It is sometimes important to understand the difference between Chekhov’s gun, and foreshadowing. Here is a short video in which a cute young lady, despite having a small piercing, knows the difference and explains it well. She also mentions red herrings. I was never expert at understanding the nuances of iconic plays and movies, but I always tried to at least improve myself. It is amazing how the brain of someone like Roger Ebert is wired to grasp movies, and sometimes, even Ebert would write in a review: “Upon second viewing, I realized that…”

OT but at least iSteve adjacent: What do you make of Nicole Shanahan? She seems to fit Steve’s profile of the Silicon Valley Adventuress (SVA), marrying the Google billionaire Sergei Brin, and then cashing out in a divorce with an undisclosed settlement. Now that Kennedy with the gravelly voice has picked her as his VP running mate.

Confusing article about a not very interesting person or book. I couldn’t follow it.

The “trick question” seems stupid.

Is the answer that it’s more “probable” that she’s just one thing instead of two? Kinda lame.

But why is it probable than Linda is a bank teller? Based on the information given, there’s more probability that she’s a feminist. So the correct answer should be: c) Linda is a feminist. (Of course, we don’t have that answer here.)

Anyway, I don’t think it has much to do with Chekhov either, it’s a test question, not a novel. If the question is misleading, it’s not proving anything except that “IQ science” is definitely overvalued.

There’s no way that Linda is not a feminist. 😛

The iconic Linda question is rubbish.

Bank teller is a completely nonsensical option. It should have been

1. Linda is a feminist
2. Linda is a feminist and escort

This is the reality.

Yeah, just as I thought. They’re wrong. Yeah, yeah, I get it. It’s less likely that a sample will meet two criteria than one and so therefore the simpler proposition is more likely.

But this ignores hierarchy. Facts are nested, they are not like geese distributed randomly on a winter field.

The primary job of the maturing intellect is to understand the rank order of facts in the world. Venn diagrams are just intersecting circles, but some circles are completely circumscribed by others, contained within or largely so. Therefore, not all elements of an argument are neutral. The trick is in knowing when what circumscribes what; what valence this element has in this particular equation.

Sorry guys. You came up short. You’ve left out wisdom, which is just this understanding of nested actual as opposed to theoretical relationships. But I can see why a couple of guys would note an error in logic or mathematical modeling and then build a theory which makes them look like the guys who draw aside the curtain. They’ll get published by the “Aha! Gotcha, morons.” crowd; those people who will embrace anything that reminds them of how smugly smarter they are than everyone else.

“Which is more probable?

1. Linda is a bank teller.

2. Linda is a bank teller and is active in the feminist movement.”

What about:

My pillow guy evicted
Trump facing legal problem
Polls showing Biden winning
Economy improving
Demographics changing
Immigration increasing
Churches going progressive
Hate speech laws getting stronger
whiteness getting criticized
criminal law reform

Dang it feels good to be a liberal!

I think the Chekov’s gun allusion demonstrates the opposite of what Steve wants here. Fiction, while perhaps amusing, is ultimately an inferior way to learn about reality compared to nonfiction. (Biographies and memoirs being the most comparable genre of nonfiction.)

Effective storytelling requires streamlining pesky complexities and dead ends. That’s fine if you only want to be entertained. But if you want to have an accurate, predictive understanding of something, you need to be aware of facts that don’t fit within a convenient narrative arc.

Kahneman, Phil Tetlock, and Nassim Taleb created a modern school of empirical skepticism that I greatly appreciate.

RIP Mister Kahneman and thirty thousand Palestinian men, women and children.

Well, I rewatched Wrath of Khan last Friday and can tell you that Checkov’s Gun did not actually go off. He pointed it at Kirk when Khan issued the command but couldn’t actually fire because Kirk’s mojo outweighed the power of the creature in his ear. Captain Tyrell, the token black in this 80s film however committed suicide rather than shoot Kirk. This was the 80s when the horror film trope that “the black guy always dies” was still in effect. However now we have so many new strange tropes on TV. I noticed this Chase Bank commercial making it look like middle aged Indian women (dot not feather) go on tropical beach vacations with a middle aged white woman, and a fat, middle aged, gay hispanic male friend. Really? When I encounter these people they always stick exclusively to their large extended clan. That would never happen, her husband and family would think she had gone nuts running off with those people. In reality she would go on that trip with her dot head husband from an arranged marriage, her kids, her parents and in laws, the whole extended clan. Has anyone met any one of these people who live like “atomized individuals” like anglos do and not stick exclusively with their own kind? JD Vance’s wife is about the only one I’ve ever heard of?

Well said.

Relatedly, every few years a researcher used to come along and rig up a study to dunk on fans and players for believing in something as stupid as the “hot hand” (or, uh, the variability of energy, focus, and coordination). “Look, the makes and misses are exactly what you’d expect from a bonomial distribution!”

Then a few years ago a paper appeared demonstrating that all of those studies suffered from subtle selection bias and once you correct for it, oopsies, the hot hand is real.

Now every few months a Twitter account with a sizeable following will post a brainteasers that tests whether you can find this selection bias yourself and all of the former gifted kids pretend it’s obvious.

It reminds me of the Monty Hall problem and the widespread objection to Marilyn Vos Savant’s solution. Even Paul Erdos told her she was wrong to switch. But now the kids are all smarter than Paul Erdos because of, uh, c-sections and this new thing called assortative mating (Elon Musk tweet reference), so we can expect future generations to make fewer of these obvious blunders.

Michael Lewis has succeeded his mentor Tom Wolfe as our top Southern center-right nonfiction author…

But has he succeeded Wolfe as “our top Southern center-right nonfiction” dresser? And might not that title go to Cape May County native, Alabama grad, “This crazy Trump… is better than Obama” Gay Talese, who’s still alive at 92? Lewis doesn’t even sport a tie, except at award presentations.

An I.Q. of 150 is quite rare: It should occur randomly only once out of every 2,330 people.

Doesn’t that depend on the test given? Isn’t a 130 on one a 150 on another? Almost, acccording to Mensa:

The minimum accepted score on the Stanford–Binet is 132, while for the Cattell it is 148, and 130 in the Wechsler tests (WAIS, WISC).

https://en.m.wikipedia.org/wiki/Mensa_International#Membership_requirement

Mensa skews old– 75% of members born before 1982. Still, this implies the membership is less “non-conforming” than the general population:

The American Mensa general membership identifies as 64 percent male, 32 percent female, 3 percent unknown, and less than 1 percent gender non-conforming or other.
https://en.m.wikipedia.org/wiki/Mensa_International#Demographics

Are trannies dumber on average, or do they just avoid the intelligent?

… center-right …

What specifically does it mean if we say someone is “center-right?”

Am I “center-right” if I believe society would be best served by having slightly more executions of unwanted adults than abortions of unwanted children?

Kinda related: Rejoice, oh Kovid Kultists, for your messiah hath cometh!

Morbidly Obese (Negress) Green Party Candidate Demands Fresh Lockdowns, Enforced By Military

https://www.zerohedge.com/news/2024-03-27/covid-propaganda-morbidly-obese-green-party-candidate-demands-fresh-lockdowns

“One of her main platform planks — along with using the state to forcibly return all United States land to Indians and slavery reparations to the tune of $4,000/month to every black funded by taxing weed profits — is the institution of 2025 COVID lockdowns, to be enforced by the National Guard and requiring the relocation of prison inmates out of the “carceral system” and into “secure housing” — a confusing proposition on many fronts.”

the rational answer is 101 Nope.

If you are going to irrationally accept uncritically all the assumptions behind the question then the rational answer is 100.

You’re reading of Fast and Slow as being “people are partly irrational!” seems generous. My reading of Fast and Slow is “if you set out to trick people, you can trick a good amount of them. The rubes!”

I was late to the Fast and Slow party, maybe 5-6 years after the hoopla, so while I slogged through each experiment trying to discern what exactly people thought was mind blowing about human nature…I couldn’t find anything. There are all kinds of old tables that basically express the same message about fool me once, shame on you; fool twice shame on me.

OT – National media doesn’t seem to be talking too much about this latest rampage attack, in Rockford Illinoise (four killed, more injured):

https://www.dailymail.co.uk/news/article-13248685/christian-ivan-soto-rockford-illinois-suspect.html

Well, it doesn’t fit the “gun violence” narrative.

One might wonder about the immigration status of the murderer.

Are trannies dumber on average,

As Steve has noticed before (but which is of course verboten for mainstream discourse), there are different types of trannies. One type is the autogynephilic male such as Dick (“Rachel”) Levine who, often late in life and after a successful career, decides to act upon his autogynephilia and assume a female public identity. Such trannies are often VERY high IQ and anything but dumb. Other types of trannies (effeminate black males) might be dumber.

The major problem with Kahneman is that he became famous before people realized that there was a replication crisis in the social sciences

His findings rely on research of others and his own research that has proved to be hard to replicate and if someone cannot replicate your psych work then the results aren’t really valid and if the results aren’t really valid…then they are fairly useless in the real world or are (at best) just-so stories that are pretty to know and tell but are just-so stories in the end

I think it’s an age thing with Mensa, honestly. Lots of autistic boys nowadays figure being a transwoman is better than being a low-status man—at least then the LGBT crowd will want to be your friend.

She’s profoundly ugly. Fetal Alcohol Syndrome optical disposition.

More generally, Lewis’s description of the breakthrough moment in the careers of the Israeli duo seems underwhelming.

Fairly sure that’s because it *is* underwhelming. I have not read his book, but the common examples are not interesting nor insightful at all. But some people have high self-esteem and good at hype.

What’s appalling here “elite” people who slobber over these supposed psych and logic 101 level insights, while our elites don’t even seem able to bring even basic logic and 6th grade math to bear on critical issues like immigration. The fools here are not the people Kahneman manages to verbally baffle, but the people who are impressed by that.

you don’t need to go much beyond the Wason 4 card test to get the idea here, although the Kahneman thing does illustrate one of Steve’s points about the 100 year friction between British scientists and jewish scientists. designing experiments to deliberately break human reasoning heuristics, principally to delve into and investigate how they work, and how they speed things up correctly in 99.9% of situations, is a long time field in psychology. the amount of nutty scenarios i had to go thru…this one guy designed a moving room experiment that breaks a human brain so much that people literally fall over onto the ground because the visual system heuristic is so wrong.

presumably robots will have none of those issues because they didn’t inherit their algorithms from a rube goldberg machine evolved over millions of years of wet brains in cases. although they will probably have their own systemic shortcomings and recurring mistakes.

one interesting new field is posing classic heuristic breaking questions to LLMs and see how much they get them wrong. what’s kinda interesting is they sometimes make the same mistakes a human would make. sometimes not. their training set is an obvious confounding factor. but it does show that maybe we’re already into universal principles of intelligence. the AIs correctly extrapolate and Democrats want to dumb them down for that, but at the same time, GPT-4 level LLMs make some of the same undergraduate STEM mistakes that professors see every year in their classes.

LOL has anybody tried the conditional hypothetical meme with AIs yet. i presume they would have no problem with it, even though they don’t eat breakfast, or any food, ever.

But then half of Mensa members should be trannies, not 1 %!

Kahneman’s reputation lost some gloss when it turned out his work suffered from its own cognitive biases. In particular, he was obsessed with the statistics of meta studies and how aggregating individally weak studies could produce strong statistical signals.

But like the trick questions Steve cited, Kahneman was basing his math on the stated results of the individual studies which he took as accepted facts to input into his logic model. But it apparently never occurred to him that the whole field was corrupted by persistent publication bias. So his genius meta analysis wasn’t cancelling out random errors of individual studies, but simply aggregating the persistent bias that ran in only one direction.

Mensa skews old– 75% of members born before 1982. Still, this implies the membership is less “non-conforming” than the general population:

The American Mensa general membership identifies as 64 percent male, 32 percent female, 3 percent unknown, and less than 1 percent gender non-conforming or other.

There’s nothing off here at all.

About 2% of men seem to be homosexuals–however they label themselves. Women’s sexuality is more fluid but if anything fewer are true lesbians who really eschew a penis attached to a healthy attractive high-status male. But real “gender non-conformity” is exceedingly rare–microscopic.

What’s going on is simply a ridiculous fad, being propagandized and promoted by the anti-national, anti-normality minoritarian wreckers on their mission to balkanize, destroy and screw up Western nations–screw up everything.

This is obvious. Biology simply could not be this bad.

