I'm not a statistician by trade, but I am a data scientist, and this is absolutely correct.
However don't disregard stats all together. They represent an absolute truth, and as such are very powerful. The bias comes in with how they are framed and talked about, so you need to be able to read a statistic and ask "ok what's the source data, where are the holes in the data, and how was it collected"
You see this shit in economics reporting all the time.
"Company beat earning expectations by 1000%!!!"
Oh they were expected to make $10 but instead they made $10,000. Not a big difference when revenue is in the hundreds of millions.
Almost all stats provided are bogus when you dig into the details. They are always dressed up to suit the makers purpose.
I'm not a statistician by trade, but I am a data scientist, and this is absolutely correct.
However don't disregard stats all together. They represent an absolute truth, and as such are very powerful. The bias comes in with how they are framed and talked about, so you need to be able to read a statistic and ask "ok what's the source data, where are the holes in the data, and how was it collected"
Yep.. There are three kinds of lies:
Either people hate Mark Twain, or love fake news. Or I’m being downvoted by shills.
"The risk of bowel cancer increased by 80%"
Then later on: from 1 out of 10000 to 1.8 out of 10000
Extrapolated from a sample size of twenty.
just have to say, I love your name. gave me a sensible chuckle.
THIS! I especially like the "increased chance of dying of" BS stats. Chance of dying is 100%, wtf is an "increased chance of dying" lol....