Brokenness of Scientific Method: Up to Half of Papers May Be Wrong

In summary, there are many issues with the application of the scientific method, ranging from conflicts of interest, data falsification, lack of proper incentives, abysmal lack of statistical knowledge, etc. As much as HALF of scientific papers published today might simply be wrong!
  • #1
Ackbach
Gold Member
MHB
4,155
92
Excellent read on the brokenness of the scientific method: there are many, many issues, ranging from conflicts of interest, data falsification, lack of proper incentives, abysmal lack of statistical knowledge, etc. As many as HALF of scientific papers published today might simply be wrong!

In summary: the idea of the scientific method is good. Its implementation right now is fraught with egregious errors, many and varied. Do not just trust anyone in a white lab coat.
 
Physics news on Phys.org
  • #2
Yes, there are any number of flaws in applications of Physics as well. (For example, the original data from Edwin Hubble was presented correctly but his best fit line for the redshift problem was completely guess work and should not have been presented as it was. More data was required to make any kind of argument. (And don't get me started on "cold fusion!") That's a nice article. Thanks for posting!

Now here's a thought. We assume that the scientific method can be used to investigate just about everything. The idea is that we should (if we have the technology) be able to set up the experiment such that a preschooler should be able to push a button and get the same results as anyone else. But is this true? Can the scientific method be used to verify the scientific method?

Ackbach said:
Do not just trust anyone in a white lab coat.
I'm a hard core theorist. I never believe them!

-Dan
 
  • #3
topsquark said:
Yes, there are any number of flaws in applications of Physics as well. (For example, the original data from Edwin Hubble was presented correctly but his best fit line for the redshift problem was completely guess work and should not have been presented as it was. More data was required to make any kind of argument. (And don't get me started on "cold fusion!") That's a nice article. Thanks for posting!

Very welcome!

topsquark said:
Now here's a thought. We assume that the scientific method can be used to investigate just about everything.

Well, many things, at least. I haven't heard anyone claim that you can investigate ethics scientifically. And consciousness is a standard topic that has evaded all scientific inquiries for a long time - I personally don't think science is going to make much progress there.

topsquark said:
The idea is that we should (if we have the technology) be able to set up the experiment such that a preschooler should be able to push a button and get the same results as anyone else. But is this true? Can the scientific method be used to verify the scientific method?

That's a great question! I would say, on the whole, not. All reasoning systems make assumptions, and typically those assumptions cannot themselves be verified within the reasoning system (other than, if possible, showing consistency). See Godel's work for why that is. Even so simple a logical system as first-order logic plus the axioms of arithmetic is incomplete: there are true statements within the system that cannot be proved from within the system. Godel went outside the system to show that the statement was true, and showed that the system itself could not prove it.

Surely if you add in all the complexities and uncertainties of an inductive process such as the scientific method, this problem would only get worse.

topsquark said:
I'm a hard core theorist. I never believe them!

-Dan

Couldn't help but smile when I saw that. Same here! Just ask my wife how often I'm wrong. ;)
 
  • #4
 
  • #5
This case was really a blatant fabrication of data. But the causes of the replication crisis are usually more nuanced. Personally I've had my antennas up for fabricated studies since the start of the pandemic. There have been a few studies claiming that tobacco smoking or nicotine has protective effects against Covid-19. At least one of these studies have been shown to have serious flaws, but there have been more studies published and even nicotine trials in France. What I find suspicious is that this is happening at a time when the tobacco industry is struggling, and even tried to register tobacco as an "essential item" in many countries to bypass laws on buying essential items only during the pandemic.
 

FAQ: Brokenness of Scientific Method: Up to Half of Papers May Be Wrong

What is the "Brokenness of Scientific Method"?

The "Brokenness of Scientific Method" refers to the growing concern among scientists that many published research papers may contain flawed or incorrect findings. This could be due to various factors such as bias, flawed methodologies, or the pressure to publish results that support a certain hypothesis.

How many papers are estimated to be wrong?

According to a study published in the journal PLOS Biology, it is estimated that up to 50% of published research papers may contain inaccurate or false results.

What are the potential consequences of flawed research?

Flawed research can have serious consequences, such as wasting resources and time on pursuing incorrect findings, leading to incorrect conclusions and potentially causing harm to individuals or the environment if the research is used to inform policies or practices.

What is the scientific community doing to address this issue?

The scientific community is taking steps to address the issue of flawed research. This includes promoting transparency and reproducibility in research, encouraging pre-registration of studies and data sharing, and implementing stricter standards for publication in scientific journals.

How can individuals ensure the reliability of scientific research?

Individuals can ensure the reliability of scientific research by critically evaluating the methods and results of studies, considering the source and potential biases, and seeking out replication studies. It is also important to support and advocate for a more transparent and rigorous scientific process.

Similar threads

Replies
14
Views
4K
Replies
11
Views
26K
Replies
1
Views
3K
Replies
94
Views
9K
Back
Top