Talk:Intelligence quotient

Summary of race relationship

Proposed summary sentence for the paragraph on race:
The scholarly consensus finds environmental factors, not race, determines IQ scores. Superb Owl (talk) 18:14, 12 October 2024 (UTC)

I appreciate you bringing your suggestion here. The issue with this sentence is that IQ differences between individuals do indeed seem to be highly heritable, i.e. genetics plays a role in accounting for why person A scores higher than person B. Some folks have naively assumed that this fact redounds to group-level differences, but it does not. This mistake has been called the "hereditarian fallacy".

If you'd like to explore a source, see e.g. "Heritability in the genomics era — concepts and misconceptions" in Nature Reviews Genetics, which lists in its table of Misconceptions regarding heritability the notion that Heritability is informative about the nature of between-group differences.

As they explain:

This misconception comes in two forms, and in both cases height and IQ in human populations are good examples. The first misconception is that when the heritability is high, groups that differ greatly in the mean of the trait in question must do so because of genetic differences. The second misconception is that the observation of a shift in the mean of a character over time (when we can discount changes in gene frequencies) for a trait with high heritability is a paradox. For IQ, a large increase in the mean has been observed in numerous populations, and this phenomenon is called the Flynn effect, after its discoverer. The problem with this suggested paradox is that heritability should not be used to make predictions about mean changes in the population over time or about differences between groups, because in each individual calculation the heritability is defined for a particular population and says nothing about environments in other populations. White males born in the United States were the tallest in the world in the mid-19th century and about 9 cm taller than Dutch males. At the end of the 20th century, although the height of males in the United States had increased, many European countries had overtaken them and Dutch males are now approximately 5 cm taller than white US males, a trend that is likely to be environmental rather than genetic in origin.

I hope that's helpful. As I stated in my edit summary, I'm all for improving the language if anything is unclear or wonky, but a lot of this stuff is the result of painstaking consensus-building across a number of related articles in the R&I topic area, so changing anything substantially is going to require some workshopping. Cheers, Generalrelative (talk) 22:30, 12 October 2024 (UTC)

I appreciate the detailed answer. How about:
"There are no biological differences based on race." Superb Owl (talk) 22:51, 12 October 2024 (UTC)

I hate to be a Debbie Downer but I also find that sentence confusing. Most scientists prefer the formulation "race is a social construct", which is what we already state at the beginning of the subsection.

Maybe it would help if you explain what you find disagreeable about the current section intro?

Among the most controversial issues related to the study of intelligence is the observation that IQ scores vary on average between ethnic and racial groups, though these differences have fluctuated and in many cases steadily decreased over time. While there is little scholarly debate about the continued existence of some of these differences, the current scientific consensus is that they stem from environmental rather than genetic causes.

Thanks, Generalrelative (talk) 23:41, 12 October 2024 (UTC)

The intro does not seem clear as it is a bit jargony as written. I think a thesis sentence could help it out a lot:
"Scientists do not believe that IQ differerences are influenced by one's race." Superb Owl (talk) 01:15, 13 October 2024 (UTC)

I don't feel particularly qualified to speak on this subject, but your proposed sentence sounds off. At least in the US, race does entail environmental differences, even if they are not genetic. CAVincent (talk) 01:29, 13 October 2024 (UTC)

I should maybe add that these environmental differences are themselves heritable, again even though they are not genetic. CAVincent (talk) 01:36, 13 October 2024 (UTC)

I suppose I should be clear for the talk history that I struck my own comments. I noted my own lack of qualification, and was graciously corrected on my misunderstanding. CAVincent (talk) 12:55, 13 October 2024 (UTC)

In vernacular usage "heritable" is sometimes used to mean anything that's often passed on from parents to children, such as wealth, a history of domestic abuse, poverty, obesity. But in the context of this article -- that is, in scholarly usage -- it does mean genetic. NightHeron (talk) 07:31, 13 October 2024 (UTC)

There is a lot of cherry-picking of sources that's gone on with respect to this sentence, and similar sentences in other articles. Of the sources for the current sentence, only one uses the word "consensus" or any similar term as required by WP:RS/AC. While it's a high-quality source, it also is fifteen years old. Of the other two sources one is from twelve years ago, and the other contains only a single sentence about group differences in IQ. Statements like this one appear in nearly every Wikipedia article related to intelligence, and the sourcing is mostly of a similar level of quality. See my earlier comment about a similar statement in another article:

All of these sources, even the Guardian article and the VOX blog post, theoretically satisfy the requirements of WP:RS and can be cited. But in a Wikipedia article about virtually any other topic, sources like these would not be used to make a statement about what academic "consensus" is, especially when more recent and higher-quality secondary sources are available, such as those linked to in the last comment here and the off-Wiki discussion cited there. The most recent source this article cites about academic "consensus" is from 2017, but the off-wiki list includes ten secondary sources about this topic that were published more recently, and nine out of ten of those newer sources (all except Harden's book) present a very different view.

