Tag: Blinded by stats
-
One headline says it
Here’s the problem with statistics, especially the mean, encapsulated in one headline: People are living longer but it’s not evenly distributed. Well then, you can’t say PEOPLE. You should say rich people are living longer, and everyone else is dying faster. Since 1980 longevity and happiness were transferred upward along with wealth.
-
Watchclock 2.0
The latest Post Office podcast discusses a massive increase in mail thefts since the Trump torture camp. When most people were relying on mail and Amazon and DoorDash, organized crime grabbed the opportunity. Mail thefts, mostly for IDs and checks, increased by a factor of 2. Armed robberies of postmen increased by a factor of…
-
Shocking but didn’t seem right
Saw this pair of graphs, credited to NYTimes. Industry was dominant everywhere in 1990, now healthcare is dominant everywhere. First response was Yup, that’s exactly what happened. Then I stood back and compared with what I know from tech history and experience. On the 1990 end, industry was never dominant in most states. It was…
-
Mean vs median
A classic case… Some literary type on Substack stated as a fact that the average person reads 12 books per year. Others came in fast to give the real fact: the normal or typical person reads 1, maybe 2. The literary type was using mean, which is always pulled up by outliers like literary types…
-
What you mean EVERYONE?
This BBC piece on Sammy starts with an obnoxious headline: Everyone got duped by Sam’s big gamble What you mean Everyone, dupe man? I wasn’t duped by any of the bitcoin shit. Plenty of others with bigger voices weren’t duped, and we all tried to tell EVERYONE that EVERYONE is a goddamn fool. This has…
-
60/40 odds
Via NewSuperstitionist: Coin flipping is controlled by the flipper. = = = = = START QUOTE: Researchers theorised that when a coin is flipped, the flipper’s thumb imparts a slight wobble to it, causing it to spend more time with one side facing upwards while in the air and making it more likely to land…
-
Metaburge
Ryan Burge ran into a twitterstorm when he posted graphs showing that church attendance is a function of status. People objected that their own experience didn’t match. Burge is trying to assert the data with rationality, but it’s a hopeless fight for the SAME REASON THAT THE GRAPHS SHOW. In the last 30 years ALL…
-
There is no coffee in China???
TIL, as they say at Reddit. I’m always suspicious of ‘there are no X in Y’ sentences. Authoritative people have said There Are No Pianos In Japan, and There Are No Basements In Oklahoma. I can debunk the latter from experience, and the former is easily disproved by Yamaha with a long history of making…
-
Why did Endicott fail?
MindMatters reviews some newer research on brainwave resonance. This line of research is ‘controversial’ but seems pretty solid now thanks to MRI. The rule is: When intelligent critters are working ‘in sync’, whether in mating or hive activities, their brain waves are literally in sync. Telepathy unquestionably happens in some circumstances. Russian and American scientists…
-
Brits don’t grok the Dry Line
DailyMail bemoans the number of counties without a grocery store. Most of them are in Texas and the Dry Line parts of the plains states. Most of them are nearly unpopulated. Texas is especially peculiar because it has something like 250 LITTLE counties, the same size as counties in Ohio or Virginia. Several of the…
-
Anecdata vs Anecfood
It’s odd… In recent years we’ve returned to correct science on overprocessed food. Our digestive system needs to work for its living, and it needs to chew up individual fibers to get the most out of nutrition. If we digest food before we insert it in the mouth, the system atrophies. This was understood in…
-
Pearson refuses to correlate
Continuing to read the 1905 Literary Digest… Karl Pearson, the master of stats, has a strangely familiar complaint. “At least 50 per cent of the observations made and the data collected are worthless, and no man, however able, could deduce any result from them at all. In engineer’s language we need to scrap about 50…
-
Avi does Carver
Avi makes a nice strong Carverian point in this piece. He’s discussing astronomical stuff like gravitational waves and spectral lines. In each case the theoreticians predicted that the signal would never be there, or at best would be ferociously hard to detect among the noise. In each case the signal was perfectly obvious after the…
-
Stacks, not Savannahs
Sharp observation from Gary Smith, discussing the perils of letting stats lead you around by the nose: Our distant ancestors benefitted from noticing that elephants could lead them to water and that wildebeest stampedes might warn them of predators. The best pattern spotters were more likely to survive and reproduce, passing on their pattern-recognition skills…
-
Math random vs real random
This distinction was generally recognized in earlier decades. Now that the Platonists are fully in charge, it seems to have been lost. I recognized it in 2000 when I was designing and running perception experiments. Lately I’ve noticed a couple of older references to the point, indicating that it was common knowledge. (1) Puthoff and…
Polistra's Mill 