Read on Twitter
Short thread: Whether it's the election or COVID, this year continues to show that a lot of people struggle with or don't care about math. It makes life harder.
A "Numbers" course (basic stats, a little accounting) should be mandatory for all high schools in this country.
A "Numbers" course (basic stats, a little accounting) should be mandatory for all high schools in this country.
Just put something together to educate people on simple concepts such as averages and sample sizes.
Look at how many stupid takes, starting with Trump, have been put out there about our COVID testing, citing total tests and ignoring the population differences among countries.
Look at how many stupid takes, starting with Trump, have been put out there about our COVID testing, citing total tests and ignoring the population differences among countries.
Another example: If you look at my Letterboxd stats ( https://letterboxd.com/KacsmarOCD/stats/) it says the 1970s and 1940s are my highest rated decades for films.
Does that mean I think films from the 40s and 70s are the best or they're my favorite decades? No, they're not
How's that possible?
Does that mean I think films from the 40s and 70s are the best or they're my favorite decades? No, they're not
How's that possible?
It's SAMPLING BIAS.
Born in 86, if I watch an old film from the 70s or 40s, I watch it because it's supposed to be a classic or highly regarded. I'm not going to watch a random 1948 film or buy a 1972 Blu-ray that's a 6/10 quality film at best.
Born in 86, if I watch an old film from the 70s or 40s, I watch it because it's supposed to be a classic or highly regarded. I'm not going to watch a random 1948 film or buy a 1972 Blu-ray that's a 6/10 quality film at best.
However, when it comes to the decades I've been alive, I will watch movies that are new and easily available to me that may not be very good. I'll buy a movie that just came out hoping it's good and sometimes will be disappointed. Or I'll watch something just for a certain actor.
So my ratings for recent decades have a lot more randomness while old decades like the 40s and 70s are a biased sample because I'm mostly just trying to watch classics.
Also, the sample sizes are uneven. I'll watch 100+ films from current years, but 25-30 from old decades.
Also, the sample sizes are uneven. I'll watch 100+ films from current years, but 25-30 from old decades.
For me to watch 100+ films from the 40s/70s, I'd have to dip my toes into some crappy and mediocre movies instead of just looking for the best of the best. That would bring down the ratings for sure.
That's just an example of sampling bias and why decades aren't apples 2 apples
That's just an example of sampling bias and why decades aren't apples 2 apples
So, it was always going to be a challenge to explain to people why our partisan voting techniques this year and the order battleground states (sometimes by GOP demand) decided to count votes would produce a Trump lead on election night that would slowly go Biden's way by Friday.
It's not fraud. It's just common sense with the different forms of votes out there and their biased distribution of Biden/Trump voters.
But we lack common sense in this country. That's why 70M+ still voted for the asshole.
/endthread
But we lack common sense in this country. That's why 70M+ still voted for the asshole.
/endthread
