IPCC Pretends the Scientific Publishing Crisis Doesn’t Exist
The Intergovernmental Panel on Climate Change (IPCC) issued a press release today. It tells us the IPCC assesses “thousands of scientific papers published each year,” and that its latest report relies on “more than 6,000 references.”
That sounds impressive until one remembers that academic publishing is in the grips of a reproducibility crisis. A disturbing percentage of the research published in medicine, economics, computer science, psychology, and other fields simply doesn’t stand up. Whenever independent third parties attempt to reproduce/replicate this work – carrying out the same research in order to achieve the same findings – the success rate is dismal.
The influential 2005 paper, Why Most Published Research Findings Are False, is now very old news. Headlines declaring that ‘science is broken’ have become commonplace. In 2015, the editor-in-chief of The Lancet declared that “much of the scientific literature, perhaps half, may simply be untrue.”
So here’s the bottom line: We know that studies about promising drugs typically fail when strangers attempt to reproduce those studies. We know that flashy physics research published in Science and Nature has been wholly fraudulent. We know that half of economics papers can’t be replicated, even with assistance from their own authors. We know political bias distorts the peer-review process in psychology. (All of this is discussed in a report I wrote in 2016).
https://www.thegwpf.org/content/uploads/2016/10/PeerReview.pdf