Since Mensa skews old it has less of this stuff. Though probably more than a comparable age adjusted sample since the whole “Mensa” concept is pretty geeky and autistic.

in the particulars here, you have to be pretty careful when investigating the psychology of finance stuff. people start doing thing in that situation which they would never normally do. a lot of the smart finance guys can correctly answer tricky psychology test scenarios in a no consequences, calm, sterile lab environment where they can think clearly with no stakes and nothing on the line.

once the stakes start being an option trade on Nvidia stock to turn the 2 million they have in the bank into 20 million dollars and what it means for life – never having to get up at 6am again, never having to listen to their annoying boss or any boss again, never having to commute again, never having to worry about money ever again, being able to fly anywhere and do anything they want any time they want, being able to buy multiple properties, being able to approach good looking women with some confidence even if they get shot down a few times, etc

well, they start making different decisions. they might decide to risk all their money on the trade. they do some ROI calculations and decide certain stuff is worth the risk that they normally wouldn’t go for in a lower stakes scenario.

but we already know this kind of thing in general. for example women make worse decisions IRL when they are ovulating, and on paper and pencil intelligence tests, their performance goes down. nature already accepts making a worse, less optimized decision in this situation, as long as it gets the biological result it is looking for, which is much more important. so i’m not sure Kahneman has discovered much new ground here, unless he has developed some sort of pretty well calibrated scale of stakes-to-ROI ratio adjustment being made in people’s heads. rationality declines 1% per X amount of adjusted implied biological benefit to the person.

Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.

Which is more probable?

1. Linda is a bank teller.

2. Linda is a bank teller and is active in the feminist movement.

More interesting than anything Kahneman says, is what’s actually going on here.

Very few people are confused that everyone is more likely to be a bank teller, than any sub-category of bank teller.

Rather the question they are really answering is something like “which is likely a better description“. Or more mathematically: relative to the total population is Linda more likely to be “bank teller” or “feminist bank teller”. Well let’s say bank teller is 1% of female employment (1/100 chance). (Note: I’m aware it is less than that now.) But bank tellers active in the feminist movement is probably only 1% of that (1/10000). Linda is maybe half as likely as a typical woman to be a bank teller (1/200), but if she is she is at least an order of magnitude more likely to be one of the few feminist ones (1/1000) than the average woman.

What Kahneman’s b.s. actually shows, is most people in their normal routine aren’t using language like a logical computer code, but more just general passing on info. We know Linda is a certain kind of pain in the ass and that’s the salient info.

My guess–again more interesting than Kahneman–is that if we put this question on the SAT suddenly most people will get it right.

Context matters.

I further hypothesize that if you put the same question alternately in the SAT verbal section and the SAT math section, a slightly higher proportion of people will get it right in the SAT math. Because the section signals to them to treat it mathematically.

Again context matters.

And when you do get a Chekhov, he dies too young (like Pushkin).
——-
There is an Uri Geller joke here to be made, but I can’t see all of it. Is the most fundamentally Israeli question, “why did you trust me, are you stupid or something?”

Mensa was a pre-internet thing. like many pre-internet social functions, it’s mostly over now. da internets have subsumed most of that function. Grady Towers was wrong about this part.

it was highly sperg even back then. when i went to Mensa meetings 25 years ago to meet new people back when i was moving to a new city every couple years, i found most of the people going were the human robot types who enjoyed spending non-work time on sperg puzzles. the other high intelligence people not doing any Mensa meetings were extremely engaged with their jobs, and put all their wechsler 140, 150, 160 effort into their industry work, which was a lot more productive and rewarding.

probably the issue with the Math Olympiad stuff as well, and the low ROI there. most real world problems and projects are not that amenable or approachable to mathematics at that sperg level, and undergrad STEM math (at least the rigorous version) is highly useful for almost all real world new things you’re trying to figure out. so the 5 Chinese guys getting medals at MO 2024 don’t have much place to put that 1 in a million level math ability – you don’t get much application out of that. they have nothing to say about the new stealth jet fighter PRC is trying to build. the engineer guys 2 steps below them in math ability already have more and better ideas about what to do.

esoteric level math already has limited real world application as Kaczynski found out. math savants sometimes make new discoveries that move society forward, but most of the time they don’t.

Their work won Kahneman the quasi-Nobel prize in economics in 2002.

Good to see that a few years nagging in the comments can have results…

Trannies are so few in number that apart from Steve’s discovery of “Steve’s Girls,” there’s not really a lot of meaning in the bits.

Agree.

(although I doubt too many people who bought it at airport bookstores ever finished it).

Brings up a good point.

Would be great to see if you publisher could get your Noticing in Hudson’s or the like. It would be a book the airport road warriors would actually enjoy and learn from.

The mean I.Q. of the population of eighth-graders in a city is known to be 100. You have selected a random sample of 50 children for a study of educational achievement. The first child tested has an I.Q. of 150. What do you expect the mean I.Q. to be for the whole sample?

Obviously it is still 100.

We don’t know the size of the city, but there are probably as many children with IQs of 50 as 150.

The minimum accepted score on the Stanford–Binet is 132, while for the Cattell it is 148, and 130 in the Wechsler tests (WAIS, WISC).

I was offered membership of MENSA, but as Groucho Marx once said…

The detractors of IQ science seem to fixate on the idea that IQ levels are the end all, be all to someone’s capacities. Many of us who believe in IQ diversity and averages also understand that it is not the silver bullet that explains all success or failure. Rather, IQ levels can be compared to the equivalent of a measure of intellectual horsepower. It’s still just a measure of what someone could potentially achieve. Plenty of us know of brilliant folks who do not attain a higher education, that instead are content to coast along working in fairly mundane career fields. We are all human, after all with individual goals and motivations.

It is still very much up to the individual and their surrounding environment as to where those abilities and talents can take them.

it did a world-historical good job at figuring out who its brightest guys were.

Citation needed.

Isn’t a 130 on one a 150 on another?

https://www.theclassroom.com/normal-range-iq-7613839.html

The Cattell Culture Fair III intelligence test is designed to test intelligence while avoiding a bias toward Western students. The Cattell test, along with the Wechsler, is used as a qualification for Mensa. The Cattell Scale, like other IQ scales, places its average IQ score at 100. However, it uses a standard deviation of 24, rather than 15, in its scoring system, allowing it to provide a more accurate assessment of scores that are farther from 100.

These days I think it is reasonable to assume mean 100 and SD 15 unless specified otherwise.

if we put this question on the SAT suddenly most people will get it right.

Most SMART people will get this right. The way the SAT is constructed, included among almost every multiple choice answer is (at least) one answer that is called “the distractor”. The distractor serves as a lure for the stupid. It is the answer that APPEARS to be the right answer at 1st glance if you don’t stop to think about it (and most people don’t) but it isn’t. On an SAT, choice #2 above would be the distractor and you would be amazed at how many people would pick it. The leading cause of getting the wrong answer on an SAT question is falling for the distractor. Stupid people would be better off guessing randomly than trying to answer (they no longer penalize guessing) because they are so attracted to the distractor. Stupid people score LOWER than random guessing on the SAT because of the distractors.

We don’t know the size of the city, but there are probably as many children with IQs of 50 as 150.

Actually, having done a smidgeon of research I see that the frequency of an IQ of 150 is about the same standard deviation as the frequency of an IQ of 60.

However, once you start dealing with IQs of 60 and below, you are probably looking at people who were deprived of oxygen during the birth process, have brain injuries, or suffer from inherited or congenital conditions that affect the workings of the brain, so testing their IQ is difficult. Also they may not be included in statistics for school populations, or may die before they reach school age.

I read Kahneman’s best-selling book, hoping that it would help me make better decisions by illustrating my cognitive biases. However, I got almost nothing out of it.

A lot of the examples, like the feminist bank teller, are just trick questions that depend on a misunderstanding about what is being asked.

Some of the examples were wrong, such as the hot hand. See this explanation:
https://www.scientificamerican.com/article/momentum-isnt-magic-vindicating-the-hot-hand-with-the-mathematics-of-streaks/

In other news:

2024 Green Party presidential candidate calls for forced injections, military lockdowns in 2025

An animatronic Star Wars-looking bitch with purple braids is nominally running for the highest office in the land.

One of her main platform planks — along with using the state to forcibly return all United States land to Indians and slavery reparations to the tune of $4,000/month to every black funded by taxing weed profits — is the institution of 2025 COVID lockdowns, to be enforced by the National Guard and requiring the relocation of prison inmates out of the “carceral system” and into “secure housing” — a confusing proposition on many fronts.

https://www.zerohedge.com/news/2024-03-27/covid-propaganda-morbidly-obese-green-party-candidate-demands-fresh-lockdowns

The rest is worth a read.

Actually the answer is 101. You know that 1 kid has an IQ of 150 and can assume that the remaining kids still have an average IQ of 100 (picking this one kid out of a hat doesn’t raise or lower the IQ of the rest) so the weighted average is (150*1 + 100*49)/50 = 101.

Thanks. This piece has a look at the issue of publication bias in his book.
https://replicationindex.com/2020/12/30/a-meta-scientific-perspective-on-thinking-fast-and-slow/

He uses R-Index to take a look at the results presented in the book chapter by chapter. Chapter 4 on priming is significantly worse than the others.

The Replicability Index (R-Index) is based on two statistics (Schimmack, 2016). One statistic is simply the percentage of significant results. In a popular book that discusses discoveries, this value is essentially 100%. The problem with selecting significant results from a broader literature is that significance alone, p < .05, does not provide sufficient information about true versus false discoveries. It also does not tell us how replicable a result is. Information about replicability can be obtained by converting the exact p-value into an estimate of statistical power.

One thing I find surprising is the implication of the final graphic that the example involved publishing only a small portion of the studies conducted (a little over 12%). Are that many studies really binned for not producing significant results?

The conclusion.

In conclusion, Daniel Kahneman is a distinguished psychologist who has made valuable contributions to the study of human decision making. His work with Amos Tversky was recognized with a Nobel Memorial Prize in Economics (APA). It is surely interesting to read what he has to say about psychological topics that range from cognition to well-being. However, his thoughts are based on a scientific literature with shaky foundations. Like everybody else in 2011, Kahneman trusted individual studies to be robust and replicable because they presented a statistically significant result. In hindsight it is clear that this is not the case. Narrative literature reviews of individual studies reflect scientists’ intuitions (Fast Thinking, System 1) as much or more than empirical findings. Readers of “Thinking: Fast and Slow” should read the book as a subjective account by an eminent psychologists, rather than an objective summary of scientific evidence. Moreover, ten years have passed and if Kahneman wrote a second edition, it would be very different from the first one. Chapters 3 and 4 would probably just be scrubbed from the book. But that is science. It does make progress, even if progress is often painfully slow in the softer sciences.

Cattell is some kind of flakey BS test that nobody uses. Anytime you see the word “fair” you should flee in horror. “Fair” means that somebody wants to substitute THEIR judgment for the way that the markets or nature have distributed things.

In real life (as opposed to MENSA membership – that and $2.00 will buy you a Coke nowadays) most serious professionals use the Weschler tests nowadays. The Stanford-Binet has SD 16 (a slight difference anyway) but the Weschler has the overwhelming market share. But anything other than those two are not widely accepted in mainstream psychological circles.

Which is more probable?

1. People who get this right have taken the SAT.

2. People who get this right have taken the SAT and are SMART.

All in all, I think we should push the fat, feminist bank teller onto the track so the trolley doesn’t kill the other bank tellers.

The probability that the the mean of the sample and the mean of the population are exactly the same is very low. In reality, the probability that the mean of the sample would be above 100 is around 50% but having a single score that is a massive outlier just increases that chance.

As an experiment, one could create 10,000 runs of 100 random numbers using a random number generator picking from a normal distribution with a mean of 100 and an standardization of 10. Then one could look to see how many of those 10k distributions contain at least on score of 150 or higher. Then one could take the average of only those distributions and create a histogram to show what samples of 100 containing at least one score of 150 or better would be. .

This would be a good first week or two homework problems for students learning modeling.

“One might wonder about the immigration status of the murderer. ”

Status: Complicated.

Interesting how they associate Rockford with Chicago as a suburb. It’s 85 miles place to place, and seeing as Rockford grew out of its own industrial base, it wasn’t dependent upon a host city.

It’s mostly a gutted out shell of a city now, as most other rust belt cities have become, but it was never merely a residential sprawl for ex-city dwellers as defines a true suburb.

