/sci/ - Science & Math

[Anonymous No.15438853]

Any geneticists willing to confirm this?

[Anonymous No.15438854]

[Anonymous No.15438858]

[Anonymous No.15438859]

[Anonymous No.15438860]

[Anonymous No.15438862]

[Anonymous No.15439151]

>>15438853
I can confirm that Stephen hsu is based as hell. Met him once. Good guy.

[Anonymous No.15439167]

>>15438853
I use statistics, ML and other kinds of regression extensively, and no he's not right. He is greatly generalising a result.

Statistics is not causation. Hypothesis testing needs to be done very, very carefully. Let's say you go back a hundred and fifty years and attempt to some kind of model to predict the likelihood of being wealthy based on genes. Anyone who's genes matches specific subsets of Jewish and upper class Anglo Saxons would have the highest chance. East Asians like Chinese and Japanese would have almost no chance. Today, this is radically different. Statistics describe the state of the population. It is not, in itself, a prediction of causality.

Causality requires understanding the sequence of events because cause and effect. Regression just gives you the state of things as recorded in the data.

[Anonymous No.15439171]

>aliens revealing themselves
retard alert. nature over nurture any day though
why is a human smarter than a dog? purely socio-economic factors

[Anonymous No.15439178]

>>15439167
Yep.
>Causality requires understanding the sequence of events because cause and effect. Regression just gives you the state of things as recorded in the data.
Hard to fathom how so many don't get the difference between descriptive and predictive analytics.

[Anonymous No.15439187]

>>15439167
This is a fair point but you chose the most environmentally controlled example in the list. Things like height and IQ aren’t nearly as environmentally controlled if you assume the person grew up in the first world.

[Anonymous No.15439191]

>>15439151

[Anonymous No.15439193]

>>15439167
You could have just said "correlation is not causation" and saved everyone the time it takes to read this drivel.

[Anonymous No.15439194]

>>15438853
Provably impossible. Most likely a larp, a slight chance of actual delusion of the described people.

[Anonymous No.15439210]

>>15439193
>You could have just said "correlation is not causation" and saved everyone the time it takes to read this drivel.
Yet every time someone does leave it at that there are hordes of retards like you screaming about how it so totally is if and only if it supports what you want to believe.

Do everyone a favor and save us the time of dealing with your delusional narcissistic smugness by lurking more.

[Anonymous No.15439219]

>>15439210
There is not one single human being left on earth who has not learned that correlation != causation and yet anytime any study is done some retard has to mention it as if it’s some great insight.

[Anonymous No.15439220]

>>15439219
>There is not one single human being left on earth who has not learned that correlation != causation and yet anytime any study is done some retard has to mention it as if it’s some great insight.
Lurk more newfag

[Anonymous No.15439245]

>>15439219
Yet few actually apply it in practice. AI is especiaply prone to hallucinate such results. It is very well possible that they applied some machine learbing on it, but it learned to guess the residence of the people and the environmental factors they may experience, and give seemingly reliable results, but it won't work when applied in practice. (this is similar to the glorified twin studies, which in fact can't distinguish factors that affect the fetus in the womb from heredity)

[Anonymous No.15439253]

>>15439219
Yet few actually apply it in practice. AI is especially prone to hallucinate such results. It is very well possible that they applied some machine learning on it, but it learned to guess the residence of the people and the environmental factors they may experience, and give seemingly reliable results, but it won't work when applied in practice. (this is similar to the glorified twin studies, which in fact can't distinguish factors that affect the fetus in the womb from heredity)

[Anonymous No.15439261]

>>1543918
The problem with IQ it was that it was always meant to detect subnormal intelligence. If you apply it to the right side of the bell curve the correlation becomes very weak. There's a big increase in income between someone with 80 vs 100 IQ, a moderate increase between 100 to 120, and from then on there's low to no correlation.

There's also a lot training you can do to see the patterns in the Ravens Progressive Matrices test. If you include verbal tests, then that has a lot to do with your English education as well. MENSA insists it's cheating because they believe it's fixed, but if its fixed why can it be trained? Many aspects of intelligence can also be trained. Like memory training can 10x-100x your recall of lists, number sequences etc.

[Anonymous No.15439263]

>>15439193
That's a statement not an argument

[Anonymous No.15439266]

>>15439187
The problem with IQ it was that it was always meant to detect subnormal intelligence. If you apply it to the right side of the bell curve the correlation becomes very weak. There's a big increase in income between someone with 80 vs 100 IQ, a moderate increase between 100 to 120, and from then on there's low to no correlation.

There's also a lot training you can do to see the patterns in the Ravens Progressive Matrices test. If you include verbal tests, then that has a lot to do with your English education as well. MENSA insists it's cheating because they believe it's fixed, but if its fixed why can it be trained? Many aspects of intelligence can also be trained. Like for working memory, memory training can 10x-100x your recall of lists, number sequences etc.

[Anonymous No.15439269]

>>15439219
The fact that people keep asking the same type of question shows that there people on earth who don't understand it

[Anonymous No.15439274]

>>15438862
The problem is that singular genes never affect just a single trait. Suppose you have two gene variants, one increases height and the other decreases height by the exact same amount. What's the difference between having both of those gene variants and not having them? Someone with both genes would have the exact same height as someone without them, and we see this exact thing. Genes which affect such traits would also tune other metabolic pathways outside of bone growth, such as the immune system, hormones, etc. so it'd be a nightmare to figure out how to perform such selection without screwing up someone's immune system or metabolites or something else.

[Anonymous No.15439383]

>>15439266
High ceiling tests show extremely high iq for research scientists, particularly Nobel prize winners. The correlation between iq and solving hard problems doesn’t stop at medium iq, it’s just that other qualities can become more important for financial success.

I’m sure they can be gamed to some degree but things like the correlation in scores between identical twins raised apart have me convinced that those tests are getting at something real. It’s not a perfect measurement but it’s the most solid thing that has ever come out of the soft sciences.

[Anonymous No.15439384]

>>15439220
Eat a dick, faggot

[Anonymous No.15439387]

>>15439253
And it’s very well possible that when I train a model to determine dog breeds it’s basing its guesses off of the background rather than the dog. So you test it thoroughly and get a diverse, representative dataset, and try to remove bias. There is no way they didn’t consider this on day one.

They are also able to predict the tallest brother out of 10 with something like 1/4 accuracy. What confounding variables could there be in there?

[Anonymous No.15439407]

>>15439266
Critique of IQ generally boils down to it being imperfect. Which really doesn't matter as long as it's the most solid metric we got

[Anonymous No.15439524]

>>15439383
Studies of PhD holders put their IQs somewhere near 125. Most Nobel prize winners have never taken an iq test and published it, iq scores you hear of them are almost always "estimated".

As pointed out by other people, twins were raised in environments but they did not substantially differ because of the situation. They were generally in the same country, had the same cultural upbringing, same curriculum etc.

The most solid thing to come out of cogsci is much more interesting, it's studies of working memory and talent acquisition. This is a specific, measurable metric of intelligence. And the latter, unlike IQ, is directly related to achievement.

[Anonymous No.15439528]

>>15439407
It's not just imperfect, it has serious errors which will compound the more it is used. Let's say you breed society for higher IQs. At a 100 your benefit starts to diminish. At 130 you basically see no gains. The irony is that it's only useful purpose is to filter retards, but it's never used for that, instead it's always used to try to make smart people smarter.

There are much more interesting results from cognitive science. For example, every genius and high achiever in any field share upbringings that are similar in key ways. They demonstrate curiosity, engage in a high level of practice over many years and are nurtured in an environment that supports them. Whether you're talking about Serena Williams or John Von Neumann these things are present in their childhood.

[Anonymous No.15439852]

>>15439266
>There's a big increase in income between someone with 80 vs 100 IQ, a moderate increase between 100 to 120, and from then on there's low to no correlation.

[Cult of Passion No.15439862]

>>15439852
>110
>that dip
Is this the smoking gun of "midwit"?

[Anonymous No.15439869]

>>15439528
a society where the average person had an IQ of 130 would be orders of magnitude better than the current one

[Anonymous No.15439874]

>>15439862
It's when people start being able to pursue worthless degrees instead of becoming a shop manager.

[Anonymous No.15439880]

>>15439266
>>15439383
>>15439407
>>15439524
>>15439528
IQ isn't just inverse, it's in a large part the measure of gullibility. The higher the IQ, the more gullible you are. Most Mensa members believe in some schizo level nomsense. High IQ people have zero sense what to think about and return nonsensical resupts, as they lack the capacity to tell if their solutions make sense. IQ has nothing to do with the ability to tackle complexity, most IQ questions are trivial. Some people learn to score high by learning what kimd of answers the tests want, but those who answer intuitively, they are idiots.

