Glenn’s article goes on to reference suggestions from Generative AI on how to mitigate the replication crisis in Science.
But above all, I liked this suggestion:
…And – and this was suggested by a commenter to an earlier blog post – not relying on scientific research for public policy purposes until it has been successfully replicated by someone else is not a bad idea. That would slow down the connection between research and public policy, but would that really be such a bad thing?
I particularly think this is apropos to Climate Science. Let’s measure model predictions against observations over the climate interval, which is defined to be a minimum of a 30 year trend, before we go over a policy cliff.
BTW observational methods in Climate Science also needs work. Ref. the UHI problem in the surface temperature record and the over-reliance on faulty proxy data (apropos. of Rand’s lawsuit).
Not relying on “the science” for public policy at all might right a lot of wrongs. What does public policy have to do, really, with “life, liberty and the purfuit of happinefs?” In any case, in modern English it would say “life, liberty and property.” Isn’t that the full definition?
If it’s fraud, then it’s already criminalized. Consider the typical definition of fraud:
intentional perversion of truth in order to induce another to part with something of value or to surrender a legal right
It wouldn’t be hard in these extreme examples to show that deliberate falsehoods were committed in order to obtain some combination of status, funding, and employment advancement. That checks the box as far as fraud crime goes.
Most cases are more wishful thinking or ideological bias combined with flawed methods like p-hacking. Not a crime.
Make the publication contain ALL the data. There’s this new thing called the internet, you can download 10,000 pages in 10-20 seconds. No need to kill trees anymore. No data, no publication.
Seriously, there are so many thousands of “journals”, each publishing thousands of papers with sometimes dozens of “authors” each, often enough spread over three continents. Who goes to jail and where? Where do you find knowledgeable yet disinterested referees? As far as replication, how many of this flood of “knowledge” will ever be read by anyone at all? All 99.99% will do is pad the C.V.’s of the authors, most of which never read the finished paper anyway. Yet one more way your tax dollars are working for you.
No, that would be a disaster.
Glenn’s article goes on to reference suggestions from Generative AI on how to mitigate the replication crisis in Science.
But above all, I liked this suggestion:
…And – and this was suggested by a commenter to an earlier blog post – not relying on scientific research for public policy purposes until it has been successfully replicated by someone else is not a bad idea. That would slow down the connection between research and public policy, but would that really be such a bad thing?
I particularly think this is apropos to Climate Science. Let’s measure model predictions against observations over the climate interval, which is defined to be a minimum of a 30 year trend, before we go over a policy cliff.
BTW observational methods in Climate Science also needs work. Ref. the UHI problem in the surface temperature record and the over-reliance on faulty proxy data (apropos. of Rand’s lawsuit).
Not relying on “the science” for public policy at all might right a lot of wrongs. What does public policy have to do, really, with “life, liberty and the purfuit of happinefs?” In any case, in modern English it would say “life, liberty and property.” Isn’t that the full definition?
Not relying on “the science” for public policy at all might right a lot of wrongs.
Thankfully Dr. Fauci has retired from making public policy…
Relying on “the seance” for public policy on the other hand, might wrong a lot of rights.
Quis custodiet ipsos custodes?
Trofim Lysenko
Title: Scientific Fraud
Should it be criminalized?
If it’s fraud, then it’s already criminalized. Consider the typical definition of fraud:
It wouldn’t be hard in these extreme examples to show that deliberate falsehoods were committed in order to obtain some combination of status, funding, and employment advancement. That checks the box as far as fraud crime goes.
Most cases are more wishful thinking or ideological bias combined with flawed methods like p-hacking. Not a crime.
Duplicating images and labelling them as being different?
“My data was lost in a flood/house fire/etc., so I can’t release it, only the conclusions?”
Some of it is wishful thinking, but some of it is criminal, or would be in e.g. accounting.
Make the publication contain ALL the data. There’s this new thing called the internet, you can download 10,000 pages in 10-20 seconds. No need to kill trees anymore. No data, no publication.
Seriously, there are so many thousands of “journals”, each publishing thousands of papers with sometimes dozens of “authors” each, often enough spread over three continents. Who goes to jail and where? Where do you find knowledgeable yet disinterested referees? As far as replication, how many of this flood of “knowledge” will ever be read by anyone at all? All 99.99% will do is pad the C.V.’s of the authors, most of which never read the finished paper anyway. Yet one more way your tax dollars are working for you.
Who will bring forth the lawsuits?
Some billionaire, or popular youtuber, should create an infotainment series with guest experts to replicate results.
I am sure nerdy nerds will watch it just like gun nerds watch penetration tests.
I thought gun nerds watch “Kentucky Ballistics” and “Demolition Ranch”?