5 ms·

So on one hand we have a scientific study, and on the other some guy on the internet with a pet theory.

dudul 11y ago
Since it's been proven that most studies are not valid since not replicable, I don't know if there is much of a difference between the two :)
- baldfat 11y ago
  Then you just threw out the baby with the bath water. This is reproducible study and what you are talking about is actually fraud that they can't be reproduced. Doesn't speak to this study or most studies like this. The numbers are there in the project. You can reproduce your results from that. You can ask people the same questions that this study did. You can take from the same pool of people for your survey. You can also take MRIs of your sample. Seems completely reproducible. It is an issue with the "So What...."
  - dudul 11y ago
    https://www.newscientist.com/article/dn7915-most-scientific-papers-are-probably-wrong/ https://www.newscientist.com/article/dn7915-most-scientific-... http://www.businessinsider.com/why-most-published-scientific-research-is-wrong-2013-10 http://www.businessinsider.com/why-most-published-scientific...
    - baldfat 11y ago
      THIS study. Not all studies. Is this a study that is valid and was the approach right. Throwing away all because of most is not a good approach.
      - dudul 11y ago
        I think you don't understand how the scientific method is supposed to work. Let's stop there.
        baldfat 11y ago
        Sorry have to say that if you think ALL science must be reproduced your missing something. Oh well just look and see if I am all wrong. Or just go with that view point.
  - jackcosgrove 11y ago
    But has it been reproduced? A study must be reproducible and reproduced, with the same results in the reproductions, to be valid.
    - baldfat 11y ago
      Reproducibility means IT can be.This isn't the claim that stem cells could be produced from normal adult cells by dipping them into acid for a 30-minute shock period.(Was a Nature Journal paper that was false in January 2014) With these you have the numbers and the research. You would have to say something was wrong in their approach and redo it. In these instances you look at the data and you make analysis of it. No need to reproduce it.
      - jackcosgrove 11y ago
        There is definitely a need to reproduce all scientific studies. No study is perfect and can be taken at its authors' word.
        baldfat 11y ago
        It is impossible to make all science reproducible. Some studies are to expensive and just not possible. Do we need 2 Hadron Colliders? Since we need to reproduce their effects? Here is some reading on irreproducible research. http://www.nature.com/news/reproducibility-1.17552 http://www.nature.com/news/reproducibility-1.17552
cblock811 11y ago
we also have snark and hostility :)
return0 11y ago
Your point? You know any studies disproving his theory?
- enraged_camel 11y ago
  That's not how this works. The burden of proof is on the person making the claim, not the other way around.
- s73v3r 11y ago
  The one who's article we're commenting on right now?
  - return0 11y ago
    The article doesn't talk about rich kids.
s73v3r 11y ago
Unfortunately, the people in charge of the social safety nets are going to be listening to only one of these. And it's probably not the science.
dang 11y ago
A comment like this breaks the HN guidelines by being both unvicil (calling names) and unsubstantive (a knee-jerk generality). Please don't take HN conversations in this direction. Not only is there nothing wrong with sharing one's experience and wondering, those things are the life of good conversation. Meanwhile, comments like this one kill the interesting conversation, and pettiness takes its place. Note that this has nothing to do with one's actual views. Look at how thoughtfully https://news.ycombinator.com/item?id=10745355 https://news.ycombinator.com/item?id=10745355 made a counterpoint. That's what we want here. This subthread is what we don't want here. We've detached it from https://news.ycombinator.com/item?id=10744567 https://news.ycombinator.com/item?id=10744567 and marked it off-topic.

dudul 11y ago

Since it's been proven that most studies are not valid since not replicable, I don't know if there is much of a difference between the two :)

baldfat 11y ago
Then you just threw out the baby with the bath water. This is reproducible study and what you are talking about is actually fraud that they can't be reproduced. Doesn't speak to this study or most studies like this. The numbers are there in the project. You can reproduce your results from that. You can ask people the same questions that this study did. You can take from the same pool of people for your survey. You can also take MRIs of your sample. Seems completely reproducible. It is an issue with the "So What...."
- dudul 11y ago
  https://www.newscientist.com/article/dn7915-most-scientific-papers-are-probably-wrong/ https://www.newscientist.com/article/dn7915-most-scientific-... http://www.businessinsider.com/why-most-published-scientific-research-is-wrong-2013-10 http://www.businessinsider.com/why-most-published-scientific...
  - baldfat 11y ago
    THIS study. Not all studies. Is this a study that is valid and was the approach right. Throwing away all because of most is not a good approach.
    - dudul 11y ago
      I think you don't understand how the scientific method is supposed to work. Let's stop there.
      - baldfat 11y ago
        Sorry have to say that if you think ALL science must be reproduced your missing something. Oh well just look and see if I am all wrong. Or just go with that view point.
- jackcosgrove 11y ago
  But has it been reproduced? A study must be reproducible and reproduced, with the same results in the reproductions, to be valid.
  - baldfat 11y ago
    Reproducibility means IT can be.This isn't the claim that stem cells could be produced from normal adult cells by dipping them into acid for a 30-minute shock period.(Was a Nature Journal paper that was false in January 2014) With these you have the numbers and the research. You would have to say something was wrong in their approach and redo it. In these instances you look at the data and you make analysis of it. No need to reproduce it.
    - jackcosgrove 11y ago
      There is definitely a need to reproduce all scientific studies. No study is perfect and can be taken at its authors' word.
      - baldfat 11y ago
        It is impossible to make all science reproducible. Some studies are to expensive and just not possible. Do we need 2 Hadron Colliders? Since we need to reproduce their effects? Here is some reading on irreproducible research. http://www.nature.com/news/reproducibility-1.17552 http://www.nature.com/news/reproducibility-1.17552

cblock811 11y ago

we also have snark and hostility :)

return0 11y ago

Your point? You know any studies disproving his theory?

enraged_camel 11y ago
That's not how this works. The burden of proof is on the person making the claim, not the other way around.
s73v3r 11y ago
The one who's article we're commenting on right now?
- return0 11y ago
  The article doesn't talk about rich kids.

s73v3r 11y ago

Unfortunately, the people in charge of the social safety nets are going to be listening to only one of these. And it's probably not the science.

dang 11y ago

A comment like this breaks the HN guidelines by being both unvicil (calling names) and unsubstantive (a knee-jerk generality). Please don't take HN conversations in this direction. Not only is there nothing wrong with sharing one's experience and wondering, those things are the life of good conversation. Meanwhile, comments like this one kill the interesting conversation, and pettiness takes its place. Note that this has nothing to do with one's actual views. Look at how thoughtfully https://news.ycombinator.com/item?id=10745355 https://news.ycombinator.com/item?id=10745355 made a counterpoint. That's what we want here. This subthread is what we don't want here. We've detached it from https://news.ycombinator.com/item?id=10744567 https://news.ycombinator.com/item?id=10744567 and marked it off-topic.