[Anonymous No.15439882]

>>15439266
>>15439383
>>15439407
>>15439524
>>15439528
IQ isn't just imperfect, it's actually inverse, it's in a large part the measure of gullibility. The higher the IQ, the more gullible you are. Most Mensa members believe in some schizo level nomsense. High IQ people have zero sense what to think about and return nonsensical resupts, as they lack the capacity to tell if their solutions make sense. IQ has nothing to do with the ability to tackle complexity, most IQ questions are trivial. Some people learn to score high by learning what kimd of answers the tests want, but those who answer intuitively, they are idiots.

[Anonymous No.15439885]

>>15439874
>>15439852
So you're just posting self refuting contradictory cherrypicked graphs as if they support some weird point? That second one especially is provably nonsense given the IQ disparity between men and women is nowhere near that high. In fact I went to the source just now and it's some pre-2011 flawed proprietary crap nobody can track down and that graph isn't general IQ it's just the quantitative reasoning slide. There's no correlation for vocabulary or other reasoning from the GRE dataset it claims to use.

Are you using bullshit on purpose for some stupid reason, or have you never read your own sources?

[Anonymous No.15439886]

>>15439528
>Let's say you breed society for higher IQs. At a 100 your benefit starts to diminish. At 130 you basically see no gains.
Ok so let's stop at 120, what's the issue ?
But your original point was about income. Are you sure that other correlations too disappear at 130 ?

[Anonymous No.15439904]

>>15439886
The higher the IQ, the more gullible schizo you are.

[Anonymous No.15439937]

>>15439882
>t. coping retards

[Anonymous No.15439960]

>>15439937
No. Painful experience with these supposedly brilliant, but in reality stupid as fuck people.
At the very best some kind of mental illness gives you a high score in the tests, but most likely it's just inverse to reality, but some actually smart people manage to score high by answering what is wanted not what makes sense. (I know because, I did)

[Anonymous No.15439979]

>>15439937
No. Painful experience with these supposedly brilliant, but in reality stupid as fuck people.

At the very best some kind of mental illness earns you a high score in the tests, but most likely it's just inverse to reality, but some actually smart people manage to score high by answering what is wanted not what makes sense. (I know because, I did)

[Anonymous No.15439982]

>>15439167
This is just a more longwinded variation on the fact that heritability depends on the degree of variance in the environment. These attempted counterexamples never address the fact that the predictive value of genetics *within* any given population is always staggeringly high. Moreover, your particular counterexample is weak, since wealth depends largely on what is valued by society, which does not have to be grounded in anything genetic. In a centrally planned economy, it almost certainly will not be.
>>15439266
>The problem with IQ it was that it was always meant to detect subnormal intelligence.
This argument is so stupid and tiresome. IQ tests have gone through 100 years of continuous research and development and now have strong support for predictive value at the upper end. There are literally 100s of inventions in society that started in one context before being adapted to a different one.
As one small example: "Beyond the Threshold Hypothesis: Even Among the Gifted and Top Math/Science Graduate Students, Cognitive Abilities, Vocational Interests, and Lifestyle Preferences Matter for Career Choice, Performance, and Persistence"
https://journals.sagepub.com/doi/10.1177/0963721410391442
>There's a big increase in income
Again, that has more to do with the central planning of income levels than about the relevance of IQ to productivity.

[Anonymous No.15440002]

>>15438853
just from the typos, I can tell this person is talking using their anus

[Anonymous No.15440009]

>>15439982
>These attempted counterexamples never address the fact that the predictive value of genetics *within* any given population is always staggeringly high.
Not that guy, I just had to comment due to how ridiculously and blatantly false this statement is. The actual predictive success is so poor it is dubbed the "missing heritability problem". What you wrote is not only false, it's so false it contradicts the past 50 years of genetics research if not more.

You aren't as wrong with the rest of your writing but you're on the level of flat earther for what you wrote about predictive value for things like GWAS. To be properly technical polygenic risk scores are nowhere near as high in the general population for any phenotype, disease or otherwise, compared to the descriptive genetic correlates of past research. Depending on the phenotype, like IQ, it can be as low as .1 or .15 compared to descriptive correlates such as .8 in twins. Think what you will, but at least get the facts straight so you're not THAT fucking wrong about it.

[Anonymous No.15440174]

>>15440009
Because it isn't genetic. Most likely there is no major variability in intelligence, there is only damage from the lack of heavy metals.

[Anonymous No.15440215]

>>15438853
it is mind blowing how damaging leftist ideology has been for science. Just amazing

[Anonymous No.15440660]

>>15438859
>>15438859
>IQ
Pretty based. Inpute.me did this for free until it got bought over a year ago. Also found out im tremendously prone to osteopetrosis.... wish I had downloaded all my sorted data

[Anonymous No.15440759]

>>15439266
>MENSA insists it's cheating because they believe it's fixed, but if its fixed why can it be trained?
No, they believe fluid intelligence is fixed, and testing it accurately requires someone not having spammed similar problems before because then you're really testing crystalline intelligence, so the score isn't accurate. You're only pumping up the score by fudging the data, not actually increasing your intelligence.

[Anonymous No.15440778]

>>15439210
I see the trope 'correlation != causation' more abused to dismiss results that people don't like, than ever to reach a greater understanding.

[Anonymous No.15440788]

Why didn't they make IQ an absolute score? It's so confusing that it's supposed to be normalized to a population, but obviously that's a moving target.

[Ghosto No.15440789]

>>15438853
Actually, chinese/Asians are based because they have good spatial memory. So they are better at everything expect inventing new math. New math usually comes from Asians not from Asia.

[Anonymous No.15440791]

>>15440788
Because the point of IQ was never "how smart are you," it's "are you smart enough to function?"

[Anonymous No.15440795]

>>15440791
But why didn't they just invent a new metric to fix this, it would be so beneficial to scientific advancement in the long run.
You'd be better able to track people's development and cognitive decline for example.

[Ghosto No.15440799]

>>15440795
An easy answer to IQ is "how much information do you obtain from the environment"

The real problem with denying IQ is exactly what you see now...bad science everywhere and no progress. This is what happens when you let people do science to create diversity, but the diversity comes at a loss of function, since no new information can be gained from the environment.

[Anonymous No.15440800]

>>15439982
>predictive value of genetics is staggeringly high
Which still doesn't mean anything. If you look at number of fingers as a phenotypical trait, the heritability would actually be very low because most variability comes from people's fingers getting cut off on accident. However number of fingers is definitely an inherited trait

[Anonymous No.15440805]

>>15440009
>Not that guy, I just had to comment due to how ridiculously and blatantly false this statement is.
You misunderstand. I don't mean GWAS, SNPs, or any of those caveman attempts to model genetics at all. I mean actual genetics. For instance, if you have one member of an identical twin pair within a population, you can predict almost everything about the other twin with high accuracy, even if they were separated at birth and never met.

I am absolutely aware that current models are awful at predicting outcomes from DNA sequences. However, for the record, with LLMs and AI in the fold, even that is going to change very soon.

[Anonymous No.15440808]

>>15440174
Oh god, this retard is here to retard up another thread with his 50 IQ takes
>We need to eat more lead and mercury guys

[Anonymous No.15440843]

>>15440805
>You misunderstand. I don't mean GWAS, SNPs, or any of those caveman attempts to model genetics at all. I mean actual genetics.
I did not. Did you mean to write exclusively twin populations? If so, that is not what you wrote, so clarifying that might help. As a side matter, declaring one thing "actual genetics" as if population genetics don't count is silly.
>>15439982
>These attempted counterexamples never address the fact that the predictive value of genetics *within* any given population is always staggeringly high.
You cannot, from twins, make this inference to general populations.
>>15440805
>For instance, if you have one member of an identical twin pair within a population, you can predict almost everything about the other twin with high accuracy, even if they were separated at birth and never met.
Which does not really tell you much. Doesn't even tell you the reactive norm of the genotype. Mostly it just tells you the obvious, that genotype correlates with phenotypes within a similar enough postnatal environment range. "separated at birth" doesn't really tend to change that much, since you're not exactly shipping one to Mongolia to live as a nomad while raising the other to be an academic.

[Anonymous No.15440874]

>>15440843
A clarification. In this context I should've written "phenotypic plasticity" rather than "reaction norm", though the former is derived conceptually from the latter. Phenotypic plasticity is less esoteric of a concept in this context, and far easier to find relevant research on for this as well.

