Science's Growing Reproducibility Crisis

​It's becoming more and more well known that reproducibility, or more accurately the lack thereof, is an enormous problem in science. At the very least, scientists are starting to agree,

"More than 70% of researchers have tried and failed to reproduce another scientist's experiments, and more than half have failed to reproduce their own experiments. Those are some of the telling figures that emerged from Nature's survey of 1,576 researchers who took a brief online questionnaire on reproducibility in research."

52% said it was a significant crisis and 38% said it was a slight crisis. Only 3% said it was not a crisis at all. 

This shouldn't be a surprise at all, for example, from the 2015 paper "Estimating the Reproducibility of Psychological Science," the abstract notes,

"Reproducibility is a defining feature of science, but the extent to which it characterizes current research is unknown. We conducted replications of 100 experimental and correlational studies published in three psychology journals using high-powered designs and original materials when available... Ninety-seven percent of original studies had significant results (p < .05). Thirty-six percent of replications had significant results..."

Indeed, the lack of replication seems to be the only, consistently reproduced result these days.

And sure, the"softer" sciences such as psychology are going to be hit the hardest. (Or one's with financial incentives, like drug tests.) And the social sciences will get wiped out. Science on the whole is having a real hard time replicating.

Science is great and all, but it's by no means infallible. Scientism, as Nassim Nicholas Taleb, calls it, is just another religion. Right now, science on the whole has a major crisis. Without reproducible results, science just doesn't tell us much. It's high time scientists look at this as a crisis and work to find out how to solve it.