We live in a sound-bite, twitter, world. I was driving and listening to the radio do it’s top of the hour summary. Verbatim, this is what was said: “15% of Americans have the norovirus due to poor catering practices.”
WTF does that even mean?
I often read, or hear, stories where statistics are trotted out as if they mean something, or are proving a point.
Since we know that 97.23 % of statistics are made up on the spot, what statistics have you heard trotted out that leave you confused?
73% of 97 women agree.
Studies made up by me, but similar to ones I have heard (since I can’t remember many).
Confounds: “vaccines cause autism” doesn’t work if the non-vaccinated kids are that way for a reason, like they live in areas with no vaccination access (pretend here that autism isn’t associated with mostly affluent people).
Confusing correlation and causation: “dry counties have fewer deaths due to auto accident” or they could just be rural.
or a common cause (& confound):“chewing tobacco associated with steroid use” yes, if they’re all baseball players.
False dichotomy: “68% of Americans don’t believe in teaching evolution” when a good portion of those polled were in a third group, people who have no problem teaching both evolution and creation/ID. You can also easily influence responses by framing the questions a certain way. E.g. “evolution occurred without God” may be objectionable for some, yet the same people would have no problem with “current species are descended from a common ancestor and species are created by selection.” That doesn’t imply that they believe God is constantly mucking around with species (ID).
Poor definition of variables: “playing violent video games causes kids to look at pictures of weapons for longer than non-violent games, therefore violent games make them more aggressive.” A famous one on this is willingness to put hot sauce in someones’ drink = aggression.
One I saw today. Admittedly, the reporting may misinterpret it, but I’ve skimmed the paper and that appears to be some of their point. But it seems your methodology better be solid if you’re going to claim that Female-named hurricanes kill more than male hurricanes because people don’t respect them. See also, the link at the bottom with questions. If the exact opposite result were found, I could come up with a cromulent-sounding but still made up explanation of why it’s still sexist.
I don’t know why this one throws you. Norovirus is the most common food borne illness in the U.S. It is contracted from improperly prepared food, i.e. “poor catering practices”, and the CDC says that 70% of cases are traceable to food service workers. Additionally, some cases are are sub-clinical, meaning that you may have the virus without showing symptoms.
What the stat means is that 15 % of Americans are walking around with this virus in their bodies because the kid behind the counter at Burger King didn’t wash his hands.