How to cherry-pick your statistics

Now, this recent post by Travellerev about my ‘trolling’ was presumably kicked off by a comment left by ‘Rua’ on this post:¬†Medical Journal Article: 14,000 U.S. Deaths Tied to Fukushima Reactor Disaster Fallout.

The research linked to claims that

An estimated 14,000 excess deaths in the United States are linked to the radioactive fallout from the disaster at the Fukushima nuclear reactors in Japan, according to a major new article in the December 2011 edition of the International Journal of Health Services.

Wow. That’s a lot. How did they come to this conclusion? Well, they fudged the numbers is what they did, as explained here (and here, here, here, here, to name but a few of the sites that have taken the time to look at the actual numbers).

Basically, the researchers ignored the data that didn’t prove their hypothesis, honing in instead on a subset of the data that did.¬†Interestingly enough, the researchers involved have subsequently admitted to their ‘errors’ in their data interpretation (and made more in trying to keep their hypothesis alive, but that’s another story), but Travellerev…?

Well, obviously, my comment that pointed this out has now been deleted by Travellerev, which makes you wonder, does she still believe in the conclusion of that report, if even the original authors have admitted it’s erroneous? How does deleting any opinions that point out potential flaws in the research help in our quest for ‘truth’?