Replication crisis?

I'd shudder to think that Psychology might be a pseudo science, but I'm not surprised that their experiments are not always replicated. Nor would I be surprised with drug company tests either. I'm sorry, but I didn't read any further than those. Those are both examples where testing is carried out in the real world, rather than in the controlled conditions of a laboratory. That means that there are uncontrollable variations in the environment that will affect the result.

William A. Wilson is a software engineer in the San Francisco Bay Area.
Software is something that can be replicated easily because it will always perform in the same way on the same data. The real world is not like a computer program.

I don't think that a lack of replication of an experiment means that "it isn't science" as the author is purporting. Something is wrong with the experimental design certainly, but that's what peer review is for, and the fact that people are repeating experiments and finding that they don't replicate shows that the process works. That is precisely how science is supposed to work, so it isn't a crisis, merely a lack of proper review.

I would see any lack of repetition as being one of two problems. Either, there is/are some variables that have not been considered in the experimental design, or there are so many variables that have not been considered, that the sample size is too small. For the first, it's go back to drawing board and redesign the experiment so that variable is now a constant. For the latter, it means a larger sample size that allows for the deviation to be statistically acceptable.
 
Evaluation of this 2016 article from Silicon Valley?


Further to this subject I heartily recommend "Science Fictions - Exposing fraud, bias, negligence and hype in science" by Stuart Ritche. Not anti-science by any means but an expose of how human bias can creep into science. A companion book to this might be "May contain lies - how stories, statistics and studies exploit our biases and what we can do about it" by Alex Edmans (even though in my opinion he includes a bias about cavemen). My view is science is under attack by fake news and false science all the time so it really needs to get its own house in order and anything that helps people pick out potential falsehoods is a good thing (ie too good to be true studies and fake findings)
 
My view is science is under attack by fake news and false science all the time so it really needs to get its own house in order and anything that helps people pick out potential falsehoods is a good thing (ie too good to be true studies and fake findings)
I agree with that. The problem is that while I fully subscribe to the 'Scientific Method' rather than to rumour, hunches and superstition, scientists themselves are human beings with human foibles. There are some bad scientists in the same way that there must surely be some good bankers, some good politicians and some good journalists. :p
 
I agree with that. The problem is that while I fully subscribe to the 'Scientific Method' rather than to rumour, hunches and superstition, scientists themselves are human beings with human foibles. There are some bad scientists in the same way that there must surely be some good bankers, some good politicians and some good journalists. :p
With regard to science the last on that list is of particular note. Often it is the way something is reported that skews it. I remember some years back now (1996) major news papers with big headlines saying "Life found on Mars" The reality was some micro fossils that the jury is still out on but veering towards not being life
 

Similar threads


Back
Top