[Anonymous No.15440875]

>>15439524
>Studies of PhD holders put their IQs somewhere near 125
I didn't say PhD holders, because any retard can sign up for a PhD at some shit school

>Most Nobel prize winners have never taken an iq test and published it, iq scores you hear of them are almost always "estimated"
Anne Roe studied 64 of the best scientists in the US in 1952 (20 biologists, 22 physicists and 22 social scientists), including Nobel Prize winners. Her findings regarding global IQ can be summarized as shown in the table I attached.

>As pointed out by other people, twins were raised in environments but they did not substantially differ because of the situation.
Twins raised apart have an IQ correlation of 0.78, unrelated siblings raised together have an iq correlation of around 0.05 (I'll go look for the exact numbers). The only way you could explain that without iq being almost 100% genetic is if the experiences in the womb have a very high impact. That is possible, and can't be ruled out (yet).

[Anonymous No.15440881]

>>15439524
>>15440875
Found the exact numbers for identical twins raised apart vs unrelated siblings raised together.

Assuming you don't starve as a child or suffer severe brain damage, iq is basically 100% genetic.

[Anonymous No.15440886]

>>15439524
While we're at it, here's the results of coaching vs not coaching kids to study for the SAT. People often give the SAT as an example of why IQ scores don't work, since the SAT is similar to an IQ test, and you can study for the SAT. This data pokes a big hole in that argument.

[Anonymous No.15440923]

>>15439885
The %female one isn't necessarily wrong, you're just interpreting it incorrectly. It doesn't imply that men have higher average IQs than women. That can be directly tested and is widely known not to be true.

I think the most obvious option is that men are more drawn to working with things rather than people (also tested and shown to be true), and that going to university for those things requires a higher iq, so men are more filtered. Low IQ men go on to different professions, that also happen to involve working with things.

Another option is that men's physical strength gives them access to professions that require manual labor, so they can make decent money without going to university. University then only makes sense for men who have a high enough iq that they can study things that will make them more money than trades.

[Anonymous No.15440936]

>>15440843
>I did not. Did you mean to write exclusively twin populations?
Lol, you're going way too deep. I literally just meant the proportion of mental and physical outcomes within a population that is explainable by genetics. This is independent of our capacity to model those outcomes from genetic sequences.
>declaring one thing "actual genetics" as if population genetics don't count is silly.
Again, by "actual genetics" I was mere emphasizing the full non-linear interactions across the genome, rather than the weak approximations used in models.
>Mostly it just tells you the obvious, that genotype correlates with phenotypes within a similar enough postnatal environment range.
Don't undersell it. It tells you that genotype very strongly predicts phenotypes within an environmental range that covers almost the entire Western first-world experience. That was my point.

[Anonymous No.15440940]

>>15440843
I think what >>15440936 means to say is that the % that can be shown to be hereditary is extremely high, even if we don't yet know how to predict it.

[Anonymous No.15440961]

>>15440881
>>15440875
I'm that other anon >>15440843. While I agree it's genetic as a matter of refuting some absurd total blank slatism, I don't think you're correctly thinking about that inference either. In particular,
>Assuming you don't starve as a child or suffer severe brain damage, iq is basically 100% genetic.
If you starve it's still 100% genetic. What you seem to be trying to get at, same for here >>15440886, is instead the notion that IQ is not very plastic with respect to a given genome or individual. I agree with you on some points. Namely, that nutrition would account for most of the observed plasticity and seems to account for most of the flynn effect as well. I'm nitpicking what appear to be some mistakes in concept, or use of jargon, as it seems to me you'd be interested in learning more and phenotypic range/reaction norm concepts would help you explain it better in future.
>>15440886
You're also making what appears to be a mistaken inference from this data. This is not demonstrating that coaching has no benefit to a given individual, only that the distribution compared to people not pursuing coaching ends up similar. The data source does not and cannot disentangle sample bias such as self-selection from the outcomes.
>>15440923
>The %female one isn't necessarily wrong, you're just interpreting it incorrectly.
I think you're misinterpreting me there. I am refuting the usual manner in which the chart is presented, hence refuting an unstated conclusion. I agree with you as concerns personality driving factors within a given IQ SD of course. If that was you, and that was not your intent, I apologize.

To be continued

[Anonymous No.15440985]

>>15440936
>I literally just meant the proportion of mental and physical outcomes within a population that is explainable by genetics.
Okay. The problem is how you phrased what I originally responded to, >>15439982
>the fact that the predictive value of genetics *within* any given population is always staggeringly high
This is false as earlier noted. The problem is those words as you've used them do not mean what you have just now clarified. It isn't predictive, but descriptive and associative. Trying to help here, what you should've written would be along the lines of "The fact that heritability is so high", or without the oft-misunderstood heritability concept, "the fact that genotype correlates so highly with phenotype in twins for similar environments". It is not predictive, it is descriptive. Since you appear to be wanting to refute a kind of blank slatism, and to be absolutely clear I am not that anon nor do I support that notion.
>Again, by "actual genetics" I was mere emphasizing the full non-linear interactions across the genome, rather than the weak approximations used in models.
Here, too, I am not sure you understand the implication of your words with respect to genetics. Reads to me as implying something about non-additive interactions, which doesn't make sense here. I think you mean "the whole of an individual genome" or "individual genotype" for short. Even so, that contradicts "population" as quoted above. Point is, you aren't doing very good with your words, and since you seem earnest that feedback with examples might help.
>It tells you that genotype very strongly predicts phenotypes within an environmental range that covers almost the entire Western first-world experience. That was my point.
Your posts so far do not support that point. Here I can disagree quite actively. Twin studies do not support inferring that genotype predicts phenotype in disparate SES within 1st world ranges, and I'm not sure why you think they can. Explain?

[Anonymous No.15440997]

>>15440940
>I think what >>15440936 means to say is that the % that can be shown to be hereditary is extremely high
Thanks. I did get there from that very same post I've now replied to as well, but I appreciate the help in case I had not.

[Anonymous No.15441012]

>>15440875
>Anne Roe studied 64 of the best scientists in the US in 1952 (20 biologists, 22 physicists and 22 social scientists), including Nobel Prize winners. Her findings regarding global IQ can be summarized as shown in the table I attached.
Brutal blackpill for midwits

[Anonymous No.15441018]

>>15440799
>An easy answer to IQ is "how much information do you obtain from the environment"
That's very obviously the inverse of IQ.

[Anonymous No.15441019]

>>15440805
>I mean actual genetics. For instance, if you have one member of an identical twin pair within a population, you can predict almost everything about the other twin with high accuracy, even if they were separated at birth and never met.
That doesn't imply genetics. If one twin has FAS, the other will have it as well, even though it isn't genetic.

[Anonymous No.15441022]

>>15440985
>It is not predictive, it is descriptive.
You appear to be caught up in an attempt to make a semantic and technical distinction where one is not needed.
The total genetics of a person are predictive within a population. That statement is literally true. If I have one identical twin, I can predict attributes about the other one. There is no confounding factor causing both the genetic sequence and the phenotype.
>Even so, that contradicts "population" as quoted above.
"An individual genome is predictive for outcomes within a population" (again, irrespective of our ability to model those predictions).
>Twin studies do not support inferring that genotype predicts phenotype in disparate SES within 1st world ranges,
Heritabilities of 0.7-0.8 for behavioral outcomes were obtained as a result of averaging over first world environmental variation. Twins separated at birth very frequently did not end up in the same SES, yet still showed overwhelming similarity.
Even more importantly, many of these studies were performed decades ago, when differences in SES had much more profound impacts on environment. A low-SES child in the 1950s could easily be malnourished, lead poisoned, and uneducated. In the present day, former concepts of malnourishment are unheard of, and even the poorest individual has the entire repository of human knowledge in their pocket. The environment has never been so non-limiting. Heritabilities are probably higher now than ever before in history.

[Anonymous No.15441026]

>>15440961
>it seems to me you'd be interested in learning more and phenotypic range/reaction norm concepts would help you explain it better in future

I'd love to read more. Any suggested reading?

The most ideal case I think would be to try to actually play with the data myself, so if you know of any good datasets I could mess around with I'd appreciate it.

[Anonymous No.15441031]

>>15441018
>>15440799
>An easy answer to IQ is "how much information do you obtain from the environment"
That's very obviously the inverse of IQ. High IQ people are almost totally incapable of doing that. That's where the "denial" comes from. IQ is pretty much the complete opposite of the usual understanding of intelligence. The higher your IQ, the closer you are to a dumb calculating machine that can't understand what it's doing and with no capacity of actual thought.

