3 ms·

Socio-economic science is rife of p-hacking. The vast majority of those studies are not reproducible. Your argument would be stronger if you'd choose a differen

by chopin 6y ago

Socio-economic science is rife of p-hacking. The vast majority of those studies are not reproducible. Your argument would be stronger if you'd choose a different example. I would assume that most studies on UBI are done by strong supporters of it which introduces enough subconscious bias to make them problematic. In the medical area (testing medications) history is full of wrong studies either by negligence or just by greed. "Trust the science" is never a good argument as being to general. I'd trust most of the science in physics, chemistry and mathematics. Apart from those there are too many shenanigans.

toto444 6y ago
What do you suggest then ? We claim that every opinion is equally worthy and just stop trying to understand the social world ?
- chopin 6y ago
  We could set higher bars for what is accepted as "scientifically valid" as an example. What you're saying is that wrong understanding is better than not to try. At least when political action affecting many is derived from science, standards must be higher than currently. For example the mRNA vaccines: As far as I know there is no or little research for possible long term side effects as the original research is for cancer medication which has vastly different risk trade offs. Nevertheless the vaccines are approved for use. I've read and seen in many discussions if one raises doubts that just "trust science" is the answer although there is no science. At least I expect when asked that a proponent can name a study which researches on the long term effects of vaccines or the underlying technology (eg. safety of packaging, which is deemed a trade secret and thus not public afaik). "Trust us" is the wrong answer as long as regulatory capture is a thing.
  - toto444 6y ago
    > What you're saying is that wrong understanding is better than not to try. You can be wrong but state clear hypotheses and a non logically flawed reasoning that allows you to make statements about the world. These can then be tested for their validity. What you claim is that the standard for the quality of the test are not high enough. That may be true or false however this is only one aspect of science. For instance many people misunderstand the use of maths in Economics. The point of using mathematical models is not to be able to make predictions about next year's GDP but to make sure that what follows from your clearly stated hypotheses about human behaviour is consistent.

by chopin 6y ago

toto444 6y ago

What do you suggest then ? We claim that every opinion is equally worthy and just stop trying to understand the social world ?

chopin 6y ago
We could set higher bars for what is accepted as "scientifically valid" as an example. What you're saying is that wrong understanding is better than not to try. At least when political action affecting many is derived from science, standards must be higher than currently. For example the mRNA vaccines: As far as I know there is no or little research for possible long term side effects as the original research is for cancer medication which has vastly different risk trade offs. Nevertheless the vaccines are approved for use. I've read and seen in many discussions if one raises doubts that just "trust science" is the answer although there is no science. At least I expect when asked that a proponent can name a study which researches on the long term effects of vaccines or the underlying technology (eg. safety of packaging, which is deemed a trade secret and thus not public afaik). "Trust us" is the wrong answer as long as regulatory capture is a thing.
- toto444 6y ago
  > What you're saying is that wrong understanding is better than not to try. You can be wrong but state clear hypotheses and a non logically flawed reasoning that allows you to make statements about the world. These can then be tested for their validity. What you claim is that the standard for the quality of the test are not high enough. That may be true or false however this is only one aspect of science. For instance many people misunderstand the use of maths in Economics. The point of using mathematical models is not to be able to make predictions about next year's GDP but to make sure that what follows from your clearly stated hypotheses about human behaviour is consistent.