If you think even the USDA hardiness map is a conspiracy theory that was 'fudged by politicians', you should see a doctor. Honestly, I can't even tell if that comment is satire or not anymore. It reads like satire.
Didn't say anything about the map.
I was talking about
@Gabler's general statement on piss poor applications of statistical methods becoming a trend.
Stereotyped weighting is one I see frequently, at least from my angle in psychology/sociology/education. A study is done, and the results don't match the hypothesis, which was based on the study official's preconceptions about the study population. Essentially they're thinking, "well everyone knows that group X does Y," and when they set out with the study they intended to prove it, but were disappointed. So in stead of admitting their hypothesis was wrong, or even arguing that there may have been a fault in their study that produced the result, they use vaguely related data from other studies or even other sources to weight the results, thus achieving numbers more like they wanted.
In education it's very common for a study to be so poorly planned and executed that it's essentially worthless. Like that one in the mid 90s that said that kids will believe anything they see online. It was conducted in class, and the teachers were required to say that the thing the students were asked to look up was true, though it actually wasn't. The source of authority purported to be measured was a bogus website, but the thing that was actually measured was the teachers' authority. People still quote that painfully out of date and horribly executed study today, but it was complete bunk.
I was actually part of that study in 6th grade. I remember how pissed I and half the rest of the class was afterward that these people were calling it science when it had such obvious flaws. If a bunch of 12 year olds can poke holes in your study with only half a thought, you're friggin doing it wrong.
As far as the map goes, so far no one is reporting more than a half a zone change, and the maps are allot less polkadotted. Overall an improvement over the 2012 version where they thought they could get it accurate to within a city block.
And anecdotally from looking at this thread it also doesn't seem to indicate a substantial warming trend that isn't explained by urban heat islands.