducing time spent sitting at work that came out two years ago, which laid out the truth in no uncertai

There is a Cochrane review [1] on the health effects of reducing time spent sitting at work that came out two years ago, which laid out the truth in no uncertain terms: "The quality of evidence is low to very low for most interventions, mainly because of limitations in study protocols and small sample sizes." And still people keep doing shit like this, performing studies with fifteen people?? [1] https://www.cochrane.org/CD010912/OCCHEALTH_workplace-interventions-methods-reducing-time-spent-sitting-work https://www.cochrane.org/CD010912/OCCHEALTH_workplace-interv...

anonytrary 8y ago
The last company I worked for had a "market research" report about mobile data users that claimed 75% of people do X, the rest do Y. The sample size was 4. What's worse is that the report was beautifully styled. It's amazing how much money can be wasted on moot work when you have funding.
- cryoshon 8y ago
  my experience is that marketing people care nothing for quality or veracity of "facts". if it looks good, it goes.
dalbasal 8y ago
Idk the exact answer, but I doubt the reason is "researchers don't understand the statistics." It is some "how the sausage is made" reason, most likely. For example: researchers have the budget/resources to to a preliminary study. They use the interesting results of this study to secure funding for a larger sample size study. Btw, a small sample size study isn't useless necessarily. You can refine methodology, with less on the line. You can also (more or less) validly exclude certain conclusions, to avoid wasting resources on a wider study. The problem is with pop science headlines, like it is in any other journalistic field. I think it's less of a problem though, than many think. Publishing in the popular press needs to be understood as "academics working on X" not "X has been proven."
- koboll 8y ago
  >The problem is with pop science headlines, like it is in any other journalistic field. I think it's less of a problem though, than many think. Publishing in the popular press needs to be understood as "academics working on X" not "X has been proven." Writing a headline that misrepresents a single study's conclusion as "proven" or what "scientists say" -- especially when the conclusion is contradicted by most others on the subject -- should be considered by journalistic ethics watchdogs as an extreme violation bordering on an outright lie. The media seems simultaneously bewildered as to why the public no longer trusts it, while incapable of recognizing and responding properly to obvious breaches of the public trust like inflated science headlines.
  - dalbasal 8y ago
    Its not just science, it's anything. Also, it's not just media, it's everything. The only thing to do is individual. Take "facts" with a grain of salt, at the very least until you have read the article. Do not take headlines as facts. You can just read academic journal abstracts, if you want a more academic and precise reading materials. There are trade offs between rigour and the digestible content intended to be browsed over coffee.
    - jcutrell 8y ago
      Let’s go one step further - unless you have some experience understanding academic journals (many here do), even the journals may suggest something that isn’t true by way of you reading into it. Some things everyone here should know (like correlations and what they mean). Other things many won’t know, like certain statistics methods, base rates for a given thing, etc. It’s easy for two people to read the same study and come out with totally different conclusions based on their individual context.
    - koboll 8y ago