[Anonymous No.15441032]

won bronze at the IMO at the age of 11- current grad student at HARVARD; pride and joy of peru

[Anonymous No.15441034]

>>15441018
>>15440799
>An easy answer to IQ is "how much information do you obtain from the environment"
That's very obviously the inverse of IQ. High IQ people are almost totally incapable of doing that. That's where the "denial" comes from. IQ is pretty much the complete opposite of the usual understanding of intelligence. The higher your IQ, the closer you are to a dumb calculating machine that can't understand what it's doing or anything else, and with no capacity for actual thought.

[Anonymous No.15441036]

>>15440961
>You're also making what appears to be a mistaken inference from this data. This is not demonstrating that coaching has no benefit to a given individual, only that the distribution compared to people not pursuing coaching ends up similar. The data source does not and cannot disentangle sample bias such as self-selection from the outcomes.

I'm struggling to understand some of your wording here, but my understanding is that the data shows that an individuals PSAT score is extremely correlated to their SAT score, whether they received coaching or not.

If coaching helped, we would expect to see an offset in the lines between the white and black points, or a different slope.

I understand that selection comes into play wrt whether or not the individual chooses to receive coaching, but just by eyeballing the data it doesn't seem like that correlates strongly to PSAT score, so is it not effectively random? Or at least uncorrelated to PSAT score? (assuming my ability to eyeball the data means anything, which I know it doesn't)

[Anonymous No.15441043]

>>15441019
This is a good point, and is currently impossible to disentangle from genetics afaik. I think its instead makes sense to separate the two factors into pre-birth and post-birth categories.

If using surrogates becomes more popular we may be able to disentangle this in the future, at least for siblings.

Actually maybe we could do it now for half siblings with the same mother vs half siblings with different mothers. Doesn't tell you about shared womb experience exactly, but tells you about roughly the prenatal environment.

[Anonymous No.15441051]

>>15441022
>The total genetics of a person are predictive within a population. That statement is literally true. If I have one identical twin, I can predict attributes about the other one.
If that were true phenotypic variation and discordance would not be a thing. Why the hell do you think this?
>You appear to be caught up in an attempt to make a semantic and technical distinction where one is not needed.
It's a very real distinction that seems more and more needed the more you write.
>"An individual genome is predictive for outcomes within a population" (again, irrespective of our ability to model those predictions).
All evidence to the contrary. At this point you just seem to be asserting this on faith, or fundamental misunderstanding about genetics on the order of thinking the relationship between phenotype and genotype is 1:1. In which case... no, no it isn't 1:1.
>Twins separated at birth very frequently did not end up in the same SES, yet still showed overwhelming similarity.
I'm sorry, where are you getting this from? ChatGPT or something? Twins very often end up in the same SES due to adoption regulations and scrutiny on adopting family income, and most research comes from adoption rather than familial separation. This is a very well known phenomenon about twin research and a very well documented confound.
>Heritabilities are probably higher now than ever before in history.
This is true. Everything else you're writing is like reading a textbook from the 1980s.

[Anonymous No.15441098]

>>15441026
>I'd love to read more. Any suggested reading?
Wikipedia and its citations are often a good starting point to find further literature with relevant jargon for people unfamiliar.
https://en.wikipedia.org/wiki/Phenotypic_plasticity
https://en.wikipedia.org/wiki/Developmental_plasticity
https://en.wikipedia.org/wiki/Reaction_norm
Given I just came upon this today to answer similar questions elsewhere and on 4chan, and it would help address some of your seeming confusion as to how predictable and variant outcomes are with respect to SES, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6698881/
The effect size is rather large with respect to SES confounding polygenic score. This may explain some of your seeming confusion as to the strength of phenotypic invariance with respect to twins as you seem to hold here >>15441022
>>15441036
>my understanding is that the data shows that an individuals PSAT score is extremely correlated to their SAT score, whether they received coaching or not.
Just so we're on the same page,
https://www.researchgate.net/publication/247851721_The_Effect_of_Admissions_Test_Preparation_Evidence_from_NELS88
"The Effect of Admissions Test Preparation: Evidence from NELS-88" by Derek C. Briggs.
While he attempts and his main purpose is in the attempt to account for selection bias, such as through use of the Heckman Model, it is at the wrong scope and not relevant to what I was referring to. Where I fucked up was in not explaining the scope of what I meant.

I'll explain the scope in plain terms. Per the above study, SES strongly confounds polygenic score for traits like IQ. It also strongly confounds academic achievement. As that is independently of observed IQ, it stands to reason there's a significant selection bias as to performance, regardless of coaching, right? So if IQ correlates highly with SAT, and SES so significantly confounds polygenic score prediction, it stands to reason there's also a large effect on SAT as well independent IQ.

[Anonymous No.15441108]

>>15441098
I'm >>15441026 but not >>15441022
Thanks for this, I appreciate it. I'll start reading now.

[Anonymous No.15441118]

>>15441108
Well god damn it but it works either way. Let me know if you have questions

[Anonymous No.15441133]

>>15441098
>>15441036
Also worth noting, in case you found my reasoning hard to believe due to cherrypicked (very bad) research, the high correlation claims tend to use fairly nonstandard non-WAIS IQ estimators. With regard to WAIS, usually WAIS-III or WAIS-R, or WAIS-IV, the correlations are only moderate. For full scale IQ, around r = 0.5 to round out most results across WAIS related comparisons with SAT over time.

There's some weird myth going around that SAT-WAIS correlation is stupidly high, like 0.8 or 0.85. Definitely not. I think the myth circulates due to somebody somewhere confusing the inter-correlations between academic tests with the correlation for WAIS IQ score, which is much lower and far more variable across publications. There's this random number crunching blog that nonetheless demonstrates what I mean in an accessible non-paywalled way https://randomcriticalanalysis.com/2015/06/18/on-sat-act-iq-and-other-psychometric-test-correlations/

There are very, very, large correlation differences between IQ and various tests and all kinds of confounds.

[Anonymous No.15441338]

bump

[Anonymous No.15441708]

>>15441133
My favorite article from that site
https://randomcriticalanalysis.com/2015/11/16/racial-differences-in-homicide-rates-are-poorly-explained-by-economics/

[Anonymous No.15441717]

>>15441708
>My favorite article from that site
Why? They're largely explained by age demographic differences in the U.S. Race isn't even a factor relative to age, and within-age economics does explain a lot of variance. Shame if he didn't get any of that and fucked it all up, but that can happen if you just crunch numbers without thinking about it.

[Anonymous No.15441719]

>>15441717

[Anonymous No.15441724]

>>15441719
And? That doesn't adjust for age demographics, that is just listing ages and rates of crime demonstrating what I said is true. Do you honestly think the crime rates are the same if you've a neighborhood with two hundred people 15-24 versus two? Density of age demographics obviously matters. The same goes for poverty with respect to age. This is stupidly old hat dude criminologists have been studying this and know this since the fucking beginning of criminology. It's damn near older of a fact than evolution is.

[Anonymous No.15441737]

>>15441724
Care to cite me whatever that claims age and withtin-age economics explain most or all the differences?
Also do you work for the JIDF?

[Anonymous No.15441801]

>>15441737
>Care to cite me whatever that claims age and withtin-age economics explain most or all the differences?
Literally every single publication to do with predicting recidivism and crime. It doesn't matter what descriptive statistics you toss out if the predictions don't work like you're inferring.

Nonetheless, null for violent recidivism descriptively in a large meta-analysis. You'll note most reported large effect sizes have incredibly high margins of error. You can see from the meta-analysis that most studies also individually report null correlations.
https://www.researchgate.net/publication/268102464_A_Meta-Analytic_Review_of_Ethnicity_Race_Gender_and_Age_As_Moderators_of_Violent_Recidivism_A_Focus_On_Correctional_Interventions

Stands to reason and that is a common no-brainer finding everyone knows, which is why predictive policing algorithms and recidivism algorithms not only don't use race but don't benefit from doing so. This is true for COMPAS reconstructions
https://hdsr.mitpress.mit.edu/pub/7z10o269/release/7
The prediction of recidivism based solely on age with and without race has a correlation coefficient difference of 0.011 and for violent crimes 0.08. Contained in that article are also many citations reinforcing the banal fact that age is long known to be the primary variable, and corresponding variables with age make up the variance superficially correlated to race. Predictive crime and recidivism algorithms end up the same because age and economy factors together explain a lot, so long as you factor in youth population density into it especially.

