What do you consider to be the most misleading statistic you have come across or can contrive?
I find many statistics wrt health very misleading. When news reports tell that certain practices increase your chances of so and so by 50%, or even 500%, what is ignored is that this increased risk is often still statistically small, or even insignificant, as compared to other risks that we take every day without giving it a second thought.
"Human DNA only differs from Chimpanzee DNA by 0.7%"
In addition, much so-called "research" in social medicine seems extremely suspect, in that (for example) one may well wonder whether sufficient controls have been developed.
and the winner is....
Parasites account for 0.01% of your body weight.
37.43% of statistics are made up on the spot.
My old favorite was one that was brought up around holidays. It went like:
"Seventy percent of all auto accidents (don't remember the exact number) occur within thirty miles of home, so be especially careful when - - - - "
This would seem to imply that if we do ninety percent of our driving within that thirty miles, this is the 'safest' place to drive? Is that what the people who made the statement wanted to convey?
No, they're saying within 30 miles is the most likely for an accident.
LOL! That's always been my favorite! The statement they are trying to make is soundly and conclusively disproved by the supporting evidence they themselves provide.
I'm also quite fond of "four out of five dentists serveyed..."; they never do make clear the screening process by which they determine what dentists they will servey.
I just bought my first electric toothbrush. While checking into this I found that one brand is most preferred by dentists, and the other is the most preferred by dental professionals.
My own creations - deliberately as absurd as I could make them:
- the GDP of the US and the RA of Pluto are correlated ... r^2 is something like 0.99
- the 'Arpian's' view that a set of 6 (or is it 8?) quasars 'near' (on the sky) an NGC spiral is, in fact, a highly statistically significant demonstration of "pi writ large on the cosmos".
I particularly like the latter - it took me only ~1 hour, yet the numbers were truly astonishing.
(if anyone is really interested, I'll dig up the PF post links)
Not really a statistic as such...but the Law of Averages is probably the most misinterpreted and misused statistical principle.
Number of US States that report "above average" high school scores : 50
Reminds me of Lake Wobegone! ". . . that's the news from Lake Wobegone, where all the women are strong, the men are good looking, and all of the children are well above average." I've only had a chance to listen to the show a few times, but this line really cracked me up.
This could be a true statement, if all the men were weak, the women were horrible looking, and the children were averaged against the men and women (who are all retarded).
Enlightening. Does the May 03 entry on Sean Carrolls blog remind you of any PF discussions?
This isnt really a statistic but more of an annoying thing people say. "It defies the laws of physics".
On all the shampoo & conditioner adverts: now with 70% less frizz, 90% less breakage, and 50% more shine!
I wasn't aware that there was a measurement of frizz that allowed it to be expressed as a percentage.
On foods- 90% fat free!!
ok, so 10% fat, doesn't sound too good now, does it? And isn't the amount of sugar thats in it being downplayed somewhat?
haha oh man now you brought up a whole line of sayings!
"Twice the flavor"
Ive always wondered if these car dealers that go "99% approved!" are really telling the truth..
Personnally, I always wanted to release a product and market it like so:
Instead of saying "50% less fat"
say the old product had "100% more fat"
You're naive buyer will assume fat free!
But only if none of the children were men or women
Of course, it is literally true that 'all the children are well above average' if the variable is 'seniority-1'!
Well, they could either A) have genetic problems or B) not be considered "men" or "women" because those usuallt refer to grown ups.
There is a direct corellation between rising global temperatures and falling numbers of pirates, ergo pirates require low temperatures.
ROFL matt thats awesome. I want to organize a march on the capitol to protest Bush's environemtal policy on behalf of all pirates around the world.
BTW 9 out of 10 doctors think 1 out of 10 doctors is an idiot.
Separate names with a comma.