      >Take "facts" with a grain of salt, at the very least until you have read the article. Do not take headlines as facts. One consequence of social media is that headlines now have outsized importance in relation to the substance of actual articles. Headlines, thanks to links and embedded snippets, get dispersed to 1000x more eyeballs than the articles themselves. So when they contradict or misleadingly oversimplify (which is often, especially since they are often written to draw clicks) it's the headline version of the truth that wins out, while the actual readers are left to chime in with "Hey, that's not what the article actually said..." in the comments below, swept away in a sea of respondents who commented only after reading the headline. Journalistic ethics haven't caught up to this. They're still playing by the rules of legacy media, where you can bend the truth a little to make a joke or to draw people in. The problem is that in the digital age, that unfortunately amounts to spreading a falsehood that usually overshadows the facts of the piece.
gls2ro 8y ago
The study you are citing is about "interventions aimed at reducing sitting time at work". It is not about the health benefits of reducing time spent sitting, but about the effectiveness of methods of reducing the time spent sitting. So the study you cited does not support your assertion about health effects: > There is a Cochrane review [1] on the health effects of reducing time spent sitting at work If you have studies that does invalidate the health benefits of standing vs sitting I am very interested to read them and find out more.
- semi-extrinsic 8y ago
  I concur the review linked is more about effectiveness of intervention methods. My point was that it's been so thoroughly highlighted that we need bigger sample sizes, and yet here we are with N=15. As for "invalidating the health benefits of standing vs sitting", here is a well-designed study [1] (N=7300 and 12-year span) that found """ Occupations involving predominantly standing were associated with an approximately 2-fold risk of heart disease compared with occupations involving predominantly sitting. This association was robust to adjustment for other health, sociodemographic, and work variables. """ I.e. I think we can reasonably claim that the jury is still out on the health benefits of standing desks etc. OTOH, that increased exercise is a good thing is undisputed. [1] https://academic.oup.com/aje/article/187/1/27/4081581 https://academic.oup.com/aje/article/187/1/27/4081581
  - Bartweiss 8y ago
    Thanks, that's a really interesting result - especially in the face of countless headlines (with weak supporting data) about the dangers of sitting. I'm curious to see what "adjustment for work variables" means, since that's such an open-ended task. I'm also fascinated to see: "Cardiovascular risk associated with occupations that involve combinations of sitting, standing, and walking differed for men and women, with these occupations associated with lower cardiovascular risk estimates among men but elevated risk estimates among women." I would have guessed that mixed position, moderately-active work was drastically better than anything else across all groups. I'd be surprised if the paradoxical gender result holds, normally those imply a weak or confounded effect, but even if it's simply neutral that means my expectations were massively wrong.
picsao 8y ago
If you have 20 small studies you can bundle the data to create one macro study
pasta 8y ago
"performing studies with fifteen people" The study only shows that blood flow to your brain drops when you are sitting for a long period of time. So I think 15 people is a huge sample in this case. The study doesn't even has any conclusions. So I think 15 people is fine. The next step could be gathering data from thousands of people to check weather this causes brain damage in any form. But that's not what the study was about. Only the title of the article suggests this.
- cryoshon 8y ago
  chiming in to agree here. for something along the lines of basic organ system operation like blood flow, you need only a small sample to show that there is an effect. example: a sample of 12 people would be more than enough to prove that people bleed if they are cut. you could look at a sample of 10k people just to be sure -- but you're not going to learn anything new. the experimental measurement of the variable is too high level. the point of this study is to corroborate past research and provide another avenue of future investigation. if a future study finds that the inverse is true, we now have reason to be suspicious of that study's conclusions.

