The DI has discovered Ioannidis too!

I realize it's fundamental to being a crank, but the persecution complex of the IDers is getting really old. The latest is Bruce Chapman at Evolution News and Views, who no longer satisfied with grasping at the mantle of Galileo, is now groping for Semmelweis and Lister as well. The idea being, as usual, if science has been slow to accept the theories of people in the past, surely the same flaws must be preventing ID from being accepted. Never mind that these other scientists actually had things like data or evidence, or did rather fantastic things like reduce the death rates in maternity wards by 90%. Further the word "persecution" in this case largely consists of not being immediately believed. Long gone are the days in which persecution meant being crucified or thrown to the lions. Nowadays, persecution apparently means actually having to provide proof for what you say. Oh the humanity!

It's just the same old Galileo Gambit being recycled to include new martyrs, who if alive today would laugh just as heartily at what the DI calls science as we do.

While nothing in this essay is new to anyone who has read Thomas Kuhn, I noticed that embedded in this little tail of hyperbole and whining was a reference to Ioannidis' work! This, of course, elevated this tired rehash of creationist nonsense from the ignore pile to the proof-that-I-was-right pile. I always knew the cranks would one day find Ioannidis' work and use it for the benefit of their Galileo Gambits.

Robert Lee Hotz in the "Science Journal" column of The Wall Street Journal two weeks ago called attention to what you might call a "study of studies" that was conducted by Dr. John Ioannidis, an epidemiologist in Greece and at Tufts University (in Massachusetts). After examining 432 published research reports from science journals (peer reviewed reports, for those of you who entertain the superstition that peer review is some kind of academic prophylactic), Ioannidis wrote in the Journal of the American Medical Association that "There is an increasing concern that in modern research, false findings may be the majority or even the vast majority of published research claims."

Mr. Hotz writes that an earlier essay by Dr. Ioannidis in the journal PLoS Medicine, "Why Most Published Research Findings are False" is "the most downloaded paper the journal PLoS has ever published." Here it is, in case you are interested.

Mainstream journals have to correct errors after publication, which, of course, is just good practice and fully in the spirit of sound science. Some papers (all peer-reviewed, remember) are retracted. However, many that are shown to be flat wrong on any number of grounds simply sit out there, uncontested. Why? Might not the sloppiness have something to do with greed? The federal government is funding scientific research like never before, and, of course it is never enough. The checks on quality seem deficient, since the people who vote the funds and many who administer them are not conversant with the scientific issues.

The DI, however, is late, as the global warming and HIV/AIDS cranks found and used his research first (for my coverage of Ioannidis see this post). The fundamental misunderstanding this crank makes is that Ioannidis doesn't show that previous papers were fraudulent, he merely shows that many effects that appear in the literature aren't replicated. It's a big difference. The data were real, they were just irrelevant. It's a problem of statistical significance. If a p < 0.05 is considered significant, a false positive effect will still appear real, and significant, about 5% of the time. Take that into account, along with the file-drawer effect and the reluctance of journals to publish negative results, and inevitably, the literature gets contaminated with a large number of false-positive results. These results should not be retracted, or disavowed, because the data are actually real. There wasn't fabrication, nor necessarily sloppiness. False positives are bound to occur with the limitations of biomedical research, which is why you don't consider single papers in isolation, but instead evaluate the literature as a whole.

The redeeming feature of science is repetition. And the mere fact that Ioannidis could do this study shows that ultimately these incorrect results were not replicated, and the literature was corrected. It should also be noted that this is largely an effect in biomedical research because of problems of human studies, variability in biological effects, costs etc. It is largely irrelevant for other scientific fields which aren't (usually) limited by things like how many cases of say, ankylosing spondylitis you can find within the time limits of a study. There's a big difference between a gene-association study in which researchers try to link a single-nucleotide polymorphism to a multi-factorial human disease and the types of observations that are made in physics. Further, even if this research did apply, replication saves the day. The problem with evolution isn't that it hasn't been sufficiently studied and replicated and confirmed across multiple different species, locations and times. Evolution has been replicated and found to be consistent in every context in which it has been studied; it is the strongest kind of theory.

So nice try DI. The mixed Galileo/Ioannidis attack is truly on the leading edge of crank attacks on science, yet like all the other cranks that have attempted the link, they once again fail to understand their source material.

Update - John P.A. Ioannidis responds after I sent him links to cranks using his work.

This is a very important issue that you are raising. I was not aware of this, but it is hard to understand how some people may use my work to fuel attacks against science per se. HIV/AIDS denialism, global warming denialism, and evolution denialism/intelligent design have nothing to do with science, they are dogmas that depend on beliefs, not on empirical observation and replication/refutation thereof. Perhaps we should just take it for granted that such "currents" may try to use anything to support their views. I think that one of the strongest advantages of science is that its propositions can be tested empirically and they can be replicated, but also refuted and contradicted, and improved. Obviously, this cannot be the case with any dogma, so all my research makes absolutely no sense in the setting of dogmatic belief. Science should gain respect in the wider public, especially because of its willingness to test and refute its hypotheses, in contrast to any type of dogma. In a letter to PLoSMed following my 2005 paper (2007;4:e215), I recently clarified that "Scientific investigation is the noblest pursuit. I think we can improve the respect of the public for researchers by showing how difficult success is." Obviously this has nothing to do with dogma (religious, political, corporate, or otherwise) that really needs no hard work and by definition cannot be countered in its absurdity.

Well, he may be shocked, but I'm not. It's part of a paradoxical behavior of the crank. While on the one hand they struggle futilely for scientific recognition of nonsense, they simultaneously try to drag science down by any means necessary to try to lower it to the level of their own discourse. Someone who is actually interested in science and is not "anti-science" as the title of this essay suggests biologists are, wouldn't be interested in smearing the reputation of science and the integrity of the process. A sure sign of a crank is one who rejoices in every perceived mistake or slight against science, as they mistakenly believe it makes their nonsense appear more legitimate.