Rockford, Joliet, Aurora, Chicago Heights, and Kankakee were all stand alone cities in Illinois that somehow became suburbs of Chicago. They really weren’t, unless Daley and Obama facilitating the mass exodus of inner city negroes to those places counts.

The singing on the bus in It Happened One Night is indicative of a cultural change

It strikes one today as a somewhat strange if not a contrived scene where people get up from their seats and start singing and then all the passengers join in but supposedly it was actually quite common behavior on long distance buses way back until the fifties Thus the scene while strange to us would be actually quite understandable to the movie audience in the thirties as something quite normal Singing on a bus actually happened a lot in the old movies since this was done on actual long distance buses

Back in the thirties, people were largely monoethnic (white Europeans) and music was not as commercialized and specialized by producers plus they didn’t have personalized tech like cell phones or even transistor radios Thus they had singalongs to pass the time-every one knew the words and everyone liked what was being sung since their preferences were common due to being of one ethnic background

As America became diverse this was lost Can you imagine starting a singalong on a long distance bus today? There would be Somalis, Mexicans, Asians all having their song preferences with nothing in common and everyone staring into their phones

This commonality started to be lost in the horrible sixties In Midnight Cowboy Jon Voight, on his way to his NYC prostitution gig, boards a bus in Texas and all the way he is completely isolated by listening to his transistor radio

A world was lost and we didn’t even notice or cared

Lots of autistic boys nowadays figure being a transwoman is better than being a low-status man—at least then the LGBT crowd will want to be your friend.

Tell these autistic boys, you may end up a low status adult male, but at least unlike a trans woman, you won’t have to shell out an extra six figures to the medical industrial complex over the course of your life.

And considering our society today, to paraphrase Rich Sanchez, the term “low status” means nothing to me, I’ve seen what gets given high status.

Bought S Hawkings “A Brief History Of Time” in London in 1988. Still haven`t read it… Have Thinking fast an slow ,paperback, not read.
But I have read Thorkild Hansens Book “Prosessen Mot Hamsun” . Excellent book , but nobody of you will ever read it as it is not translated into english.
You could do like Pete Buttigieg , learn Norwegian. He did it to read one of Erlend Loe`s books in the original language . Loes “Naïve. Super” is available in english,recommended

Jack D could tell you that going by societal contribution, that’s a steep rate, but then Jack D could rebut him that it’s racist to judge minorities by societal contribution.

Well, I rewatched Wrath of Khan last Friday and can tell you that Checkov’s Gun did not actually go off.

Well, that right there, was your mistake. You should have watched ” Wrath of Kahneman.”

Between the short-lived priming fad and the CIA paragraph on magick, it looks like our dying civilization has entered the credulous mysticism phase.

The average IQ will be 101. Linda is a bank teller and is active in the feminist movement.

If she had stopped focussing on Leftist political causes, it’s unlikely that she would find herself still a bank teller at 31.

‘… Kahneman was a good guy…’

You started off by saying he worked for the IDF. Which is it?

I think not. The remaining children are no longer a random selection: removing one of the sample for a non-random reason (viz his IQ is unusually high) renders the 99 a non-random selection.

Not that any of this matters; the first thought of Captain Sensible would probably be “150? Is this really a random selection?” The good captain might then peek at the IQ of a second child. “180? Stop, stop, the game is rigged!”

Kahneman, Phil Tetlock, and Nassim Taleb created a modern school of empirical skepticism that I greatly appreciate.

Of those three I prefer Tetlock. He seems the only one focused on reality as it is (prediction) rather than over hyped exaggeration of nonetheless real effects. Kahneman and Taleb added to the conversation, but I would be hesitant to rely on them.

If you are going to irrationally accept uncritically all the assumptions behind the question then the rational answer is 100.

Could you elaborate? Here is the original question.

The mean I.Q. of the population of eighth-graders in a city is known to be 100. You have selected a random sample of 50 children for a study of educational achievement. The first child tested has an I.Q. of 150. What do you expect the mean I.Q. to be for the whole sample?

I see the reasoning behind answering 101, but not 100.

P.S. Curious because I know you to be a smart insightful person.

Fiction, while perhaps amusing, is ultimately an inferior way to learn about reality compared to nonfiction.

I’ve found the same thing. Some well-written fiction is entertaining but I find real life much more interesting.

WNB

You are making an assumption that the remaining mean is still 100. The mean of the remaining 49, given the one identified is 150, is less than 100.
The equation should look more like this (150*1+X*49)/50=100

My Shanahan take from last week.

When she VP debates Kamala, the stage will have two different mixed race Bay Area Molls Flander!

Throw in Kristi Noem and we’d have the battle of three adulteresses, with Noem’s paramour, however, being two years younger than her.

— Oliver Cromwell (@Lorlordylor) March 19, 2024

I’ve had the hot hand a few times in my life. 4 or 5 times, felt like I could not be stopped, which is not my usual experience.

And a reasonable income stream as an escort!

Actually both 100 and 101 are wrong. In a city with an infinite number of 8th graders, the answer would be 101. Because the remaining 49 children would be from a pool that still had a mean of 100, and 49*100 plus 1*150 is 5050, so the mean is 101. However, no city is infinite, and the fact that one child of 150 has been taken out, causes the mean of the remaining city to be a little less than 100, so the correct answer would be somewhere between 100 (if the 50 selected were the only kids in the city, thus guaranteeing there would be -50 somewhere in 49) to 100.99999 (if the city is large enough that taking out one 150 kid has only a very tiny impact on the mean of the remaining city.

You are supposed to think “49 with an average of 100 + 1 with an IQ of 150 => 50 with an average of 101”.

Chinese politician Ma Shitu also died today (in his time zone), at the age of 109.

https://en.wikipedia.org/wiki/Ma_Shitu

He was one of the world’s oldest living people, certainly the oldest person at the time of his death who was any sort of public figure before he became old, and certainly one of the oldest men. There are under 40 men in the world alive, older than he was (and about 250 women).

We were wondering today if high IQs are more or less self-aware. They say we are least aware of our most conspicous traits—which would be why familiarity breeds contempt (Why do you do that? Do what?). Seems that if our personalities are mostly hard-wired, and you are by definition unaware of what you do on autopilot, then your personality largely manifests unconsciously. This made me think the only really self-aware people would be actors when they are acting. We concluded that high IQs probably are less conscious of their personality because more occupied with the object of their attention but more insightful and therefore capable of modifying their own behavior/habits when they reflect on their lack of self-awareness.

But the most sefl-aware medicine is surely humility and nothing blinds like pride. Which is good to remember when your book-touring for the first time.

> this one guy designed a moving room experiment that breaks a human brain so much that people literally fall over onto the ground because the visual system heuristic is so wrong.

David N. Lee:

https://en.wikipedia.org/wiki/Sensor_fusion

> presumably robots will have none of those issues because they didn’t inherit their algorithms from a rube goldberg machine evolved over millions of years of wet brains in cases

No, they very much have to use sensor fusion as well. You might have a sensor that is slow to react, and a sensor that is noisy, and one that is fast with little noise but it tends to drift over time. You can combine them into a single “virtual sensor” that works better than any of the individual sensors. That is precisely what living organisms do, too.

> what’s kinda interesting is they sometimes make the same mistakes a human would make. sometimes not.

“They’re just questions, Leon. In answer to your query, they’re written down for me. It’s a test, designed to provoke an emotional response… Shall we continue?”

she is ok, not a raving beauty but certainly not ugly

O/T. Here’s a good article on how to escape from a car that plunges into water. Not required reading if you live in the San Fernando Valley, but yes, required reading if you live where there are roadside ponds or rivers… and especially if the household includes a teenage boy who drives.

https://www.roadandtravel.com/safetyandsecurity/sinkingcar.htm

https://www.bbc.co.uk/news/world-europe-68685292.amp

OT but World War Hair related Mister S.

I think 100 makes sense, because we know the average is 100, and you’re going to have outliers in both directions, so why would we be so sure one very high outlier moves the average?

I could also see an argument that a mirror outlier of that magnitude is unlikely, so the average of the sample could be moved by one outlier. But I don’t think that is necessary implication given the terms of the problem.

Christian nationalism running amok!

That’s a really good ratio for a knifeman.

You can often uncover a fallacy by stressing the example. What if the first 25 kids you tested were all 125? Would the expected mean of the total group of 50 still remain at 100?

Trannies are so few in number that apart from Steve’s discovery of “Steve’s Girls,” there’s not really a lot of meaning in the bits.

Assuming any “bits” are left!

However, “non-binary” and “they” do seem to be a trend among the young today, Gen א or whatever. (א being the symbol associated with infinities, toward which “genders” are approaching.). Or is this just the latest iteration of “lesbian until graduation”?

And news from Biden’s America, rather than Sailer’s:

US government orders Google to identify who watched certain YouTube videos

https://www.lifesitenews.com/news/us-government-orders-google-to-identify-who-watched-certain-youtube-videos/

Coming to a Library near you.

While Nicole Shanahan looks 100% Han Chinese her Wikipedia page says she’s only half Han. While her surname is three-fourths Han.

I’m puzzling over this one:

Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations.

Which is more probable?

1. Linda has a gun on her wall.

2. Linda has a gun on her wall and is active in the feminist movement.

Agreed. That is a good technique. I understand the reasoning you gave replying to Jonathan Mason (currently comment 45).

But I spent a fair amount of time reading dearieme’s (usually insightful) comments in James Thompson’s blog and would like to hear his response. I suspect there is an interpretation of the ground rules I (we?) am/are missing.

P.S. Perhaps not a coincidence that both of the commenters advocating an answer of 100 have a British background?

Sounds like kind of a lightweight, really, Steve. That people didn’t get the conditional probability, big deal. They probably miss all kids of little double negative SAT questions also. That’s not noteworthy. And the thinking fast and slow book was trash.

For once I agree with the resident troll.

There’s no way that with the information given, Linda is a “bank teller”. She might be a barista, a whore, or some kind of NGO worker. And she’s definitely a feminist. No children, either. But maybe a cat or two.

https://www.npr.org/2024/03/28/1237218459/census-race-categories-ethnicity-middle-east-north-africa

Next U.S. census will have new boxes for ‘Middle Eastern or North African,’ ‘Latino’
MARCH 28, 2024

On the next U.S. census and future federal government forms, the list of checkboxes for a person’s race and ethnicity is officially getting longer.

The Biden administration has approved proposals for a new response option for “Middle Eastern or North African” and a “Hispanic or Latino” box that appears under a reformatted question that asks: “What is your race and/or ethnicity?”

Going forward, participants in federal surveys will be presented with at least seven “race and/or ethnicity” categories, along with instructions that say: “Select all that apply.”

They call it maths instead of math but AFAIK it works the same.

OT but at least iSteve adjacent: What do you make of Nicole Shanahan?

She’s a chick with enough money to pay for getting RFK on the ballot. So definitely a smart move by RFK, whatever her personal politics are, or aren’t.

With financing secured, RFK will be a serious candidate. The big question is therefore whether he will tend to draw more support from Trump or Biden. Reasonable minds differ on this. But Shanahan being a semi-non white woman may tend to peel off lefties who are into that kind of thing.

One iSteve angle that is sure to emerge is whether she’s actually “white” and, if so, why do Kamala (or Obama) count as “Black.”

Here is a short video in which a cute young lady, despite having a small piercing, knows the difference and explains it well.

As the small piercing likely does not affect her knowledge, perhaps this sentence should be different.

“Here is a short video in which a cute (despite having a small piercing) young lady knows the difference and explains it well. ”

“Here is a short video in which a cute young lady knows–surprisingly, considering her small piercing–the difference and explains it well.”

Another way to look at the reasoning I reject. Whatever the IQ of the first child tested (unless it be 100), it will make you modify your expectation of the mean I.Q. for the whole sample.

In other words you’re letting the chance selection of whom to test first alter your prediction of the mean for the whole set. How can that make any sense?

Indeed, having tested the first child and done your arithmetic it hardly seems worth testing the others – you already have your answer.

Like everybody else in 2011, Kahneman trusted individual studies to be robust and replicable because they presented a statistically significant result.

Somebody (Scott Alexander?) summarized all the cognitive bias research as (to paraphrase): “A bunch of junk except for the evidence that one overwhelming bias dominates everything — confirmation bias.”

The big problem with publication bias (even more so than other p-hacks) is that it pollutes the whole body of social science research, and renders all the study results dubious and unreliable.