anonytrary 8y ago

The last company I worked for had a "market research" report about mobile data users that claimed 75% of people do X, the rest do Y. The sample size was 4. What's worse is that the report was beautifully styled. It's amazing how much money can be wasted on moot work when you have funding.

cryoshon 8y ago
my experience is that marketing people care nothing for quality or veracity of "facts". if it looks good, it goes.

dalbasal 8y ago

Idk the exact answer, but I doubt the reason is "researchers don't understand the statistics." It is some "how the sausage is made" reason, most likely. For example: researchers have the budget/resources to to a preliminary study. They use the interesting results of this study to secure funding for a larger sample size study. Btw, a small sample size study isn't useless necessarily. You can refine methodology, with less on the line. You can also (more or less) validly exclude certain conclusions, to avoid wasting resources on a wider study. The problem is with pop science headlines, like it is in any other journalistic field. I think it's less of a problem though, than many think. Publishing in the popular press needs to be understood as "academics working on X" not "X has been proven."

koboll 8y ago
>The problem is with pop science headlines, like it is in any other journalistic field. I think it's less of a problem though, than many think. Publishing in the popular press needs to be understood as "academics working on X" not "X has been proven." Writing a headline that misrepresents a single study's conclusion as "proven" or what "scientists say" -- especially when the conclusion is contradicted by most others on the subject -- should be considered by journalistic ethics watchdogs as an extreme violation bordering on an outright lie. The media seems simultaneously bewildered as to why the public no longer trusts it, while incapable of recognizing and responding properly to obvious breaches of the public trust like inflated science headlines.
- dalbasal 8y ago
  Its not just science, it's anything. Also, it's not just media, it's everything. The only thing to do is individual. Take "facts" with a grain of salt, at the very least until you have read the article. Do not take headlines as facts. You can just read academic journal abstracts, if you want a more academic and precise reading materials. There are trade offs between rigour and the digestible content intended to be browsed over coffee.
  - jcutrell 8y ago
    Let’s go one step further - unless you have some experience understanding academic journals (many here do), even the journals may suggest something that isn’t true by way of you reading into it. Some things everyone here should know (like correlations and what they mean). Other things many won’t know, like certain statistics methods, base rates for a given thing, etc. It’s easy for two people to read the same study and come out with totally different conclusions based on their individual context.
  - koboll 8y ago
    >Take "facts" with a grain of salt, at the very least until you have read the article. Do not take headlines as facts. One consequence of social media is that headlines now have outsized importance in relation to the substance of actual articles. Headlines, thanks to links and embedded snippets, get dispersed to 1000x more eyeballs than the articles themselves. So when they contradict or misleadingly oversimplify (which is often, especially since they are often written to draw clicks) it's the headline version of the truth that wins out, while the actual readers are left to chime in with "Hey, that's not what the article actually said..." in the comments below, swept away in a sea of respondents who commented only after reading the headline. Journalistic ethics haven't caught up to this. They're still playing by the rules of legacy media, where you can bend the truth a little to make a joke or to draw people in. The problem is that in the digital age, that unfortunately amounts to spreading a falsehood that usually overshadows the facts of the piece.

gls2ro 8y ago

The study you are citing is about "interventions aimed at reducing sitting time at work". It is not about the health benefits of reducing time spent sitting, but about the effectiveness of methods of reducing the time spent sitting. So the study you cited does not support your assertion about health effects: > There is a Cochrane review [1] on the health effects of reducing time spent sitting at work If you have studies that does invalidate the health benefits of standing vs sitting I am very interested to read them and find out more.

semi-extrinsic 8y ago
I concur the review linked is more about effectiveness of intervention methods. My point was that it's been so thoroughly highlighted that we need bigger sample sizes, and yet here we are with N=15. As for "invalidating the health benefits of standing vs sitting", here is a well-designed study [1] (N=7300 and 12-year span) that found """ Occupations involving predominantly standing were associated with an approximately 2-fold risk of heart disease compared with occupations involving predominantly sitting. This association was robust to adjustment for other health, sociodemographic, and work variables. """ I.e. I think we can reasonably claim that the jury is still out on the health benefits of standing desks etc. OTOH, that increased exercise is a good thing is undisputed. [1] https://academic.oup.com/aje/article/187/1/27/4081581 https://academic.oup.com/aje/article/187/1/27/4081581
- Bartweiss 8y ago
  Thanks, that's a really interesting result - especially in the face of countless headlines (with weak supporting data) about the dangers of sitting. I'm curious to see what "adjustment for work variables" means, since that's such an open-ended task. I'm also fascinated to see: "Cardiovascular risk associated with occupations that involve combinations of sitting, standing, and walking differed for men and women, with these occupations associated with lower cardiovascular risk estimates among men but elevated risk estimates among women." I would have guessed that mixed position, moderately-active work was drastically better than anything else across all groups. I'd be surprised if the paradoxical gender result holds, normally those imply a weak or confounded effect, but even if it's simply neutral that means my expectations were massively wrong.

picsao 8y ago

If you have 20 small studies you can bundle the data to create one macro study

pasta 8y ago

"performing studies with fifteen people" The study only shows that blood flow to your brain drops when you are sitting for a long period of time. So I think 15 people is a huge sample in this case. The study doesn't even has any conclusions. So I think 15 people is fine. The next step could be gathering data from thousands of people to check weather this causes brain damage in any form. But that's not what the study was about. Only the title of the article suggests this.

cryoshon 8y ago
chiming in to agree here. for something along the lines of basic organ system operation like blood flow, you need only a small sample to show that there is an effect. example: a sample of 12 people would be more than enough to prove that people bleed if they are cut. you could look at a sample of 10k people just to be sure -- but you're not going to learn anything new. the experimental measurement of the variable is too high level. the point of this study is to corroborate past research and provide another avenue of future investigation. if a future study finds that the inverse is true, we now have reason to be suspicious of that study's conclusions.