Relationship between violent crime, age, and poverty, is boringly well worn https://journals.sagepub.com/doi/full/10.1177/2158244015573359
>When poverty is controlled, the traditional age-curve persists only for high-poverty populations, in which young people are vastly over-represented, and homicide rates are elevated for all ages.
Oh look, it's age and economy. Do I need to keep going?

[Anonymous No.15441863]

>>15438862
this has some serious implications for inbreeding if it allows for the selective elimination of deleterious recessive genes.

[Cult of Passion No.15441866]

>>15441863
Would explain why its so common in ancient history.

[Anonymous No.15441885]

>>15441866
The gods of Egypt were the descendants of multiple generations of sibling pairs.

[Anonymous No.15441920]

>>15441801
Absolutely none of this was what I asked for. Your first article was specifically for violent -- which obviously is an extremely biased sample as everyone here already committed a [violent] crime.
Funnily enough, though the effect size was small it still found
>"Whites were less likely to recidivate violently than 'non-Whites' (Correlation = -0.064; p<0.001)."
and the effect size for age (-0.026) was smaller compared to the effect size for race (-0.064).
Though of course sex was the larger predictor. The sheer dishonesty of citing this article as evidence for what I asked for is astounding. You are undoubtedly jewish.
And you continue to do the same with your second article is about trying to reconstruct a proprietary algorithm that is, AGAIN, on violent recidivism. Same issues as above, and note what was said
>we show that COMPAS does not necessarily depend on race [math]\textbf{other than through age and criminal history}[/math]

[Anonymous No.15441927]

>>15441920
I think I recognise his writing style.
He shows up in a lot of these threads and argues you off course.

[Anonymous No.15441932]

>>15441920
And continuing your dis-ingeunity and dishonesty streak, your last article is making the case that age is not a significant factor in violent crime but that it is merely
>That is, adolescent risks are artifacts of the reality that the overwhelming majority of serious adolescent crimes, including homicide, and other risk outcomes are concentrated in the poorest demographics—those with poverty rates of 20% to 25% and higher—in which middle-aged and older adults are seldom found.
In other words
>Under this revised theory, young people do not “age out” of crime, they “wealth out.” The failure to consider SES as a determinant in offending may also explain why studies attempting to forecast crime trends from the age structure of the population have proven so quickly and notoriously inaccurate (Levitt, 1999; Males & Brown, 2013b).
With respect to race, their views are summarized here, citing Gould of all people (jewish too, of course)
>In that light, the “integrated theory” purporting to merge biological- and developmental-stage theories to speculate why adolescents “act the way they do” (Shulman et al., 2013a) is simply the latest resurrection of 19th century notions that combined supposedly innate biological- and phylogenic-stage notions into theories speculating why races, ethnicities, and genders “acted the way they did” (see (((((Gould))))), 1981). One only need consider the fact that homicide rates are 18 times higher among the poorest than the wealthiest teenagers—or that 51,000 California 15- to 19-year-olds in the lowest poverty brackets experience an average of just 1.1 murders per year (a European level), or that 40-agers and 50-agers subjected to the same poverty levels as older teenagers display the same levels of homicide (see Table 1)—to come to the conclusion that innate biological and developmental imperatives are at best marginal drivers of late adolescent and young adult offending.
Now fuck off you lying kike.

[Anonymous No.15441934]

>>15441717
>>15441719
>>15441724
>>15441737
>>15441801
Purely age related factors

[Anonymous No.15441942]

>>15441934
god damned yt pensioners holding up corners shops

[Anonymous No.15441952]

>>15441863
>The world of genetics is about to be revolutionized, curing disease and perfecting humanity
>So you’re saying I can fuck my sister?

[Anonymous No.15441956]

>>15441920
I picked that one for two reasons. One, violent crime and most other crimes tend to be rather distinctly different and criminologists tend to separate the two. Two, because the abstract doesn't explain the figures given and I like how easily you idiots walk into being lazy about your dishonesty.
Here's what it says about the overall racial correlates,
>>Hispanic ethnicity (correlation = −.243; 95% CI [−.601, − .196], p = .277; Figure 3) was not significantly associated with violent recidivism. Whites were less likely to recidivate violently than “non- Whites”7 (correlation = −.064; 95% CI [−.126, −.042], p < .001; Figure 4) whereas Blacks (correlation = .052; 95% CI [−.035, −.139], p = .243; Figure 5) were not sig- nificantly associated with violent recidivism.
Hispanics, no correlation. Blacks? No correlation. The white correlation? Significant with an incredibly small effect size.
>Funnily enough, though the effect size was small it still found
-0.064. You are claiming victory over a statistically significant -0.064. This is why I gave you that article. You stepped in it.
>And you continue to do the same with your second article is about trying to reconstruct a proprietary algorithm that is, AGAIN, on violent recidivism. Same issues as above, and note what was said
Same dishonesty. Only I directly quoted the effect there, too. "correlation coefficient difference of 0.011 and for violent crimes 0.08". You are claiming victory over... 0.011 and 0.08.

The only people who can't see you for what you are already belong to your religion. For everyone else, you left no doubt. The funny thing is my third citation is exactly about age-poverty and crime... and you mysteriously don't mention that at all.

[Anonymous No.15441959]

>>15441932
>bla bla bla I disagree because you're a jew
Not an argument. Gave up so soon? Did you forget to actually read while trying to cherrypick? The relationship between age and crime as pointed out in that article is co-causal. That is why the author points out age alone is not a considerable independent effect.

[Anonymous No.15441967]

>>15441956
>>15441959
And as expected, you attack not the core of the argument -- that being the ridiculousness of using violent recidivism as evidence for what I asked for -- but rather on the effect sizes. People are wise to your tricks Shlomo. Only I am stupid enough to continue to entertain you.

[Anonymous No.15441969]

>>15441956
The goal here is not to trick others with difficult language and papers. And it’s not to fool them into looking stupid or “stepping in it”. It’s to try to figure out the truth.

Can you try to state in simple, understandable terms why you think that SES is a bigger factor than race?

[Anonymous No.15441972]

>>15441952
It works for some species.
High mate affinity, better infant care investment. More of your efforts promote more of your genes to survive.

If you had a lot of wealth and power already then retaining as much of it within your genetic line as you possibly could would be appealing, just look at all the royals and banker families. But now instead of having to dilute it to avoid bad recessive disorders you can simply remove the bad genes.

[Anonymous No.15441973]

>>15441967
>And as expected, you attack not the core of the argument -- that being the ridiculousness of using violent recidivism as evidence for what I asked for -- but rather on the effect sizes.
Third link is what you asked for. I directly contradict your assertion and your "retort" amounts to irrelevant frothing and claims of "muh jews".
>People are wise to your tricks Shlomo. Only I am stupid enough to continue to entertain you.
I like to share these threads with my fellow biologists because we like to laugh at internet racists.

[Anonymous No.15441978]

>>15441973
>I like to share these threads with my fellow biologists because we like to laugh at internet racists.
This feels like meme material

[Anonymous No.15441984]

>>15441972
Unfortunately (for you it seems, though fortunately for most of us) those recessive and nonlinear interactions that are caused by inbreeding will probably be the hardest for the model to capture.

[Anonymous No.15441991]

>>15441984
Anon....I'm just pointing out to you because you might be a little retarded that the people with the money and power to do this sort of thing like to hold onto it within their bloodlines as much as possible..

[Anonymous No.15441994]

>>15441932
If it’s Stephen Jay Gould he is a well known retard.

[Anonymous No.15441997]

>>15441973
Maybe I misread. Please show me where in that 3rd article evidence was shown hat controlling for SES and age, race played no role.
>>When poverty is controlled, the traditional age-curve persists only for high-poverty populations, in which young people are vastly over-represented, and homicide rates are elevated for all ages.
Yes, that makes sense. So the variance on age is seemingly explained. What does this say of race? Has anyone ever claimed SES does [math]\textit{not}[/math] play a role? I certainly have not. In any case, then you'd be left to explain how SES differences come about etc..
But wait, if according to your 2nd article age played almost no role but it was merely SES (i.e. "wealthing" out of crime, not "growing" out of it). But in that original article here>>15441708 it found that SES was a very poor explanation of racial differences in violent crime at all SES levels (in particular, homicide). How does that work out?
>>15441994
It is.

[Anonymous No.15442000]

>>15441991
I know they do. I’m just poking fun at you because it’s a funny first thought to have when presented with this information. But you might be right. Th world will change in all sorts of ways and this might be one of them.