Actually, Chekhov was right re. pathetic fallacy of human cognition. People ascribe meaning to everything in their field of perception & experience.

While, in reality, most things or occurrences in human lives have absolutely no relevance to anything. They just are.

However, it uses a standard deviation of 24, rather than 15, in its scoring system, allowing it to provide a more accurate assessment of scores that are farther from 100.

This is like Spinal Tap changing the numbers on their amps so they could “crank it to 11.”

Hispanic or Latino cannot be white? Nor Jews?

White category doesn’t include Boers, Russians, Argentinians or Jews….

Meh …

The battle of three adulteresses …..

My pillow guy evicted Trump facing legal problem Polls showing Biden winning Economy improving Demographics changing Immigration increasing Churches going progressive Hate speech laws getting stronger whiteness getting criticized criminal law reform

Dang it feels good to be a liberal!

I don’t typically bother feeding the troll but will in this case ask, in all seriousness, no snark, what exactly is liberal (vs. conservative) about most of that list? My pillow guy is just a nut, his demise my give pleasure to a few but is not a topic of interest for most folks. Trump is not a conservative. A strong economy helps the party in power at election time regardless of party. Hate speech laws are NOT getting stronger and the progressives at SCOTUS are three. “Whiteness” as a concept is beloved of a tiny fraction of public intellectuals but not given any thought by 99% of the population. And a lot of locales are backtracking on the police reforms promulgated in the era of Floyd.

By all means, take what enjoyment you can in life. But self-deception is harmful.

It is all but certain that she is a feminist.

As to whether she is a bank teller is equally probable or improbable. There is no way to determine.

Since bank teller is paired with feminist in the second option, then that is the more probable.
________________________________________________________________

Linda was a bright, campus radical.

Which is more probable?

1. Linda is active in the feminist movement and, incidentally, a bank teller.

2. Linda is a bank teller.
________________________________________________________________

If A then B.

1. A therefore (B and C)

or

2. A therefore C
_________________________________________________________________

However if it is a known fact that all of the kids in the city have a mean average IQ of 100, then this takes into account already that there are some with IQs of 150 and some with IQs of 60.

However if there are 100,000 kids in the city and we divide them randomly into groups of 50, then some groups will have an IQ mean average of less than 100, and some more than 100. Same goes for groups of 49.

All of these problems depend on assumptions. But,

Linda’s life story makes it highly probable but not a dead certainty that she is a feminist activist.

So you draw one of those Venn Diagram thingies. You draw a circle for “the set of women with Linda’s life story who are bank tellers” and another circle for “the set of women with Linda’s life story who are feminist activists.” The sheet of paper is “the set of all women with Linda’s life story”, so the circle for “same life story who are feminist activists” may fill a large part of the paper but not the whole sheet.

The two circles may have a great deal of overlap, but there may be a small sliver of the bank teller circle not in the feminist activist circle. The intersection of the two sets is smaller than the bank teller circle, ergo, Linda being a bank teller without further qualification is the more probable statement then being a bank teller with the qualification.

The circle for “feminist activist” may even be much larger than the one for “bank teller.” So long as the bank teller is not entirely contained in the feminist activist circle, the answer to the question holds.

So how do I know this? I am not perhaps the best qualified person to teach the probability course, but I agreed to teach it, so I studied up on all of this preparing lecture notes. So goes the teaching profession.

Much as it is difficult for many on of us on iSteve to admit this, Jack D is correct on this one.

Kennedy thinks that he has appeal to independent, antivax, contrarian Men of Unz type guys who might otherwise vote for Trump.

I can’t speak for Dearieme, but my first intuition, with no conscious thinking, was 100, too. Then I thought about it briefly and came to 101. And now that I’ve really thought about it, I’m back to 100.

Here’s logic of the intuition. Under the given facts, you’d expect your sample mean to be 100 before looking at any individual data points. So why should you adjust your correct a priori expectation just because you look at one data point and it turns out not to be 100, when that’s what you would expect anyway given a typical IQ distribution? Try replacing 150 with any other number. Say, 105, so now your expected sample mean is 100.1? Or 90, so now it’s 98.8? Tying your expectation to the randomness of the first data point you look at is only going to lead you astray.

Now what if the first data point is 1000 (assuming this IQ level exists for the sake of argument). That’s when I briefly switched to the 101 camp. But then I remembered the given population mean of 100, which imposes a constraint any presence of 1000-IQ people. So maybe in this strange city, every tenth person has a 1000 IQ and the rest have 0? In that case, you wouldn’t adjust your sample mean upon finding your first data point to be 1000, either.

Jack D’s hypothetical of the first 25 kids’ having an IQ of 125 violates the given facts. Either the sampling is not random or the population mean is not 100. The hypothetical has no relevance to the question as given.

I will note that this is a different problem from the expected cumulative winnings of serial bets on a coin toss. You start out expecting 0, but if you win the first toss, it is now 1, and so on. From simulations I’ve seen, the more you play, the further you move away from 0.

Anyway, that’s my non-rigorous answer. Not fool-proof, perhaps, but it gives me confidence to see Jack D on the other side.

“ One iSteve angle that is sure to emerge is whether she’s actually “white””

She could pass for full asian so I don’t think so.

The first child tested has an I.Q. of 150. What do you expect the mean I.Q. to be for the whole sample?

I see the reasoning behind answering 101, but not 100.

IMHO, Whoever answered 100 is probably misreading the question as if it were asking “what’s the mean IQ of the rest of the sample” (i.e. the next 99 picks not counting the first 150 pick). But of course it’s actually asking for the anticipated mean of all 100 picks. As you and res noted, however, the initial 150 outlier would pull up the expected average a tad.

In fact, it seems to me, the extra 50 points of the initial outlier divided by the sample size of 100 would bump up the average by .5 points (50/100) to 100.5. So maybe you could say both answers are right depending on which way you round.

(Although if you really wanted to nerd out on the game theory aspects, I suppose you’d have to notice that the mean of the remaining sample is ever so slightly below 100 because it no longer includes the 150 outlier — but that wouldn’t be worth doing unless you had serious money riding on the outcome).

She’s profoundly ugly.

And you live in China?

Fetal Alcohol Syndrome optical disposition.

Some of us are old enough to remember when the mentally retarded were called “Mongoloid”. Due to their look.

Social and play behaviors of institutionalized mongoloid and nonmongoloid retarded children

Behavioral comparisons of mongoloid and nonmongoloid retarded persons: A review.

Mongolian idiocy

Quick, is this a sick Western child, or a healthy Han one?

Errata: I just noticed the sample was supposed to be 50, not 100. So, yes, 101 not 100.5 is correct. My bad. Hey, at least I proved my own point that it’s all about reading the question correctly, not so much the math. (I also thought I was responding to deriane so there’s that also)

What do you make of Nicole Shanahan?

It’s way too early to have a “Nicole” on a presidential ticket. Or even a “Jennifer”. What’s next? An “Ashley”? A “Brittany”? A “Taylor”?

Come to think of it, that Taylor will be 35 by January 20…

Most of you talking about the Linda question are being morons about it.

YES it’s a trick question.
BUT trick questions can still have CORRECT ANSWERS.

Almost all the subjects who were given an explanation of the trick admitted that they had MADE A MISTAKE.

It’s an interesting finding WHICH ways of wording questions CAUSE PEOPLE TO MAKE MISTAKES.

But if you say it’s not a mistake, you’re an idiot.

You know that 1 kid has an IQ of 150 and can assume that the remaining kids still have an average IQ of 100

That’s not how it’s phrased, or at least not how I read it, so in that sense the problem is ill defined or badly phrased.

If “the mean I.Q. of the population of eighth-graders in a city is known to be 100“, then consider the extreme case where there are only two eighth graders. If the selected child has an IQ of 150, the only way for the mean IQ of this “population” to be known to be 100 is if the other kid has an IQ of 50. They should have said that all the kids have an expected mean of 100, and if that were the case I’d agree with you.

Or else, instead of IQ, they’d have gone with the number of “heads” each child obtained from 200 flips of a fair coin, then your approach would work for that case, too (unless they’d have similarly said the total number of heads obtained was exactly 100*N where N is the number of kids who participated, in which case, the same principle would apply).

RIP Mister Kahneman and thirty thousand Palestinian men, women and children.

What the Jews are doing to the Palestinians of Gaza is worse than the Holocaust. Six months of terror bombing of a densely populated area the size of Philadelphia. With the full knowledge of the United States and Western Europe. Indeed, the United States is supplying the Jews with the weapons of murder.

It is the greatest televised massacre of innocent civilians in the history of the world.

… and she calls for dismantling the police force; and replacing them with citizen vigilante groups.

What could go wrong?

In SF yesterday I heard a soccer coach address a 6-year old girl as “Rudnick.”

What the Jews are doing to the Palestinians of Gaza is worse than the Holocaust.
==
I can never figure out if advocates of the Arab cause are idiots or the world’s most brazen liars.

Actually Jack, the Weschler isn’t recognized at all.

The mean I.Q. of the population of eighth-graders in a city is known to be 100. You have selected a random sample of 50 children for a study of educational achievement. The first child tested has an I.Q. of 150. What do you expect the mean I.Q. to be for the whole sample?

The trickiness of the questions arises from the Bayesianism vs frequentism conflict. Kahneman is taking the strict frequentist approach by stipulating that the sample is in fact random and the ≈ 4 sigma outlier on the first observation has no special significance.

The Bayesian would counter that we know ahead of time that the population mean IQ is 100 and so this first observation is extraordinarily unusual, far exceeding normal bounds of significance.

Kahneman can stipulate all he wants, but I can tell you that any professional gambler who witnessed a 4-sigma event on their first shot at an angle would find that outcome of the very utmost interest.

Also, Kahneman is a giant within high-finance literature. I’m not super well versed in it, but I can’t think of anyone more cited among more disparate books and articles.

Welcome to the future of politics.

It’s a trick question. There are no longer any bank tellers named Linda — or any other normal-sounding American names, for that matter.

I find it very interesting: This person, Shaelin Bishop, appears to be female but nevertheless has rather boyish vibes. (She might be non-binary based on her Twitter profile.)

Do her boyish vibes indicate that she is a lesbian? And just how old was she when she made this video?

Here is what she looks like right now:

https://www.youtube.com/@ShaelinWrites/videos

Much more boyish. Further evidence for the lesbian hypothesis? What do you think?

> Whatever the IQ of the first child tested (unless it be 100), it will make you modify your expectation of the mean I.Q. for the whole sample.

No. Professor Kahneman has promised you that the remaining 49 are (also) sampled from a population with an average IQ of 100.

It’s like those stories from first grade when we learned how to add and subtract: “if you have 3 apples and Tom gives you 4 apples, how many apples do you have?”

The correct answer is 7 apples. It’s not “but Tom doesn’t have 4 apples” or “but I ate one” or “why would Tom do that?”

It’s been interesting to see who got these questions right and who didn’t. Most people did as well or as bad as I expected but there were a few surprises: you (I am as surprised as res), HA, and Mrs. Cat.

Modern songs are not made for choral singing. Try to sing a pop song acapella some time. You will find it sucks to sing. And probably just sucks period.

Hymns and anthems and drinking songs and folk songs were for group singing.

Does anybody know Oh Susanna anymore? The Old Gray Mare? I Been Working on the Railroad?

My Darling Clementine?

Of course not. They are too good.

“Linda is a bank teller” just means that she’s a bank teller. It doesn’t say if she has red hair or likes gardening or maybe speaks French.

“Linda is a bank teller” is independent of her being a feminist.

Fetal Alcohol Syndrome typically doesn’t manifest as profound retardation. I think the running average IQ for those diagnosed is 85, give or take a few points. You will get all of tell-tale facial features like she has, though.

Okay, let’s try this.

Linda is a campus radical.

Linda is a feminist.

Linda is a bank teller.

Those are our three items.
______________________________________________

We will look at eight hundred women who are campus radicals.

Of them, 768 are feminists and 32 are not feminists.

Now let us look at the number of female campus radicals who become bank tellers.

It turns out that out of every 32 female campus radicals only one becomes a bank teller.

So, amongst the feminists there will be 24 bank tellers and amongst the non feminists, 1 (one).

But that’s the same ratio of feminists to non feminists.

So whether she becomes a bank teller or a wall-paper hanger matters not the least in determining the probability. That is not the deciding factor. It is the ratio of feminists to non-feminists that is the deciding factor. The author has disguised this fact by cleverly inverting the order in which the factors are presented.

