3 ms·

Proof by anecdote

jamiequint 5y ago
Did you get confused and think "There is so much hate, arguments, misinformation, angy, and harmful messaging on FB platforms." was data?
- jimsimmons 5y ago
  If you keep up with current affairs the past few weeks have shown that FB’s alleged problems are supported by data which is disregarded by it’s leadership
  - TheOtherHobbes 5y ago
    For example: https://www.frontiersin.org/articles/10.3389/fpsyg.2019.02711/full https://www.frontiersin.org/articles/10.3389/fpsyg.2019.0271... https://bmcresnotes.biomedcentral.com/articles/10.1186/s13104-016-2186-0 https://bmcresnotes.biomedcentral.com/articles/10.1186/s1310... https://hbr.org/2017/04/a-new-more-rigorous-study-confirms-the-more-you-use-facebook-the-worse-you-feel https://hbr.org/2017/04/a-new-more-rigorous-study-confirms-t... Facebook's own research: https://www.theverge.com/2021/9/29/22701445/facebook-instagram-mental-health-research-pdfs-documents https://www.theverge.com/2021/9/29/22701445/facebook-instagr... It's nice - and of course timely - that there are some heartwarming feel good stories in this thread. But the plural of anecdote is not science. And the science is pretty damning.
    - Karrot_Kream 5y ago
      Papers 1 and 3 are about correlating _duration_ of FB usage with happiness. Look at the papers referenced here [1] for the same effect as applied to TV. These papers, sans the new controversy over the mental health of young women on Instagram, _don't_ actually point to any stronger of a conclusion than "consuming large amounts of media is correlated with bad mental health". I'm not saying that FB and this huge corporate capture of the internet is _good_, but making arguments like this doesn't actually make coherent sense. [1]: https://buddingpsychologists.org/binge-watching-mental-health/ https://buddingpsychologists.org/binge-watching-mental-healt... (I didn't read the article, but I did skim the referenced papers)
- missedthecue 5y ago
  If it agrees with my priors it's objective fact. If else, it's an anecdote
gabyar 5y ago
Not a proof. But a useful anecdote. No one is doing large scales studies of how many people meet their partner on FB, so anecdote is all we have for that. Most of the criticisms are also anecdotes, but you seem fine with those.
wly_cdgr 5y ago
It's just my experience, but I am not very special or unusual, so it would be surprising if many people haven't had similar experiences
GuB-42 5y ago
This is a case where an anecdote is proof, at least a partial proof. If the proposition is "Facebook brings at least some good to the world". A single confirmed anecdote of Facebook bringing a single good thing to the world is a proof. In the same way that all you need to prove that black sheep exist is to show a single black sheep. The proposition is "Facebook brings a lot of good to the world" is a bit trickier, we need to define "a lot", but it can still be proven by "a lot" of anecdotes. It is not a generality we are trying to prove, so we don't need to have data representative of the general case. The part about Facebook being the best for dating is stated as a personal opinion.

Proof by anecdote

Did you get confused and think "There is so much hate, arguments, misinformation, angy, and harmful messaging on FB platforms." was data?

jimsimmons 5y ago
If you keep up with current affairs the past few weeks have shown that FB’s alleged problems are supported by data which is disregarded by it’s leadership
- TheOtherHobbes 5y ago
  For example: https://www.frontiersin.org/articles/10.3389/fpsyg.2019.02711/full https://www.frontiersin.org/articles/10.3389/fpsyg.2019.0271... https://bmcresnotes.biomedcentral.com/articles/10.1186/s13104-016-2186-0 https://bmcresnotes.biomedcentral.com/articles/10.1186/s1310... https://hbr.org/2017/04/a-new-more-rigorous-study-confirms-the-more-you-use-facebook-the-worse-you-feel https://hbr.org/2017/04/a-new-more-rigorous-study-confirms-t... Facebook's own research: https://www.theverge.com/2021/9/29/22701445/facebook-instagram-mental-health-research-pdfs-documents https://www.theverge.com/2021/9/29/22701445/facebook-instagr... It's nice - and of course timely - that there are some heartwarming feel good stories in this thread. But the plural of anecdote is not science. And the science is pretty damning.
  - Karrot_Kream 5y ago
    Papers 1 and 3 are about correlating _duration_ of FB usage with happiness. Look at the papers referenced here [1] for the same effect as applied to TV. These papers, sans the new controversy over the mental health of young women on Instagram, _don't_ actually point to any stronger of a conclusion than "consuming large amounts of media is correlated with bad mental health". I'm not saying that FB and this huge corporate capture of the internet is _good_, but making arguments like this doesn't actually make coherent sense. [1]: https://buddingpsychologists.org/binge-watching-mental-health/ https://buddingpsychologists.org/binge-watching-mental-healt... (I didn't read the article, but I did skim the referenced papers)
missedthecue 5y ago
If it agrees with my priors it's objective fact. If else, it's an anecdote

gabyar 5y ago

Not a proof. But a useful anecdote. No one is doing large scales studies of how many people meet their partner on FB, so anecdote is all we have for that. Most of the criticisms are also anecdotes, but you seem fine with those.

wly_cdgr 5y ago

It's just my experience, but I am not very special or unusual, so it would be surprising if many people haven't had similar experiences

GuB-42 5y ago

This is a case where an anecdote is proof, at least a partial proof. If the proposition is "Facebook brings at least some good to the world". A single confirmed anecdote of Facebook bringing a single good thing to the world is a proof. In the same way that all you need to prove that black sheep exist is to show a single black sheep. The proposition is "Facebook brings a lot of good to the world" is a bit trickier, we need to define "a lot", but it can still be proven by "a lot" of anecdotes. It is not a generality we are trying to prove, so we don't need to have data representative of the general case. The part about Facebook being the best for dating is stated as a personal opinion.