[Anonymous No.15442007]

>>15441994
>these skulls can't possibly have different volumes
>this old white guy was clearly lying
>let me just reweigh these lead balls.
>oh no my finger slipped
>just as I expected the old racist white supremacist clearly faked the results

[Anonymous No.15442033]

>>15438853
this honestly reeks of fisherian runaway selection except on an industrial scale. i doubt this will end well fpr the chink.

[Anonymous No.15442036]

>>15441969
>The goal here is not to trick others with difficult language and papers. And it’s not to fool them into looking stupid or “stepping in it”. It’s to try to figure out the truth.
And each citation serves that purpose as well if you're willing to do the bare minimum with honesty. The trap was only for those dishonestly refusing to put up such effort. Which are you?
>Can you try to state in simple, understandable terms why you think that SES is a bigger factor than race?
Age and SES, together, explain almost all the correlations, while adding race independently of that adds no predictive power to recidivism or crime prediction for violent or non-violent crime. Of course within that bracket there are many other aggravating factors, such as untreated psychiatric conditions, which is why SES and age explain and predict crime and recidivism so well.

As for how that also answers his question directly, that is because most of the variance is due to repeat offenders in a kind of pareto principle of crime. The vast majority do not commit crimes, and most of those who do will have prior criminal history of some sort. For example, at the federal level depending on the year. This 2018 report, for example, 3/4ths had prior convictions. https://www.ussc.gov/research/research-reports/criminal-history-federal-offenders there are a lot of reports available on the commission page for federal and they tend to repeat this theme.

This is also the case for violent crimes in other countries and for long periods of time. Sweden for eample https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3969807/
This is also the case in the United States https://www.researchgate.net/publication/318842093_Ravenous_wolves_revisited_A_systematic_review_of_offending_concentration
>>15441997
This is relevant for you as well. More on that in a minute.

[Anonymous No.15442058]

>>15441997
>Maybe I misread. Please show me where in that 3rd article evidence was shown hat controlling for SES and age, race played no role.
1. Most crimes are committed by a small group of offenders
2. Odds of offense are overwhelmingly predicted by age and SES together
3. Odds of recidivism are also overwhelmingly predicted by age and SES with virtually zero added predictive power due to race
>But in that original article here>>15441708 it found that SES was a very poor explanation of racial differences in violent crime at all SES levels (in particular, homicide). How does that work out?
Like most things, by doing it wrong. This is demonstrable by simple fact predictions work best with combined age and SES whether recidivism or otherwise, and this should obvious given the strongest pathological correlates of crime that can be treated successfully have lower treatment rates the lower your SES. His analysis has multiple contradictions with all of the other analyses published and all the corresponding factors known about behavior and pathology and crime for age and SES. So what's more sensible? Without even needing to get into the jargon and methodology, what makes sense? He fucked up, or the whole field of criminology and everyone associated are wrong because nobody thought to use CDC data?

Think about it another way. Given most crimes are committed by a repeating minority, and your risk of initial crime is strongly correlated to age of first offense AND SES for that age, how could it make sense that age and SES are not the largest factors? Descriptively, factually, most of our criminals are repeat offenders who began at very young ages. His analysis cannot be true because his reasoning contradicts reality, even if you disagree for some reason about prediction. You don't even need to know anything about the particulars of methods and errors of inference involved to know, based on descriptive fact, his analysis can't make sense.

[Anonymous No.15442068]

>>15442036
Aren’t recidivism rates irrelevant to this discussion? You’re essentially saying that whites who commit crime are more likely to reoffend that blacks/Latinos who commit crime. But the question is to predict criminal behaviour in the first place, not rates of reoffending.

[Anonymous No.15442074]

>>15442036
>As for how that also answers his question directly, that is because most of the variance is due to repeat offenders in a kind of pareto principle of crime. The vast majority do not commit crimes, and most of those who do will have prior criminal history of some sort. For example, at the federal level depending on the year. This 2018 report, for example, 3/4ths had prior convictions. >https://www.ussc.gov/research/research-reports/criminal-history-federal-offenders there are a lot of reports available on the commission page for federal and they tend to repeat this theme.
Once again, you are deliberately obfuscating the issue, presumably so people here give up and don't bother arguing. Note what you said, that most offenders are repeat offenders. My original question was
>Care to cite me whatever that claims age and withtin-age economics explain most or all the differences?
What you cited was a paper that found race was still a statistically very significant (p<0.001) predictor of recommitting violent offenses, though with a small effect size, which was still larger than age, but much smaller than sex.
Then another paper on violent recidivism. In other words, among people who are already criminals the influence of race as an explanatory variable on repeat violent offenses was diminished though still present and very statistically significant. No fucking shit.
>>15442068
> You’re essentially saying that whites who commit crime are more likely to reoffend that blacks/Latinos who commit crime
No, it's the opposite. Though the effect size is small. Read my above response.

[Anonymous No.15442090]

>>15442058
I am having a hard time with you because the way you write makes it seem like you're trying to have a genuine intelligent discussion. And I am all for that; believe it or not I am very open to changing my mind in light of new evidence.
>1. Most crimes are committed by a small group of offenders
Correct. Most of any race are still law-abiding. But this still does vary greatly by race.
2. Odds of offense are overwhelmingly predicted by age and SES together
No. Your third paper just explained the between age variance, with the caveats that
>The “age–crime curve” does exist in two ways: (a) at every poverty level, murder rates drop off in the 50s and 60s, and (b) at poverty levels of 20% or higher, the traditional age–crime curve prevails

>3. Odds of recidivism are also overwhelmingly predicted by age and SES with virtually zero added predictive power due to race
Addressed here >>15442074
>This is demonstrable by simple fact predictions work best with combined age and SES whether recidivism or otherwise
care to show me the otherwise, while also showing no race influence?
As for the randomcriticalanalysis article, he has his comments open, and he responds. Why don't you argue with him about that? I will gladly read the debate.

[Anonymous No.15442091]

>>15442090
Forgot pic rel from the third paper.

[Anonymous No.15442099]

>>15442074
Sorry I got the direction of the effect wrong, but my question still stands. Isn’t recidivism an irrelevant topic of discussion since you’re already only dealing with known criminals? Shouldn’t we be asking about the general population of each group?

[Anonymous No.15442101]

>>15442074
>Once again, you are deliberately obfuscating the issue, presumably so people here give up and don't bother arguing.
Your failure to understand continuously repeated points like you've a 10 second memory is not my "obfuscating the issue".
>My original question was "Care to cite me whatever that claims age and withtin-age economics explain most or all the differences?"
>What you cited was a paper that found race was still a statistically very significant (p<0.001) predictor of recommitting violent offenses
Which I have subsequently explained in stages relevant to objections with still more research. That meta-analysis does not perform a multivariate regression or anything else, but rather correlates as to independent effect. Remember that third article you didn't read because "muh jews"? Remember what I wrote about it? >>15441959 "The relationship between age and crime as pointed out in that article is co-causal."

If you still aren't getting it, >>15441801 study 1 demonstrates the lack of effect as independent variables. Study 2 demonstrates for predicting recidivism race doesn't account for shit. Study 3 demonstrates the co-causal nature of age and SES, which would have to be the case per study 1 showing no real independent effect and study 2 showing how irrelevant race is in the prediction from age-SES. You claim I am obfuscating when I am simply explaining what you need to know to understand what the research means and why the actual academics don't agree with you on this.
>>15442090
>No. Your third paper just explained the between age variance
You're doing the 10-second memory thing again. Keep two things in your head at once.
1. Most crimes are due to a small number of people and repeat offenders >>15442036
2. Your odds of being a repeat offender depend on age-SES at first offense
Check that graph here >>15442091 and tell me what you see. Age-SES having the highest youth offense rates. Hence being the most predictive correlates.

[Anonymous No.15442107]

>>15442099
Right, that's more or less what I said.

[Anonymous No.15442108]

So poor young blacks with single mothers are more likely to be criminals?

[Anonymous No.15442114]

>>15442099
>Sorry I got the direction of the effect wrong, but my question still stands. Isn’t recidivism an irrelevant topic of discussion since you’re already only dealing with known criminals? Shouldn’t we be asking about the general population of each group?
I don't know if you just refuse to read my explanatory post or what. >>15442036
Most crimes are due to repeat offenders who become repeat offenders from a young age. Literally we are talking about the most relevant group. That IS the most relevant group. That is your largest share of resultant crimes per the citations in that second post talking about that pareto distribution.
>I am having a hard time with you because the way you write makes it seem like you're trying to have a genuine intelligent discussion.
I am. I am also being exceptionally patient about this, as evidenced by the fact I am bothering to try in the first place. There's a reason academics don't fucking bother and places like /sci/ become shitholes, and it is hardly due to mere elitism. You get more genuine scientific inquiry at a flat earther convention.