So Peter Lund (above) is correct, but draws the wrong conclusion.

Yeah, what I’m supposed to think is pretty clear. But I’d rather think critically about it – as exemplified by sf middleroader. I’ve not had time to reflect on his argument but at least he doesn’t “game” the problem by asking himself “what am I supposed to say?”

People must surely still sing “The sexual life of the camel”?

> You could do like Pete Buttigieg , learn Norwegian. He did it to read one of Erlend Loe`s books in the original language . Loes “Naïve. Super” is available in english,recommended

Erlend Loe is fantastic! You don’t have to learn to read Norwegian, although that’s what I did. Most of his books are translated into many languages, not just English.

Hamsun? I read Sult (Hunger) and it was great. Maybe I should read some more of his stuff.

So what if you are wrong, Mr. Sailer?

The more germane question to me is “Why aren’t humans rational about noticing when they are being hoodwinked?”

Because it would appear ideologues tend to be incapable of distinguishing between what they would like to be true and what is true. They sincerely believe what whatever they say is indeed fact. Think about when a person offers a sourced position, only to be blatantly accused, without offering sources themselves or by claiming the sources used are “fake news”, of outright lying or acting in bad. At the same time, if the position reinforces that worldview, that person is prone to believe it without further investigation. And when a person automatically assumes that their ideological opponent does not hold to that standard, then they are more prone to dismiss any evidence that counters their narrative.

I can never figure out if advocates of the Arab cause are idiots or the world’s most brazen liars.

You are the most brazen liar on iSteve.

“An IQ researcher FOR THE ISRAELI MILITARY … a good guy . . . .”

You’re disgusting, Sailer.

Even at your age, when you could either retire or write honestly and speak freely, you still kiss ass and praise people who serve a regime that mass-murders children and women, intentionally repeatedly bombs hospitals and high-rise apartment buildings and schools, and is now starving people to death and celebrating it. What are you?

Well said, Corvinus, but of course that all applies to you (and me) as well.

The assumption is that it is a city with a large pool of kids and that 50 is just a random sample of the group. If there were only 50 kids in the city and you were told that the mean of the group of 50 was 100, then the mean would be 100 no matter how many kid you sampled 1st, but they said it was the mean of the whole city, not the sample group.

In a sense this is like the Monty Hall Problem, which a lot of people also don’t get. One you have additional information (the IQ of the 1 kid, what is behind Door #1) then you have to factor in the additional information.

Maybe the setup was poorly phrased for people who are not used to doing statistical problems but anyone familiar with statistics would understand that “city” means a large group and “50” means a small random sample relative to the size of the group. Anyone can count things census style where you count every single head. The power of statistics is that you can take a small sample and make conclusions about a large group.

How could picking one kid magically lower the IQ of the rest?

Let’s rephrase this problem. You have a large bin of apples (thousands of apples) and you are told that the average weight of the apples in the bin is 100 grams. 50 apples are then pulled from the bin. You weight the 1st apple and it weighs 150g. Has the weight of the other apples been reduced now or is their average weight still 100g?

If I were to learn a language to read in the original, it would be Russian.

but it gives me confidence to see Jack D on the other side.

That’s hilarious. Does it give you confidence to see Kahneman on the other side, because 101 was also Kahneman’s answer and the guy won the Nobel Prize? I’m just trying to explain it to you, but I see it’s a lost cause.

This really gets to the bottom of what Kahneman was trying to say – people don’t think rationally, they think emotionally. Whether or not you disagree with me about other things should have no influence on the answer to a math problem. This is like the Nazi insistence that there was such a thing as German Physics and Jewish Physics.

So why should you adjust your correct a priori expectation just because you look at one data point

Errm this is exactly what you are supposed to do. And you probably do but don’t realize it.

Either the sampling is not random or the population mean is not 100.

Not true. You can test this your self. You know the correct a prior of the average throw of a die is 3.5. Take a single random sample of 10 throws and you will not get exactly 3.5 and yet the average is still 3.5.

The question is about a single sample which can be random and still deviate from the average.

Of course, you’re describing confirmation bias, the elimination of which is a main target of the scientific method.

I believe one of the most worthwhile things someone can do is familiarize themselves with both the scientific method and the catalogue of prevalent biases and fallacies.

“Evidence Based Technical Analysis“, a book that cites Kahneman liberally, dedicates the first 200 pages to the most fascinating exposition of the scientific method I’ve ever read, although it ultimately concludes that every traditional chartist approach is likely to be a waste of time.

Another fascinating and highly original work dealing with fallacies and biases is the section of “Poor Charlie’s Almanac” entitled “The Psychology of Human Misjudgment”.

Isn’t there a lot less randomness in shot blocking vs shooting though?

Anyone can get hot from deep. But lots of shot blocking and defensive dominance generally is a strong indicator of a big skill disparity imo.

Brian Scalabrine of “The Scallenge” discusses it here.

I’m just trying to explain it to you, but I see it’s a lost cause.

You were not explaining anything to me. And with this comment, it appears you can’t.

I wrote four sizeable paragraphs addressing the substance of the question, one of which paragraphs specifically rebuts your red-herring example. And what do you do? You pick up on a throwaway comment in my very last sentence to accuse me of thinking “emotionally”. You don’t argue in good faith, and you project your bad faith on other people. Just look who brought up Jewishness here. You! And yet it’s the “Men of Unz” who are obsessed with Jews. This is why it’s good to have you on the other side. Even when you’re right (there’s always a possibility, maybe even here), your antics are the opposite of convincing.

Funny, it occurs to me just now, after writing the above, that you may be right. Here’s how I would go about making that argument. Suppose the sample size is two, and the first data point happens to be 150. If this is a real city of any real size, then of course the expected value of the yet unseen other data point will be 100, which makes the expected sample mean 125. You do need to adjust your expectation.

The only way for the expected sample mean to remain 100 is if the city also has a population of exactly two, which is an extreme edge case. With a population of 3, the adjusted expected sample will be 112.5. With a population of 4, then 116.7. And so on. It converges on 125 fairly quickly.

Unless someone comes along and conclusively rebuts my new reasoning, I’m happy to admit that my old reasoning is wrong.

Now, you see, Jack D, you could’ve made that argument and embarrassed me. Instead, I got to save face by correcting myself while you showed yourself to be a jerk once again. Thanks!

Now, it could be argued that what is important is not the more general case of how many female campus radicals become bank tellers, but more specifically, what is the ratio of (female campus radical feminists) vs. (female campus radical non-feminists) who become bank tellers. And this does indeed yield a different probability for each, obviously. But until this ratio outnumbers the (female campus radical feminist) vs. (female campus radical non-feminist)–and there is no reason to believe that it would ever become close–then the deciding factor is as I have said

Greater Rockford and greater Kankakee remain discrete settlements separated from greater Chicago by countryside; ditto Kenosha, Wisconsin. The line between suburb and exurb runs through McHenry, Kane, Kendall, and Will Counties in Illinois, then Lake and Porter Counties in Indiana. Real estate development extended outward and enveloped formerly discrete communities. Has happened everywhere in the country no matter who the political officeholders were. Around Chicago, there are about four suburban municipalities with a wretched public order problem – Harvey in Cook County’s southern townships and Hammond, East Chicago, and Gary in Lake County, Indiana.

Thank you for replying.

In other words you’re letting the chance selection of whom to test first alter your prediction of the mean for the whole set. How can that make any sense?

I can see how that conflicts with intuition. Another aspect of that is shown by comparing these two cases.
1. Picking, testing, and looking at the result of one person at a time. Done 50 times. (also see PS)
2. Picking 50 people, testing them all at once, then looking at the results one at a time.
(my understanding of the original question is that it is neither of those, there you pick 50 people then test and look at the results one at a time, close but not identical to case 2.)

One can argue those cases are all the same, but I find the differences affect my intuition.

For case 1. I find the 101 answer more intuitive (modulo sf middleroader’s point). I think an analogy with multiple flips of a fair coin makes that clear. If you know the first of 10 results it should affect your estimate of the total number of heads to expect. And that estimate becomes more accurate the more coins are flipped. (a version of one of Jack D’s points)

For case 2. I get a strong dose of your intuition (subtly rephrased). How can looking at a single result alter the mean of the sample? I think the reconciliation is the observation can’t alter the actual mean, but it can alter your estimate (the expected mean).

I also find it disconcerting to have one’s estimate placed at the mercy of the random selection of the first example. Nonetheless, I think that is how it works.

All of that said, I agree with you, Steve, and others that seeing a 1 in 1000 outlier in the very first instance should cause a questioning of the assumptions (in particular, population mean and sample randomness).

What do you think?

P.S. Anyone interested in sf middleroader’s point should learn about sampling with and without replacement. Note how this could affect case 1. but not case 2. BTW.
https://towardsdatascience.com/understanding-sampling-with-and-without-replacement-python-7aff8f47ebe4

“…anyone familiar with statistics would understand that ‘city’ means a large group and ’50’ means a small random sample relative to the size of the group.”

Having finished Lewis’s book a few days ago, I can state that Kahneman and Tversky claim that these irrationalities are generally consistent regardless of how familiar with statistics one is (though the experts do manage to be less irrational than the general public). Therefore, that “anyone familiar with statistics” qualifier and what that implies (i.e. how precisely is the mean “known to be 100” unless someone previously measured it as such, given how much IQ varies from place to place and even country to country?) is kind of a sticking point. Again, they could have gotten around this just by dealing with fair coins and the number of heads one gets.

Sailer notes that these inconsistencies are not really all that, and I see his point, but I think we have to remember that we’re talking about psychology (and “soft science” in general), so the bar is pretty low when it comes to having anything that is quantifiable to some extent (and moreover, isn’t “racist”), and in that sense K&T (or T&K) deliver in spades. Lewis notes that a while back, about 90% of economics papers had a behaviorist bent, and he attributes that resoundingly to them (Thaler’s Nobel winning work as well). Cass Sunstein’s “nudge” policies — featured prominently in the Obama administration — likewise owe much to their work.

There are other perhaps more practical implications of their heuristics and experimental koans. For example, telling a cancer patient that he has a 90% chance of surviving surgery is more likely to be met with “yes, in that case, let’s try that”, than when you instead tell him there’s a 10% chance he’ll die in surgery, whereupon the response is more likely to be “you’re nuts if you think I’m going to let that surgeon get anywhere near me”. How should one deal with that? So, rather than quibbling over the specifics of any one problem, I think understanding how they reshaped the landscape, so to speak, is probably a worthy endeavor. But like I said, it’s soft science, so don’t set your expectations too high.

I am curious about your take on my reply to dearieme given that you reject the coin flip analogy. I would note that the average result of the coin flip would tend towards 0.5. I think that is the proper comparison for average IQ. As opposed to net differential (which IS the proper measure for a betting game though).

Speaking of coin flips and expected values, the St. Petersburg Paradox is interesting.
https://plato.stanford.edu/entries/paradox-stpetersburg/

A fair coin is flipped until it comes up heads the first time. At that point the player wins $2^n, where n is the number of times the coin was flipped. How much should one be willing to pay for playing this game?

I would argue it fails in the real world due to two points made there.
1. No counterparty can support arbitrarily large bets.
2. The utility function for money is nonlinear (per Arrow, being bounded is the important point).

It’s an interesting finding WHICH ways of wording questions CAUSE PEOPLE TO MAKE MISTAKES.

I wonder if THAT is the reason finance types value Kahneman’s work so much. In other words, not how to anticipate (and avoid?) thinking errors. Rather, how to provoke them.

It’s not “but Tom doesn’t have 4 apples” or “but I ate one” or “why would Tom do that?”

LOL.

Or “Mary says it’s 7 and I don’t like Mary and like to be different so I will say 6.”

People think that it’s hard to construct an SAT that will separate people who can do math from people who can’t but actually it’s not that hard because people are their own worst enemies. All you need to do is find all the people who can’t think clearly and send them down a blind alley. They will fall for it every time. This is actually more effective than finding the people who know the right answer.

A math problem is like a self contained little world with its own stated assumptions. Maybe they are stupid assumptions but they are the ones that you have to work with. You have to play the game strictly on its own terms but some people can’t decenter enough to do that.

