Andrew Wakefield published a study linking immunization injections with autism. The Lancet now says that it should never have been published because of a "fatal conflict of interest". At the time Wakefield was being paid to collect evidence to support possible compensation claims. Ten of his coauthors have issued a retraction, though Wakefield has refused. I think it was unethical for Wakefield to conceal his conflict of interest.
George Ricaurte published a study alleging that MDMA (Ecstasy) causes brain damage. It turns out that he actually used a different drug in his experiments. Mark Kleiman has the details. I agree with Kleiman---Ricaurte is guilty of outrageous misconduct.
A while ago, I wrote about Steve Milloy's junkscience.com and observed
Unsuspecting visitors might think that Milloy's site is devoted to criticizing shoddy science, but they would be wrong. If you look at what he "debunks" you will find that the real criterion for deciding what is "junk science" is not the quality of the work, but the political agenda that it might support. Studies that support a right-wing agenda are endorsed, while studies that don't are harshly criticized.
Junkscience.com only covered one of the cases above. Can you guess which one? The one about the study that would help lawyers sue a drug company, or the one about the study that supported drug laws?
Junkscience.com was all over the Wakefield story, with at least eight postings, while completely ignoring Ricaurte.
Of course, junkscience.com is hardly in a position to throw stones at Wakefield for concealing his conflict of interest, when they themselves conceal their funding by tobacco companies. (For example, here is Milloy asking Brown and Williamson for $50,000 to fund his efforts.)