[Anonymous No.15442149]

>>15442114
>That IS the most relevant group. That is your largest share of resultant crimes per the citations in that second post talking about that pareto distribution.
What you are doing is akin to saying that once you control for IQ/SAT, the influence of race on school performance is minimal. Or perhaps more accurately, since we are dealing with an "elite" group in either case, SAT scores poorly predict grades once admitted into Caltech. No shit since everyone there has an SAT in the 1500-1600 range.

[Anonymous No.15442156]

>>15442108
>So poor young blacks with single mothers are more likely to be criminals?
That, as an aside, appears to have more to do with whether or not the father is in jail rather than independently single mothers if you want the largest associated variance with it. Also mundane very well known stuff per examples like https://link.springer.com/article/10.1007/s40865-
021-00183-7
>>In studying the relationship between family structure and delinquency, several factors need to be controlled for. The first factor is parental crime, because several studies find a positive correlation between parental crime and juvenile delinquency (e.g., Besemer et al., 2017). The second factor is the sex of the child, because boys have typically been responsible for the majority of youth crimes (Peterson et al., 2007; Messerschmidt, 2013). The third factor is the income of the household. There are contradictory theories and results regarding household income. General strain theory (Agnew, 2006) suggests that delinquent behavior is affected by the strain caused by a relative lack of resources, which is more common in single-parent households.

Curiously and contrary to many expectations children in this large sample with parental separation before 12, due to death or otherwise, did not have increased risk of delinquent behavior. That falsifies single mothers as the independent cause, or at least in this large study. All the results control for parental (mostly father) crime. The most consistent correlated independent factor appears to be single or dual parental criminality, though in various analyses it is not always the strongest.

Link split due to spam filter

[Anonymous No.15442161]

>>15442149
>What you are doing is akin to saying that once you control for IQ/SAT, the influence of race on school performance is minimal.
That only makes sense to say if you're saying the lions share of racial effects are attributable to demographic effects. Literally your attempt at a criticism presupposes agreeing with me as you've worded it. I have no idea what you think you're trying to say.
>Or perhaps more accurately, since we are dealing with an "elite" group in either case, SAT scores poorly predict grades once admitted into Caltech. No shit since everyone there has an SAT in the 1500-1600 range.
No. In this case the vast majority of people in that hypothetical world are going to caltech. As for crime the vast majority of crimes are due to repeat offenders. Get the issue with your analogy?

[Anonymous No.15442162]

>>15442101
Yeah, I am not misunderstanding anything. You are repeating the same things over and over again.
Address this
>Then another paper on violent recidivism. In other words, among people who are already criminals the influence of race as an explanatory variable on repeat violent offenses was diminished though still present and very statistically significant. No fucking shit.
and this rewording >>15442156

[Anonymous No.15442171]

>>15442161
>As for crime the vast majority of crimes are due to repeat offenders.
Which are a very small group as you said here>>15442058
>1. Most crimes are committed by a small group of offenders
>3. Odds of recidivism are also overwhelmingly predicted by age and SES with virtually zero added predictive power due to race
Hence the analogy with Caltech -- them being a very small group of students with the commonality of having been admitted into Caltech (analogous to having committed a crime).
>That only makes sense to say if you're saying the lions share of racial effects are attributable to demographic effects. Literally your attempt at a criticism presupposes agreeing with me as you've worded it. I have no idea what you think you're trying to say.
If one thinks IQ/SAT differences reflect underlying racial genetic differences etc. But you already knew that is what I meant.

[Anonymous No.15442173]

>>15442162
>Yeah, I am not misunderstanding anything. You are repeating the same things over and over again.
I am repeating it because you are clearly not understanding it. Else you would not mindlessly regurgitate that quote as if it's some "gotcha" for a fractional effect size.

What, exactly, do you think that gets you? I genuinely don't get what you think you're saying by it. You do realize you are effectively trying to cling to "but there's a 0.0026% variance still associated with whites not recidivizing!" right? Do you get how insane that is?

[Anonymous No.15442177]

>>15442173
Not the core of the issue with your argument (though you are forced to acknowledged a very statistically significant relationship exists). As this friend pointed out>>15442099

[Anonymous No.15442193]

>>15442177
>Not the core of the issue with your argument (though you are forced to acknowledged a very statistically significant relationship exists). As this friend pointed out>>15442099
I am not acknowledging it, I am pointing out how absurd your straw grasping is. Your failure to justify it, as I originally said, clearly shows your dishonesty.
>>15442171
>Hence the analogy with Caltech -- them being a very small group of students with the commonality of having been admitted into Caltech (analogous to having committed a crime).
You must be trolling. I cannot believe you genuinely failed to connect those very simple dots.
You wrote,
>Or perhaps more accurately, since we are dealing with an "elite" group in either case, SAT scores poorly predict grades once admitted into Caltech. No shit since everyone there has an SAT in the 1500-1600 range.
To which I explained,
"No. In this case the vast majority of people in that hypothetical world are going to caltech. As for crime the vast majority of crimes are due to repeat offenders. Get the issue with your analogy?"

It does not matter that the total "population" of repeat offenders is small. What matters is they account for most of the crime. What matters is race does not meaningfully predict nor explain this, and therefore fails to predict or explain the most significant contribution to crime. Namely, repeat offenders. You have to further believe, somehow, that race still predicts crime just only for non-recidivism based on... your ass I guess?

[Anonymous No.15442224]

>>15442193
Nope. Anyone with two brain cells here understands what I was saying. And so you do too. Once again willful obfuscation. Also you did not address this
>Then another paper on violent recidivism. In other words, among people who are already criminals the influence of race as an explanatory variable on repeat violent offenses was diminished though still present and very statistically significant. No fucking shit.
This is the crux of the argument you're making. It's worthless with respect to what is being argued here.
Anyone who reads this will understand.

[Anonymous No.15442238]

>>15442224
>Nope. Anyone with two brain cells here understands what I was saying. And so you do too. Once again willful obfuscation. Also you did not address this
I'm genuinely unsure whether you really are trolling or you really can't think with proportions.
>This is the crux of the argument you're making. It's worthless with respect to what is being argued here.
Most crimes are due to recidivism. If recidivism does not have a racial component as I've clearly demonstrated, then the majority of crime cannot be causally attributed to race. This is very simple If A, then B. A, therefore B. You tried rejecting my premise and failed, and you tried rejecting my conclusion and failed. What you're left with tantruming.

[Anonymous No.15442273]

>>15442238
>If recidivism does not have a racial component as I've clearly demonstrated, then the majority of crime cannot be causally attributed to race.
lol

[Anonymous No.15442549]

>>15442114
>Most crimes are due to repeat offenders who become repeat offenders from a young age. Literally we are talking about the most relevant group. That IS the most relevant group. That is your largest share of resultant crimes per the citations in that second post talking about that pareto distribution.

Ok fine but if you’re trying to predict who is a criminal, you can’t start by controlling for whether they’re a criminal.

You’re only looking at the population of criminals to decide who is going to commit a crime. And then you find that within that group, race has very little effect. That may be true, but it isn’t helpful.

How about looking at the general population? How to things like race and SES affect crime rates then? I genuinely don’t know but my guess is that it doesn’t fit your biases because you’re ignoring it.

[Anonymous No.15442561]

>>15442238
This is like saying the majority of kids who got As on the midterm got As on the final. When we try to predict the grades on the final, and only look at kids who got As on the midterm, we find that race has no effect.

Ok but how about you look at all the kids, and find that race had a huge impact on who got As on the midterm so it’s not a fair variable to control for?

And yes, this is true even though the majority of As on the final are caused by kids who got As on the midterm.

[Anonymous No.15442595]

>>15442561
To make this even more concrete, let’s say you have 100 people who come to a cafeteria every day. 50 Latinos and 50 whites. They can order burgers or burritos. On day 1: 40 Latinos and 20 whites order burritos, the rest order burgers.