All this stuff is well and good but what they did here is give a simple 8th grade math problem involving weighted averages which has nothing to do with psychology and people can’t think clearly enough to give the one and only 1 right answer to that math problem that any 6th grader in Shanghai could answer no problem. There are many ways to overthink a math problem or underthink it or be clueless about it but there’s only one way to actually answer it properly. It’s not like deciding whether Uncle Putin is or is not a good man.

“Is the answer that it’s more ‘probable’ that she’s just one thing instead of two? Kinda lame.”

There’s no “if” about this — that’s exactly what’s going on, lame or not. This is the Conjunction Fallacy which is now commonly regarded as “the Linda problem”.

The “base rate” fallacy is probably a more useful and less “duh” Kahneman and Tversky problem, and Lewis gives it more weight. (Though this quote is from Khahneman’s book):

“An individual has been described by a neighbor as follows: ‘Steve is very shy and withdrawn, invariably helpful but with little interest in people or in the world of reality. A meek and tidy soul, he has a need for order and structure, and a passion for detail.’ Is Steve more likely to be a librarian or a farmer?”

Most people, when asked this question, will immediately answer that Steve is more likely to be a librarian. His personal qualities are those we expect of a librarian, after all. And yet this answer would be wrong, because there are more than 20 times as many male farmers in the US than librarians.

But again, I find the “And yet this answer would be wrong….” to be presumptuous. For example, let’s consider a reported rape of in Sweden and ask whether, in a case where the victim does not know the perpetrator, if the perpetrator is more likely to be a recent immigrant or a native Swede. Going out of our way to avoid the base rate fallacy and knowing that only a minority of people in Sweden are recent immigrants we should expect the rapist to be a native born Swede, right? Yes, this is an anomaly, which by its very nature means it’s less rare. Maybe the explanation to the base rate fallacy is that man-bites-dog stories loom larger in our memories than than dog-bites-man stories precisely because they’re so surprising, but how these problems are phrased go a long way towards getting people to fall for them.

Sorry, got cut off by the 5 minute mark. Yes of course, anomalies are by their very nature MORE rare and therefore loom larger — often inordinately so. Dying from a terrorist attack in a plane scares us more than dying in a car wreck, and deaths from nuclear malfunctions scare us more than the deaths from gasoline or refinery-related accidents, all of which can lead to irrational behavior. But still, for reasons I noted, the above farmer/librarian example is not the obvious fallacy that Kahneman makes it out to be.

“any 6th grader in Shanghai could answer no problem.”

Again, if they’d have just gone with flips of a fair coin, it wouldn’t have changed the statistics one iota, but there’d be no confusion whatsoever. I.e. it seems as if they’re trying to stir up trouble like some journalists use poor phrasing to generate clickbait. If they phrased things correctly and unambiguously, fewer people would get it wrong, and the fallacy wouldn’t be as prominent or publishable.

I’m more than familiar with complicated quiz questions involving Bayesian statistics (e.g., how getting a positive AIDS test doesn’t mean as much as one might fear given prevalence and what not, and you actually have to crunch the numbers out), and to the extent one is taking a statistics test and knows what the test taker is looking for — be it conditional probability, statistical independence and avoiding any “hot hands” fallacy (which is something else that Lewis’s book on Kahneman makes a big deal of) — then there’s less confusion. I’m guessing that Shanghai student has probably worked through a dozen homework problems phrased in a similar way and on the basis of that says “oh, I know what they’re trying to get me to do.” Asking someone outside that context might require more careful phrasing or at least some back-and-forth. I suspect the people programming AI bots will know what I’m talking about.

JackD says:

“This is like the Nazi insistence that there was such a thing as German Physics and Jewish Physics.”

It is?

“people don’t think rationally, they think emotionally.”

Do tell.

Well, I think I provided the clearest answer to Linda the radical feminist bank teller, but apparently, for some reason, my voice is being squelched.

Ich gebe auf.

Suppose the sample of children is 1000 not 50. Then an individual result of 150 would be much less surprising.

Do you now recalculate the expected sample mean as (1 x 150 + 999 x 100)/1000?

Or: consider two responses to the original result of 150. (i) There’s something wrong here. (ii) This must be the result of pure chance because that’s the basic assumption of all the statistics I’ve been taught.

When I was learning statistics I did notice two things. (a) The lecturers were forever making assumptions that they did not tease out or even, often, state as necessary assumptions. They had obviously been thoroughly indoctrinated with them in student days. (b) They often made assumptions that they did not test at all. Perhaps could not test.

Compared to learning, say, mechanics, it came across as a slippery business. It’s true that in mechanics you’ve got to learn that some keywords imply particular assumptions, but this was always explained in advance and – in my experience – used consistently. You didn’t have to wonder “Does this chap use different conventions than the chap who taught us last year – without notifying us?”

Of course this meant that once you’ve cracked their codes, as it were, Stats was an easy subject – money for old rope. But by God I found it frustrating until I saw how artificial the impediments to learning were.

One other observation on – well, not exactly statistical reasoning, but reasoning by a statistician. In my first academic job I was sitting one evening having coffee with other young academics. A slightly older chap joined us, a statistician. He said “Why are you all chattering about house prices? You shouldn’t buy houses, buy equities!” “Why do you say so, Eric?” asked one of us. “Because equities always go up.”

It’s as well that I’ve met intelligent statisticians since then.

I think the Chekov’s gun allusion demonstrates the opposite of what Steve wants here. Fiction, while perhaps amusing, is ultimately an inferior way to learn about reality compared to nonfiction. (Biographies and memoirs being the most comparable genre of nonfiction.)

Schopenhauer agreed = Truth.

Does it give you confidence to see Kahneman on the other side, because 101 was also Kahneman’s answer and the guy won the Nobel Prize? This really gets to the bottom of what Kahneman was trying to say – people don’t think rationally, they think emotionally.

In this case, it’s called “induction” and it works pretty well.

Amos.

People think that it’s hard to construct an SAT that will separate people who can do math from people who can’t but actually it’s not that hard because people are their own worst enemies. All you need to do is find all the people who can’t think clearly and send them down a blind alley. They will fall for it every time. This is actually more effective than finding the people who know the right answer.

Good advice, if your goal is to become a high school guidance counselor.

How could picking one kid magically lower the IQ of the rest?

It is possible, because Kahneman failed to say that the first kid was chosen randomly from the 50. Maybe the kid with the highest grades was chosen for the first IQ test. In that case, the remaining students could be assumed to be slightly below average.

This may seem like a minor quibble, but Kahneman was all about minor quibbles like this.

’30’s audiences would definitely understood the scene better than today, granted.

(e.g. today, too many people talking or walking about on a Greyhound, could either be: child molesters, terrorists, hijackers, or simply up to no good, etc.)

But the point still remains: it has nothing directly/indirectly to do with It Happened One Night. It could easily be edited out of the film and nothing really is lost. This was the point that Steve quoted about Chekov’s Gun: Chekov, following his own rule, would have had to chuck this scene out the window via editing. Nothing to do with the narrative, plot. Perhaps it develops the main characters in that they’re relaxed and caught up in “just being”, but not necessarily. Isn’t needed, coul be edited out.

It’s almost in the vein of Horsefeathers, where Groucho suddenly gets up from the couch, steps out of character, addresses the audience “I’ve GOT to stay here. But there’s no reason why you folks can’t go to lobby until this thing blows over!” (Chico was playing the piano for a dame).

And of course immediately the next scene jars them back into plot story mode.

But again, the scene isn’t necessary. It can be edited with nothing lost, nothing gained, especially if one follows Checkov’s Gun.

Also, keep in mind that the Great Depression was on in the ’30’s. Post war audiences had something that most busriders didn’t have then—the car. You do see this kind of thing in Vacation, where the Griswalds all sing to pass the time during long drive to their destination, Wally World. But it gets old fast. Perhaps that’s the thing. Do you want to be cooped and cramped up for several hrs in a car with family, or cooped and cramped up on a large bus with total strangers?

They are, except when they’re not. Like Forrest Gump said to the jogger jogging by him as stuff got on him: “It happens”

And in this case, the singing on the bus scene (the original point) in It Happened One Night could be edited out without any problems to the narrative or plot. Capra obviously had it there in his own picture for a specific reason, not just to fill up the time.

Chuck it out, its not necessary.

The 30,000 number is false. This number is from the Hamas government and can’t be trusted. In particular, the Hamas numbers started out accurate but have diverged in time from the death rate of UN female workers which is known with a high degree of certainty and from the death rate reported by Gaza hospitals. The UK Telegraph did a statistical analysis and determined (surprise, surprise!) that Hamas was overegging the cake.

This appears to further strengthen the case for using the UNRWA female staff deaths numbers, and the closely matching numbers from hospital records, as a proxy for the actual mortality numbers in Gaza. If we did this, it would suggest that around 18,000 (not 32,414) had died in Gaza since Oct 7.

https://www.telegraph.co.uk/global-health/terror-and-security/unrwa-staff-death-toll-gaza-israel-hamas-war-data/

In SF yesterday I heard a soccer coach address a 6-year old girl as “Rudnick.”

Playwright Paul Rudnick wrote film criticism even more outrageous than Steve’s for the US edition of Premiere magazine, under the pseudonym “Libby Gelmen-Waxner“. He/she called Pretty Woman “a recruiting poster for prostitution”, which I thought was right on the mark. What a waste of a Roy Orbison song.

In a problem like this, you have to work with the facts as they are given to you. You can’t invent additional assumptions . It’s possible that the 50 students were all from a school for the gifted but the statement of the problem didn’t mention this so you are not free to add imaginary conditions that were not stated.

If the first student was not randomly chosen the setup of the problem would have said so. Since nothing was stated, you have to assume that he was randomly chosen from the group of 50.

You are welcome to interpret this problem any way you want but if this is how you approach math tests you are not going to do very well on them. See Peter Lund’s comment #123.

I can see why math teachers want to tear their hair out.

I loved Hunger but I also loved Growth of the Soil, which is very different.

BTW if you liked Hunger, you might like the novels of Peter Kocan. He is a relative unknown but I think he’s great. The Treatment and the Cure, and Fresh Fields.

Suppose the sample of children is 1000 not 50. Then an individual result of 150 would be much less surprising.

Sort of. Much less surprising to have it in your full sample, but just as surprising to see it as the first example (perhaps not exactly, see Roger’s comment above).

Do you now recalculate the expected sample mean as (1 x 150 + 999 x 100)/1000?

I think so. What else can you do but ask lots of questions which are unlikely to have satisfactory answers? Or choose not to play ; )

Or: consider two responses to the original result of 150. (i) There’s something wrong here. (ii) This must be the result of pure chance because that’s the basic assumption of all the statistics I’ve been taught.

Agreed this is the real world crux. Seeing an immediate 1 in 1,000 (actually more like 1 in 2,330, but as we both know that precision is an illusion, those pesky assumptions again) outlier should make (i) higher probability.

Also agreed about much of the rest of your comment. That is not helped by so many users of statistics attempting to mislead.

P.S. Did the older chap understand the idea of financial leverage? I would say that and the respective tax dis/advantages are more the issues there. Well, those and the difference that a house offers a place to live (or rent) as well as ongoing expenses.

That’s a strong point. Staisticians: always slipping in assumptions or omitting them, in silence. They are a bad lot.

“Professor Kahneman has promised you that the remaining 49 are (also) sampled from a population with an average IQ of 100.”

Oh, if he’d have said that, the answer would have been easy. But that’s not at all what he promised and that’s what I’m knocking. He said “KNOWN to have a mean of 100”. I don’t have a problem with the “correct” answer, but why would someone awkwardly phrase the test like that given the standard phrasing that is used in statistics about expected values and so forth — precisely to make things clearer — unless they’re purposely just messing with you?

Moreover, at first glance, having rambled through all the other examples Lewis cited in his book, I though this was going to be the one about Bayesian inference — i.e., about how many times the coin comes up heads in a row before you’re supposed to realize it is highly unlikely that the coin is fair or that the flips aren’t really random. That’s why I seized on the “mean is known to be 100”, and was also thinking that maybe the expected value for the remaining kids in the class might actually need to be adjusted upward. I mean, you gotta tell me if the pebbles in the urn are being replaced after each drawing and stuff like that — it’s not too much to ask.

