Politicians and policymakers often rely on peer-reviewed academic research to justify public policy decisions, without realizing that academic peer-review rarely, if ever, involves verifying the data and calculations.
This is because in many cases, the authors simply refuse to release such materials in the name of confidentiality, proprietary concerns, or other issues. While on the one hand, this all seems very noble and high-minded, the flip side is that science depends upon independent replication of results - and without access to data and algorithms, independent replication isn't possible. Things get really sticky, then, when public policy decisions are rendered on the basis of conclusions published in peer-reviewed journals.
Oddly enough, the "peers" reviewing the material for publication tend to frown upon any conclusions that don't reinforce their own work or their own ideas. This means that if your results are at odds with conventional orthodoxy, you will (a) likely not be published, and (b) likely be removed from grant consideration processes. I've had personal experience in this regard: back in the early 1980's, I worked with a prominent physical chemist on an especially intriguing issue - chemocommunication in elephants.
Her discussion of the flehmen response in elephants was thoroughly dismissed by experts, as they all "knew" that elephants do not exhibit such behavior. Eventually, her article made the cover of SCIENCE - one of the most respected of peer-reviewed scientific journals - (but only after she altered the discussion from "flehmen response" to "flehmen-like response"). That issue opened my eyes, as it was subsequently demonstrated that the "peers" were dead wrong. Moreover, after my observation of a behavioral correlation involving pacinian corpuscles and a chemocommunicative "gating mechanism" relating to substance placement, it is now generally accepted that chemocommunication plays a prominent role in elephant behavioral dynamics. The much-pooh-poohed flehmen response does, in fact, occur among elephants, as do a range of other chemically-mediated behaviors.
Although humans have been living and working with elephants for some 3,000 years, the vast majority of our present state of knowledge regarding the behavioral biology of these animals has been elucidated only during the past thirty years. And the "peer-reviewers" had it mostly wrong.
In terms of public policy decisions based upon peer-reviewed work, then, it's reasonable to go in with a healthy dose of skepticism. AlGore and other "man-made global warming" enthusiasts routinely claim that "the science is in". But the science is never completely in, which is one of the things that makes it such a fascinating field of endeavor. And it is a huge mistake to base public policy decisions upon the sort of pronouncements that AlGore and his ilk so passionately promulgate, day in and day out.
AlGore and company have promoted mass hysteria by pushing - among other things - Michael Mann's 1998 "hockey stick" graph (as did the United Nations IGPC) as proof of an anthopogenic basis for global warming. Mann, however, faked the data.
In 2006, the U.S. National Research Council investigated the issue and concluded that Mann’s study failed key tests of statistical validity. Mann’s conclusions were also deemed insupportable by an expert panel led by Edward Wegman, professor of statistics at George Mason University and chairman of the National Academy of Sciences Committee on Theoretical and Applied Statistics, which was convened at the request of the U.S. Congress.
It took eight years for the NRC to figure it out, primarily because Mann refused to identify the data and algorithms used to fudge his conclusion. But hey - at least it was "peer-reviewed".