You know that the majority of burrito orders tomorrow will be from people who ordered burritos today, so you decide that “previous burrito orderers” is the only sample that matters. Of that sample, 30 Latinos and 15 whites order burritos the next day. All of a sudden race has no effect on who orders a burrito, because it has no effect on “burrito recidivism rates”

[Anonymous No.15442968]

bump

[Anonymous No.15443018]

bump

[Anonymous No.15443040]

>>15442238
>Most crimes are due to recidivism. If recidivism does not have a racial component as I've clearly demonstrated, then the majority of crime cannot be causally attributed to race. This is very simple If A, then B. A, therefore B. You tried rejecting my premise and failed, and you tried rejecting my conclusion and failed. What you're left with tantruming.
Imagine how stupid you have to be to believe this

[Anonymous No.15443057]

>>15442561
>This is like saying the majority of kids who got As on the midterm got As on the final. When we try to predict the grades on the final, and only look at kids who got As on the midterm, we find that race has no effect.
Nope. It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who started younger and poorer are far and above the most likely to continue getting A's. Or, in this case, commit crimes.
>>15442549
>How about looking at the general population? How to things like race and SES affect crime rates then? I genuinely don’t know but my guess is that it doesn’t fit your biases because you’re ignoring it.
Ignoring it? That's a funny way to say "addressed it from the very start". 10 second memory again? >>15441801

But as you've continued to project, your biases ensure you keep ignoring it. I guess because it mentioned a jew once and that triggered you. Either way for significant causal factors of juvenile crime, that significant predictor of later recidivism, >>15442156

I've spoon fed you multiple times with a variety of studies, in each and every case needing to disentangle your kneejerk attempts to save your bias. These associations occur independent race, independent nation, and as further proof of your bias and unwillingness to learn you didn't even check between articles for nationality. For that matter, you didn't read any of the article citations where many of them talk about race or ethnicity or minority status and cite other literature answering your question. No amount of data as to the age-SES effect will persuade you, you will just endlessly reach for more ad hoc excuses. What stops you further claiming "Oh but muh IQ muh income still race causing it"? Nevermind GWAS associations find no such significant effect sizes for what few functional variants are common between populations either. Once my trip was over I was hardly going to waste free time.

[Anonymous No.15443060]

>>15443057
Correction: Obviously I meant for variants common within but not between populations.

[Anonymous No.15443072]

>>15443057
Answer this one >>15442595
You won't because it shows how ridiculous your argument is. You can try to hide behind statistics and big words all you want but your argument is terrible and everyone here can see that.

[Anonymous No.15443079]

>>15443072
>You won't because it shows how ridiculous your argument is. You can try to hide behind statistics and big words all you want but your argument is terrible and everyone here can see that.
I did. The "more concrete" continuation of the dishonesty is the same false analogy as already responded to >>15443057
>Nope. It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who started younger and poorer are far and above the most likely to continue getting A's. Or, in this case, commit crimes.

[Anonymous No.15443080]

>>15443057
>It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who started younger and poorer are far and above the most likely to continue getting A's. Or, in this case, commit crimes.

Ok, what if you have a class of 100 kids, 50 white and 50 asian. And 20 of the asian kids have gotten As before while 10 of the white ones have.

If I now only look at those kids to see who gets an A on a particular test, and find that 16 of the asian kids do, and 8 of the white kids do, have I ruled out any racial connection to grades? Because that's what your argument says, and you're not fooling anyone. I don't even think you believe it yourself, but who knows.

[Anonymous No.15443084]

>>15443079
So do you agree that there is no racial connection to lunch choice based on your criteria? Or disagree?

[Anonymous No.15443093]

>>15443080
>>15443084
I'm not wasting my time endlessly refuting dishonest strawmen when I already demonstrated to any reasonable persons satisfaction your unwillingness to admit to the dishonesty. It is sufficient to show the sort of inability, or deliberate unwillingness, you have to engage in either reasoning or earnest reaidng as I already have done.
>>15443057 >>15442561
>This is like saying the majority of kids who got As on the midterm got As on the final. When we try to predict the grades on the final, and only look at kids who got As on the midterm, we find that race has no effect.
"Nope. It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who started younger and poorer are far and above the most likely to continue getting A's. Or, in this case, commit crimes."

All your subsequent "challenges" are in the same vein as this dishonesty. Like I already wrote here >>15443057, adding yet more ad hoc excuses because nothing can ever satisfy you when you're not honest in the first place.

[Anonymous No.15443109]

>>15443093
Most people who got As were Asian
> No you idiot, most people who got As had gotten As in the past
Ok but most people who had gotten As in the past were Asian

In other words you're saying that most people who commit crimes are criminals. No shit, retard.

If you're trolling then good job, I fell for it, but if you're serious, then you're the embodiment of the overeducated midwit. You're trying to hide your terrible understanding behind big words.

[Anonymous No.15443110]

>>15443084
>>15443093
And in case you really are 10 second man, the predictor for initiation of crime is also the same age-SES factor with many aggravating aspects discussed all through literature, cited before >>15442156 as well as mentioned for example before >>15442036 like psychiatric conditions going untreated.

Altogether GWAS refutes your faith too. >>15443057 The actual mechanisms existing across nations, cultures, and races, also refute your faith. Recidivism prediction refutes your faith. All you can do is selectively ignore posts, forget points, falsely strawman arguments, and tapdance around it all like every other science denying radical. All while demonstrating a total refusal to honestly pursue answers to your doubts or questions and using them as grounds for dismissal of further study instead. That is not what an honest person does, that is what an ideologue does when he knows he can't meaningfully defend what he wants to lie about.

All while, and this is was the funny part to me motivating me to continue, I remain the sole person actually citing or discussing any of the literature and relevant science. For which I am accused of "using big words" and "obfuscating". No, kid, you're just an angry ignorant little flat earther who doesn't want to just honestly ask about things he finds confusing and thinks that tricks anyone.
>In other words you're saying that most people who commit crimes are criminals. No shit, retard.
Nope. >>15443093
>"It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who started younger and poorer are far and above the most likely to continue getting A's [and start getting them]. Or, in this case, commit crimes."
Added the obvious in brackets since you forget everything 10 seconds later.

[Anonymous No.15443119]

>>15443109
>but if you're serious, then you're the embodiment of the overeducated midwit. You're trying to hide your terrible understanding behind big words.
"Earth is flat! Stop using math! Math is big words! Overeducated midwit! Angles don't prove anything! Stop hiding!" lol I'm going to finish enjoying my drink kid. I'll let you know when anyone who does any real work or research cares about your whining.

[Anonymous No.15443121]

>>15443110
I'm (i.e. >>15441737 >>15441920 etc. ending with>>15442273) not responding to you anymore. You're arguing someone else. I feel sufficiently like I trounced you. Addressed almost all your points, line by line, link by link. You now accuse me of dishonesty which I find hilarious.
Have a good one. I will bookmark this thread and bring it up if I ever recognize you again in a thread, as you have a very particular tone and way of writing that is easy to distinguish.

[Anonymous No.15443123]

>>15443121
>Have a good one. I will bookmark this thread and bring it up if I ever recognize you again in a thread, as you have a very particular tone and way of writing that is easy to distinguish.
lmao the fucking narcissism. Please do kid I'm soooo scared you'll bring up your cringey failure to understand anything.

[Anonymous No.15443125]

>>15443110
But you're only looking at people who have already committed crimes. Do you not see how that is throwing your sample off terribly?

What if 90% of people who have committed a crime before have blue eyes, but once you've committed a crime, your odds of doing it again are not affected by whether you have blue eyes. Do you seriously think that means that blue eyes have no correlation to crime?

Please actually refute this example instead of just namecalling. Maybe I really am too stupid to get it, but the fact that you keep dodging my examples makes me think you know you're wrong and just can't admit it.

[Anonymous No.15443126]

>>15443123
That's pretty aggressive of you, calm down. You have a biology degree or something? I have a math BS and finishing my math/stats MS. Cheers.

[Anonymous No.15443129]

>>15443119
I think you're a little mad that your model can't even handle simple toy examples.

[Anonymous No.15443131]

>>15443125
>But you're only looking at people who have already committed crimes. Do you not see how that is throwing your sample off terribly?
This lie has already been refuted >>15443057
>>15443093
>"Nope. It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who [are younger and poorer have the most first-A's] and far and above the most likely to continue getting A's. Or, in this case, commit crimes."
My drinks out and you're boring. Off to better things.

[Anonymous No.15443135]

>>15443131
>"Nope. It's like saying of those kids who get As most, almost all, got As before. When you want to predict who is going to start getting A's, those who [are younger and poorer have the most first-A's] and far and above the most likely to continue getting A's. Or, in this case, commit crimes."

You're saying of all the people who commit crimes, almost all have committed crimes before. I don't disagree. But if 90% of those people share a certain trait then that trait obviously correlates to committing crimes.

Jesus no wonder you're in bio, you must have gotten filtered by basic math early on.

[Anonymous No.15443294]

>>15443123
You're so pathetic lmao