If you disagree, be honest and tell me what you would you say if the problem instead stated that the first THREE test scores you drew were 150 or above? How about the first TEN? Would you really blindly keep assuming that every remaining child has an expected score of 100, or would you update your Bayesian priors just a little, or else look for a sentence that implies that maybe this is really a problem about someone maybe cheating, or something like that? Fundamentally, it’s the same problem, whether it’s one score of 150, or else, three or ten, isn’t it? So why would any of those alterations make you do things differently? At what point does rare become suspiciously rare, and why isn’t THAT what the problem is maybe wanting you to figure out? And if there are any people who would really swallow ten as easily as they’d swallow one, then I’m guessing if they went to the next room and participated instead in a Milgram shock experiment, they’d likewise just keep blithely zapping, regardless of the screams they heard on the intercom, safe in the assurance that nothing about the test or the surroundings could possibly go awry.

I suspect this also helps explain why self-driving cars crash into two-seater bicycles and such. The people who program those have to make an exorbitant effort to not get taken in by everything that everyone supposedly knows but never bothers stating because the real world doesn’t care about any of those assumptions.

To summarize:

When two possibilities are coupled, it is the most probable possibility which will rule the outcome.

Now this seems to fly in the face of what we have long heard (but it doesn’t):

“Stevie boy.”

“Yes, Daddy.”

“It’s the weakest link of the chain that breaks. And then you have to deal with consequences.”

“Uh huh.”

“So if you’re driving down the highway and its 10:30 at night and nothing’s open and you’re out in the middle of the dessert and you’re riding on four bald tires that are so worn that the threads are showing through, it’s not likely that what stops you cold will be a bird feather flying through the air filter and plugging up a fuel passage in your carburetor. Got it?”

“Yes, Dad. Even I can see that………What’s a carburetor?”

“Never mind. The point is, sure as shooting, one of the bald tires will blow. So if there’s a weak spot, boy, that’s what has to be fixed first.”

“I see.”

“In society too. We have to take care of the weakest link. We’re all in this together.”

“Is that why Negroes get affirmed action? We learned in Sunday School that they are less well off than us and that they need help, a boost, a ‘leg up’ the pastor said.”

“Yes, son. That’s exactly what it means. The weakest among us must be helped out, strengthened, supported till they get back on their own two feet.”

“But Dad?”

“Yes, son”.

“The other day, at school, out on the playground, we were playing basketball and I complained about a black kid who fouled me all the time and he threw the ball at me, at my face. Then he and his two buddies surrounded me and pummeled me. They even kicked me.”

“What? Did you fight back? Give them as good as you got?”

“No Dad. There were too many of them. They ganged up on me. They were too strong for me.”

“Did you report them to the principal?”

“No.”

“No? Why not?”

“Well, as you said, Dad. They are supposed to be weaker so we must do everything we can to help them out…….but Dad?”

“Yes, son.”

“When they were beating on me, they sure didn’t seem like they were weaker to me.”

“No, I don’t suppose they would’ve.”

“So, how are they supposed to be weaker and not weaker at the same time?”

“Well……..Well, they weren’t literally weaker and not weaker at the same time, as you have observed. They’re weaker one moment and stronger another moment. Got it?”

“Sure.”

“See son, weaker and stronger are just words we use to fix or assign a value to some trait, but that thing isn’t really one or the other. I mean it sort of is and isn’t. A thing can be weaker at one time and stronger at another. The World is not static. Everything moves and changes. And words just try to pin down a value on some quality at a particular time. Does that make sense?”

“Kinda……Dad?”

“Yes, son.”

“I think I’m getting a headache. Can we go golfing?”

“Sure son. Where do you want to go?”

“Can we go to that course over in the next county? We’ve never played it before and I hear it’s got really challenging fairways and super-smooth greens. I’m tired of playing the same old neighborhood courses. I want to see how this one compares to Sandy Trap and Rough Patch. Larry told me it was designed by Jack Nicklaus! Can we go?”

“Why not? Let’s see what’s on the other side of the hill.”

You can’t invent additional assumptions.

Ah. So why can Kahneman?

SOMETIMES YOU WANT TO GO

As you may have notice, I myself had flipped, as it were, on the main question before this comment of yours was published. But here’s just a short note on the coin flip problem.

First, just so we’re on the same page, in my conception of the problem, you get 1 if you win a toss, and lose 1 if you lose, so the expected payoff of each toss is 0 (not 0.5), as is the expected cumulative winnings at the start of the game.

In simulations I’ve seen, the cumulative sum gets very far from 0 after a high number of tosses (say 10k) due to naturally occurring wining or losing streaks, which are not completely offset by streaks in the other direction.

The reason I rejected the analogy with our problem was because unlike the coin, the sample and the population are not rest after each draw, so to speak. After sampling the 150-IQ person, he’s not thrown back into the pool so that he might be sampled again. This lowers the mean of the rest of the population somewhat, which then lowers the expected mean of the rest of the sample. But of course, if the population is large enough relative to the sample size, it doesn’t matter. You might as well treat the population as being reset just like the coin, and calculate accordingly.

Actually I do quite well on math tests written by mathematicians who formulate questions precisely. The problem with Kahneman’s examples is that he is sloppy with his questions, and the reader is left to guess about what he means.

The problem does say that the 50 were chosen at random, but conspicuously avoids saying how the first student was chosen.

Someone said above: “Professor Kahneman has promised you that the remaining 49 are (also) sampled from a population with an average IQ of 100.”

The problem could have said that, if the goal were to test understanding of statistics. No, it is worded to be a trick question, not to test understanding of anything.

One can argue those cases are all the same, but I find the differences affect my intuition.

My conception of the problem was always Case 2, so maybe I was affected, too, although I shouldn’t have been. The math should be the same for both cases.

Since I’ve changed my mind, I think where I (and maybe others) went wrong is I subconsciously mistook the “correct” answer that you want from the sample for the correct answer to question that Kahneman was asking.

The first is 100 and will always be 100 no matter what any one or ten or hundred data points say. The sample is supposed to give you the population mean. That’s its job.

But of course, nothing guarantees that the sample will always do its job. If your sample size is one, and the 150-IQ kid happens to be the one, then 150 is the “correct” sample mean, which is what Kahneman was asking, even though it is a far cry from being the “correct” estimate of the population mean. There’s no point in telling the sample that it should stay strong in the face of randomness.

If your sample size is two, in a large population with a mean of 100, then your expected sample mean before looking at the remaining data point is 125. And so on. For a sample size of 50, the answer is 101.

For case 2. I get a strong dose of your intuition (subtly rephrased). How can looking at a single result alter the mean of the sample? I think the reconciliation is the observation can’t alter the actual mean, but it can alter your estimate (the expected mean).

Your phrasing sounds off here. Any data point necessarily affects the sample mean, of which it is an integral part. You might say the data point “can’t alter” the sample mean in the sense that the sample mean has no existence independent of the data point, but that’s just philosophical. Since the data point is part of the mean, it should likewise be part of the expected mean. Before you know its value, you make a reasonable guess, but once you know the value, you incorporate it into both the mean and the expected mean.

“The 30,000 number is false.”

Similar to the Israelis claiming that the Palestinians were burning and eating babies alive (Biden erroneously quoted this type of thing before it was unceremoniously walked back few days later)

“This number is from the Hamas government and can’t be trusted.”

What was THAT? WHO can’t be trusted? You mean like the Middle East nation recently convicted of genocide in the World Court?

Jack, baby, come on. When in a hole, stop digging. Israel is the aggressor, pure and simple, and the whole world knows it. Petty arguing over numbers during Israeli genocidal slaughter is besides the point.

The big picture, is that Israel is the aggressor and is in the wrong. Time to have a ceasefire, and stop starving civilian populations. Stop the barbaric slaughter already.

Oh I get it. Jack’s doing an early April Fool’s joke on everyone. But it’s still March isn’t it?

Israel isn’t the victim here. Time to do the ceasefire, like civilized adults do.

“I would argue it fails in the real world due to two points made there.
1. No counterparty can support arbitrarily large bets.”

In certain crackpot ventures, such as investment banking or Jewish war aims, no single counterparty is needed. The US taxpayer and public will be made to support the insane bet every single time.

I had started submitting this comment or a similar one, earlier, but decided against it because I thought it would be too confusing. I wanted to present my more specific argument first. So, why do I disagree with Mr. Kahneman? Or rather how?

Well, when Mr. Kahneman lays out his options he leaves unspoken the non-feminist category. He says this:

oh definitively, in Chekov’s personal artistic style and in his playwright experience the bus singalong would need to be paired down as being superfluous to the plot

Jack, baby, come on. When in a hole, stop digging. Israel is the aggressor, pure and simple, and the whole world knows it. Petty arguing over numbers during Israeli genocidal slaughter is besides the point.
==
There is no genocidal slaughter and Israel is not the aggressor. Anyone who is not a malicious twit understands this.

Even so even today a lot of people still take a bus to go long distances, after all it still is a major industry-either it’s more economical than a single driver journey or simply some people don’t have reliable car or even a car period

I would say the death of joint bus singalongs is basically another example of the Bowling Alone hypothesis-the more racially diverse a society gets the more isolated individuals in that society get

This article was from The Telegraph, not from the IDF. Is The Telegraph lying too?

Was Israel the aggressor on Oct 7? Was the US “the aggressor” at Hiroshima? You have to look at the big picture – who started this war? Sow the wind, reap the whirlwind.

Read this interview with leftist Israeli historian Benny Morris:

https://www.wsj.com/articles/history-goes-to-war-in-the-holy-land-israel-leading-historian-palestine-hamas-66384695

If you can get past the paywall. He is a total leftist and even he agrees that Israel has no choice but to get rid of Hamas. No nation can permit itself to be attacked in the way that Hamas attacked on 10/7 and survive.

Be honest and say you want Israel to disappear. But they are not going to allow themselves to be slaughtered again, not without a fight. And if they want to continue to exist then Hamas has to go.

Testing child #1 @150 does not change the population mean but it does change the known SAMPLE mean. As I said before, what if you had tested 49 kids instead of 1 and the 49 were coming in at 102 and there was only 1 kid left. Would you STILL say that the sample mean had not changed?

The sample mean is equal to the population mean only probabilistically and within with the confidence interval. It doesn’t HAVE to be exact 100. The sample mean could be 98 or 102. Kahneman is just asking you to recalculate the sample mean in light of known information.

Agree. Hopefully more Congressional chaos can confound Mike Johnson’s treasonous buckling to the forever wars. The chaos which so amused the lyingpress and so upset respectable attender of parties Hugh Hewitt is the best our government has behaved in memory.

Can you guess, vot is in my pocket? WATCH you don’t miss it!

Appeciate your sentiment, sir. But a ceasefire is not even the beginning of partial and belated justice. In fact, it will allow the aggressors, the “israeli” maniacs, to rearm and heal in time for another step in their genocide of the people they seek to exterminate or expel.

Time to disarm and disband “israel”, then start prosecuting, convicting and lawfully executing several hundred thousand “israeli” military thugs for mass murder of women, children, elderly, infirm and other noncombatants — as civilized adults do.

FREE PALESTINE

DISARM “israel”

DISBAND “israel”

Give ALL the land back to the Palestinians

Deploy US troops to destroy the “israeli” military and infrastructure if need be. It’s the least the US can do after arming, aiding, and shielding the murderers and torturers for decades.

music was not as commercialized and specialized by producers

You have this ass backward. The songs that everyone knew the words to were a PRODUCT of commercialization and technology (as was the bus itself for that matter). People all over the country knew the words to the same songs because they had heard them on the radio or seen them performed at the movies (and before that, because they had been published in NY by sheet music companies and put on phonograph records or player piano rolls). Commerce can be unifying as well as isolating.

“An IQ researcher FOR THE ISRAELI MILITARY … a good guy . . . .”

You’re disgusting, Sailer.

Even at your age, when you could either retire or write honestly and speak freely, you still kiss ass and praise people who serve a regime that mass-murders children and women, intentionally repeatedly bombs hospitals and high-rise apartment buildings and schools, and is now starving people to death and celebrating it. What are you?

Here we can begin to understand why it is so important to Sailer that the treatment of Jews during World War 2 be considered to be “the worst crime in human history.”

Only you could have read that comment of mine and failed to see that I was actually agreeing with you. Not only that, you repeated the very arguments that I’d made there back to me!

This is what I was talking about when I said it’s good to have you on the other side. Even when you’re right, you’re crazy.

Tin Pan Alley, Minstrel Shows. Even Evangelical hymns were commercialized and pop-culture.

Angloid rightwing cope nostalgia, including the dissident type, is kind of pathetic.

Your phrasing sounds off here. Any data point necessarily affects the sample mean, of which it is an integral part. You might say the data point “can’t alter” the sample mean in the sense that the sample mean has no existence independent of the data point, but that’s just philosophical. Since the data point is part of the mean, it should likewise be part of the expected mean. Before you know its value, you make a reasonable guess, but once you know the value, you incorporate it into both the mean and the expected mean.

In case 2 the sample mean exists before you see the first result. You just don’t know its value and are trying to estimate it. That is why I said: “the observation can’t alter the actual mean, but it can alter your estimate (the expected mean).”

Yes I was agreeing with you. Only you could have read my comment and not seen that.

Also, I don’t know if anyone mentioned that if you have a sample of n IQ scores with the first score being 150 and a population mean of 100, the expected sample mean should be ((n-1)*100+150)/n.

“You have to look at the big picture – who started this war?”

Israel did. 1920’s when they first migrated to Palestine, there were riots primarily started by Jews and into the 1930’s as well. Then in 1948, Israel launched various excursions into Palestinian towns and took their land. And so it goes down through the decades.

Big picture thinking doesn’t always equal Israel is always 100% in the right. There must be a reason why a large swath, chunk of the non-Islamic world sides with Palestinians.

Yes, yes, but PRE-late 19th century (in other words, PRE phonograph, and PRE-sheet music), Americans learned the same songs, sometimes with regional lyrics (different lyrics, different stanzas etc) to the songs. From there, each generation memorized the lyrics, and passed them down to the next generation.

Not always cut and dry.

there were riots primarily started by Jews

This is a lie.

There must be a reason why a large swath, chunk of the non-Islamic world sides with Palestinians.

There must be a reason why a large swath, chunk of the non-Islamic world is a corrupt shithole that hates America and democracy too. Since when do Americans take their marching order from Congo?

There is no point in going back to the 1920 in Palestine any more than there is of going back to the 1920s in German or Russia. That’s all ancient history now. However it came into existence, Israel exists now and it is not going away.

By attacking on October 7, Hamas unleashed the gates of hell upon themselves just as Japan did on Dec. 7. The Palestinians should be glad that Israel doesn’t nuke them the way that the Americans nuked the Japanese the minute we had nuclear weapons. They should be glad that Israel has only killed 18,000 people so far, mostly Hamas fighters (don’t buy the fake Hamas casualty #’s) when America killed millions of Japanese under similar circumstances.

If you want your enemy to be really, really pissed at you and want you dead, the best way is to do a surprise attack and kill a lot of their civilians. Compared to the way that America treated the Japanese (we put our Japanese population in concentration camps) the Israelis have behaved admirably but nothing is going to be good enough for the people who (not so) secretly want Israel to lose. Y’all are not humanitarians, you are just on the other side. You are rooting for the Japs to win.

Printed sheet music existed from earliest American days.

Here is “Pop Goes the Weasel” from 1854 (printed in Buffalo, NY)

https://www.loc.gov/resource/music.mussm2-sm1854-231520/?sp=1&r=-0.809,0.115,2.618,0.941,0

The Library of Congress has even earlier stuff going back to 1820 and before that stuff was brought over from England.

A little of that stuff did occur, some ballads were passed down, but many of them were based on popular, printed broadsides.

England ceased being a traditional society from the 1500s on. The Reformation and conversion to Capitalism, and subsequent Puritan, Low Church, and Whiggish efforts ended local saints cults and traditional communal relationships. The Enclosure movement and the Clearances were about rationalizing the means of production and getting rid of inefficient, i.e. traditional, economic and social systems.

A person or community can’t be “traditional” in a hyper-individualized, Protestant, Capitalist system. A person can larp, however. That’s basically what the royal family has been doing to justify their status.
Tolkien, a larper, himself commented on how the English really didn’t have a grand epic tradition as found in traditional societies, prompting him to create one.

America was founded by the time this process was well under way. If anything, America was even further down the road of hyper-individualism, rootlessness, and economic rationalism, masked imperfectly by larping. Witness the Southern planter moving ever further looking for virgin cotton land, comparing himself to Walter Scott’s heroes.

some ballads were passed down,

This happened more in isolated places like Appalachian hollows. In any place that was in regular communication with the outside world, people wanted to wear the clothes that were in style in the big cities, sing the songs that were popular there, etc.

Yojimbo/Zatoichi are Japanese names. I looked them up. I don’t know if he’s Japanese but it’s a distinct possibility.

https://en.wikipedia.org/wiki/Winter%27s_Bone

A tiny little masterpiece. Mostly because of JL.

Still waiting for the JL / Margot Robbie remake of Persona. A gold mine, I tells ya.

I doubt it. Zatoichi Meets Yojimbo is the name of a Japanese samurai type movie. Lots of Western guys have fantasies about themselves as brave samurai warriors, which is actually kind of funny because the average samurai was maybe 5’3″ tall.

“A person or community can’t be “traditional” in a hyper-individualized, Protestant, Capitalist system.”

Actually, the US does have a traditional localized community–the Amish. They’ve been here since at least the mid. 1700’s and for the most part, still act and behave accordingly. Directly traceable back to the Protestant Anabaptist sects of Germany during the 16th century.

In a sense, this is what Dickens was getting at thru Feziwig in a Christmas Carol: One tends to hold on to the localities, to preserve a way of life–one that one has loved and knew from father to son, father to son. Scrooge didn’t listen and went with the times. Feziwig couldn’t compete with the new modern ways and was bought out when it was basically too late to make a major profit, but profit wasn’t chiefly what Feziwig wanted most. He wanted to preserve the localized communities.

Dickens wrote this during the first half of the 19th century, as the Victorian Age was just beginning, but, it was clearly too late in England for the localized pre-modern traditional ways, just as it was clearly too late during the time when a Christmas Carol was set–about 30 yrs earlier (beginning of 19th century).

America was about a couple generations behind England in the Industrial Revolution and was still fairly localized and regional in its communities. Not nearly as hyper-individualized as you describe, until after the Civil War.

But even then, the period of say, 1870-1920 as the capitalist system in the US was getting under full sway and starting to uproot localities, etc. (to say nothing of runaway immigration from non-traditional places that had immigrated to the US in the past), that period would appear to be totally local community oriented when compared to 2024.

Didn’t happen on a consistent basis nor as often as we’d like to imagine. Remember, no internet, no TV, technology was quite fairly primitive. The further away from town and town living a community was, the last thing on their minds was aping the latest fashion imported from London or Paris.

Yes, there actually was more to early America than BOS, PHI, and NY as most Americans didn’t live in these places but lived on farms. And of course for most of the 19th century the US was more focused on taming the frontier, moving West, and living off the land, eking out a basic subsistence.

Actually the US’s population was majority rural until about 1920 or so.

Appalachia managed to preserve older versions of America, but I would continue to maintain America wasn’t a “traditional” society in an old world sense by the time Appalachia was settled.

Appalachian ballads were, once again, based on popular broadsides, some of them not even that old.

Actually, they were aping the latest fashions. The local elites did very much so. If this weren’t so, you would have seen Appalachian people dressing the same for centuries. Devil Anse Hatfield dressed like a late 19th-century person, he wasn’t even dressed in the fashion of the 1830s in those famous photos of him.

The Appalachians were not Ladakh or the Caucasus, genuinely isolated, traditional places. They weren’t Gorals in the Carpathians. They were influenced by Boston, Philadelphia, and New York through broadsides, fashion, and religiously. They were cul-de-sacs of an essentially modern, rootless, society

1st and 2nd Great Awakenings, you guys every heard of those? The spread of the Baptist church by New England preachers?

Dickens and the Victorian Christmas culture was larping. No different than Highland lairds or Queen Victoria wearing kilts as they destroyed the last remnants of pre-modern Highland life.

You can’t have a genuinely traditional society in a capitalist, Evangelical society.

The Amish are a weird exception that proves the rule. Would you say the Appalachians are as coherent, and traditional as the Amish?

Here is Devil in his later years (circa 1920). I think he could walk down the street today and not attract much attention with his clothes (even his beard is back in fashion). Men’s suits have changed very little in the last century.

His suit could use a good pressing though. I guess there were no Korean dry cleaners up in the holler.

https://scontent-lga3-2.xx.fbcdn.net/v/t39.30808-6/411191203_951428356345410_4323887006452932142_n.jpg?_nc_cat=101&ccb=1-7&_nc_sid=5f2048&_nc_ohc=T8vY2VWst5QAX_l1nNF&_nc_ht=scontent-lga3-2.xx&oh=00_AfCOkXJ4x8vFGBSK6r_SS-g7FdQ--aZN3N5QBjCYvdqX3A&oe=6610C59E

All of European “traditional dress” (dirndls and lederhosen and all that crap) is all a LARPing revival, not just kilts. So are the “Indian” outfits that “American Indians” (who are usually mostly white) wear at pow wow and such. Mass produced fabric and clothing is so much cheaper than traditional stuff that the authentic stuff disappeared from the market long ago.

It’s not a black & white thing. The Appalachians weren’t truly isolated like Papuans living in the jungle but neither were they in daily contact with outside society. Some of the Appalachian music that was collected in the early 20th century had roots going back to Elizabethan times. Certainly their dress was more modern but some of their music was quite old.

Some of the embroidery patterns one sees in Eastern Europe probably has ancient or medieval origins, but you are right, a lot of the over the top “folk” costumes of Central Europeans in particular looks recent, some of it looks like somebody just went crazy with 17th and 18th century styles. It certainly doesn’t look medieval. Medieval clothing was much simpler. The lederhosen is just larping.

Certain clothing styles in India, however, seem to have more continuity.

But, I’ve read in forums that the “Indian” style you see in the diaspora with shalwar and kameez is inauthentic in that many Hindu Indian-Americans who wear them at weddings, particularly South Indians, had recent ancestors who more likely wore dhotis. Shalwar was more of a North Indian or Muslim thing. Same goes for female pajamas.

No, Israel did not. Jews migrated to the Near East to build farms, business, and public institutions. They also employed Arabs in those enterprises. Israel has continued to build and prosper (and providing a congenial matrix for its Arab citizens). What prevents the Arabs on the West Bank and Gaza from building and prospering is their own dispositions and objects.

He wasn’t Chekhov, but who is?

You and “Alice from Queens” might hit it off.

“1st and 2nd Great Awakenings, you guys every heard of those? The spread of the Baptist church by New England preachers?”

Baptists of the 18th century were Reformed Calvinists, which still maintained a connection to New England colonial culture. This was true especially during the 1st Great Awakening.

“Dickens and the Victorian Christmas culture was larping.”

Dickens’ character Feziwig was looking back at an earlier time, when things were simpler, less stressful, less complex. A Christmas Carol, written at the very beginning of Victoria’s reign, was not set during the Victorian Age, but in the late Georgian Age. Feziwig’s memories recalled to life, so to speak, took place in his childhood, (ca. 1750-60), or at the very beginning of the Industrial Revolution in England.

No Larping going on. Nostalgia, absolutely. For a (belived to be) much simpler time as well as much more community based.

“You can’t have a genuinely traditional society in a capitalist, Evangelical society.”

Certainly you could. You most definitely could, and of course the US did have a genuinely traditional society, specifically in Ante Bellum days. After the Civil War, it slowly and consistently eroded until here we are today.

“The Amish are a weird exception that proves the rule.”

It also contradicts your assertion that no such traditional society existed in a capitalist, Evangelical nation. The Amish most definitely were not Catholic, they were Protestant.

Facts.

“Would you say the Appalachians are as coherent, and traditional as the Amish?”

What, now? Of course not. 200-250 yrs ago? More or less….yes, they were, with some notable differences. Certain swathes of Appalachia were certainly more cohesive. It was the frontier, which was strongly Protestant, and yet had to make their way during an ever changing world. So they maintained their cohesive communities as best as they could. Obviously not for very long (especially when compared to the Amish, who successfully maintained their traditional communities well into the second half of the 20th century—so for nearly 300 yrs, the Amish maintained a traditional, localized community that was rural and religious based. Not too shabby)

The pre-Industrial, simpler world you speak of was already drastically changed as a result of the Enclosures, Dissolution of the Monasteries, Cromwell’s interlude, and the Glorious Revolution.

Anglo-Saxon societies, including rural people albeit at slightly delayed pace, were subject to fads that speed along in the course of literal decades. Traditional societies in other parts of the world, including Europe, were much slower in their pace of change.

Well if the Establishment lets him run, he may be right about that.