Various people have been objecting for years that these statements about academic "consensus" are based on a very small number of sources which were selected to support a specific viewpoint, and past such objections have usually been dealt with by piling on yet more of these old or mediocre-quality sources (the equivalent statement in the race and intelligence article now cites nine of them, including one that was published more than half a century ago). But maybe now we finally have the critical mass of editors needed to change this trend.

CAVincent: you seem to be new to this topic, so I'd like to hear your opinion about whether we should make use of some of the newer sources I mentioned, and modify these parts of the article(s) accordingly. 64.127.212.41 (talk) 11:30, 13 October 2024 (UTC) Striking WP:BLOCKEVASION. Generalrelative (talk) 22:20, 20 October 2024 (UTC)

There is zero ambiguity about the scientific consensus in top-quality scholarly sources, e.g. :

Recent articles claim that the folk categories of race are genetically meaningful divisions, and that evolved genetic differences among races and nations are important for explaining immutable differences in cognitive ability, educational attainment, crime, sexual behavior, and wealth; all claims that are opposed by a strong scientific consensus to the contrary.

The consensus among Wikipedia editors is clear as well. It will not be relitigated here. Generalrelative (talk) 15:25, 13 October 2024 (UTC)

Superb Owl: Perhaps the IP's comment offers some perspective as to why we use "scientific" language here. Scientists speak with great precision for a reason. This topic area is a case where white supremacists are highly motivated to misrepresent what the science says, which is why we need to be extra vigilant to say things precisely as the best sources do. If you'd like to read more about this context, I suggest this recent article in The Atlantic: "The Far Right Is Becoming Obsessed With Race and IQ". Generalrelative (talk) 15:29, 13 October 2024 (UTC)

Thanks for sharing the research, the WP:RS article and your perspective - I still believe that it would be very helpful to include a clear succinct summary of the scientific consensus in the first sentence. Paraphrasing that quote you added:
According to a strong consensus of scientists, there are no genetically meaningful differences between racial groups related to IQ or intelligence. Superb Owl (talk) 15:40, 13 October 2024 (UTC)

I like this suggestion a lot, but it's still not 100% there in my opinion. How about we just import the sentence from the first paragraph of Race and intelligence:

Today, the scientific consensus is that genetics does not explain differences in IQ test performance between groups, and that observed differences are environmental in origin.

We could also add this other sentence from the lead:

In recent decades, as understanding of human genetics has advanced, claims of inherent differences in intelligence between races have been broadly rejected by scientists on both theoretical and empirical grounds.

The current sources, along with Bird et al. easily support this. How does that sound? Generalrelative (talk) 16:20, 13 October 2024 (UTC)

I still think your language is a bit too science-y but we are getting much closer. Curious to hear what others think/propose. I also would prefer the paragraph giving background to come after the main paragraph establishing that there is no relationship which seems to be a better summary paragraph of the section. Superb Owl (talk) 17:51, 13 October 2024 (UTC)

That's fair. Let's leave the question open for the time being and see if other editors care to weigh in. Generalrelative (talk) 18:43, 13 October 2024 (UTC)

Various people have been objecting

Your use of WP:WEASEL WORDS makes it unclear who these various people are, but if you mean Wikipedia editors, this topic has been litigated and re-litigated repeatedly and the consensus of Wikipedia editors on this subject has been made repeatedly quite clear. This comment boils down to WP:IDIDNTHEARTHAT, which is doubly apparent in your claim that any and all new sources added as a result of these litigations are mediocre, while not even pretending to acknowledge the criticisms of the quality of the sources you feel contradict the current Wikipedia consensus.

If, on the other hand, you mean Justapedia editors, well, there's a reason they don't comment over here much, except periodically from IPs in futile attempts to relitigate long settled issues.

As for your link to that Justapedia thread, I'm sorry, but a thread on a far-right Wikipedia clone is not a reliable source, and I don't just mean this in the sense of the term used on WP:RS, but also in the conventional definition of the term "reliable". Since anyone purporting to calculate the relative percentage of major secondary sources that agree vs. disagree with Murray must necessarily make some decisions about which sources to look at, which qualify as "major" (the terminology of the post), and which qualify as neutral, for, or against, the methodology by anyone from that site is inherently suspect because of the known biases of the vast majority of Justapedia's editors and articles. But I am not merely casting vague aspersions here; the immediate effects of this bias are apparent in their selection criteria and their justifications, enumerated in three items which I will likewise address point by point:

1. They begin by limiting their "survey" to just two fields known to be a magnet for white supremacists and their fellow travellers.
2. They then proceed to summarily disregard the views of those that agree with the stated consensus of the American Anthropological Association that "race" is an arbitrarily defined social category (the author's words, not the AAA's), since, according to them, this means such researchers have nothing much of substance to say about the empirical research on this topic. How convenient that the people most likely to disagree with you are immediately judged to be irrelevant. They further justify this by citing WP:MNA and the corresponding Justapedia policy—apparently, a policy designed to encourage one to make necessary assumptions largely tangential to the topic at hand in the writing of an encyclopedia article is justification for assuming as true the most basic sina qua non premises and conclusions of the supporters of a hypothesis when conducting a literature survey on the relative scientific support vs. criticism of that hypothesis. Imagine your reaction if we applied this principle in reverse while conducting the same sort of literature survey: we disregard any scientist and their writings who assume that race has biological substance on the grounds that "we must make necessary assumptions".
3. Finally, in a marked demonstration of their complete lack of self awareness, they actually claim that the use of the stated selection criteria is in order to avoid the sort of cherry-picking of sources about this topic that's occurred at Wikipedia. Because immediately ruling out anyone who believes in a position fundamentally at odds with the one you are trying to advocate for is definitely not cherry picking.

Brusquedandelion (talk) 10:06, 19 October 2024 (UTC)

Just FYI Brusquedandelion, the IP you're responding to was blocked as a sock. I didn't hat the comment since it had been referenced in my exchange with Superb Owl, but at this point their comment above should probably be struck so no one else wastes time on them. Generalrelative (talk) 13:41, 19 October 2024 (UTC)

Update: I went ahead and struck it. Generalrelative (talk) 22:21, 20 October 2024 (UTC)

We don't need to hedge our sentences on Wikipedia with "scientists believe". Scientific consensus can be written in WP:WIKIVOICE, appropriately cited of course. Brusquedandelion (talk) 04:16, 16 October 2024 (UTC)

This sentence:

The issue with this sentence is that IQ differences between individuals do indeed seem to be highly heritable

But the proposed sentence says nothing about heritability, only that environmental factors, not race, determines IQ scores. Race is not the same thing as genetic inheritability. Brusquedandelion (talk) 04:13, 16 October 2024 (UTC)

My objection to the sentence in question is not that it's wrong but rather that it might be confusing to the reader, because it seems to imply that environmental factors are the only determinants of individual differences in IQ test performance, which most scientists agree is not the case.

On the other hand, environmental factors do appear to be the only determinants of differences in average IQ test performance between population groups. And we should indeed state that emphatically. Generalrelative (talk) 07:58, 16 October 2024 (UTC)

This is not true. A scientific discussion about evidence is not a discussion of politics or social policy. The safe and obvious assumption is that population differences are a result of genetics and environment. There are several reasons this is clear to anyone with sufficient verbal abstraction ability:

intelligence is highly, highly polygenetic. Groups separated for tens of thousands of years are not going to have the exact same variance of genes. Can you name me a biological trait that is only environmentally determined?

just because something is true, it doesn’t mean you have to say it. But the ethical answer never is too tell a lie when the truth is expected. RationalFactor (talk) 19:54, 19 October 2024 (UTC)

Hi RationalFactor,

You've done a great job illustrating why we don't rely on the original analysis of our volunteers, especially not about what it is "safe and obvious" to assume. Instead, we follow what the reliable sources say, and this has already been adjudicated. If you truly have good-faith questions on the matter, I will direct you to this handy explainer.

Cheers, Generalrelative (talk) 20:57, 19 October 2024 (UTC)

There is no evidence to suggest it’s only environmental, except for some falsified data from Gould a long time ago. There are only arguments made by some scientists, but they do not pass scientific scrutiny.

After some minimal comprehension of biology and statistics, it’s clear that such a claim would not even be considered were it about any other topic, like sprinting or height or propensity to heart disease or hairiness or hormone production or ability to digest lactose or alcohol response or immunology or propensity to certain cancers or heat tolerance.

No environmental change outside the extremes of neglect, poverty, iodine deficiency, injury, or poisoning in a developed nation has ever been found to even influence IQ scores. RationalFactor (talk) 20:34, 20 October 2024 (UTC)

Importantly, this was the consensus. However, since these are all university jobs, it couldn’t possibly be the consensus today.

I think informed people who are open minded and good faith could in fact be confused, because in the past people hoped IQ was only 50% genetic, and in nearly any university this is presumably still the number given as fact. It might even be given in this article. If it were true, there would be room to say only environment; HOWEVER, even then, it’s not true science to now stray from the null: mix of environmental and genetic. RationalFactor (talk) 20:42, 20 October 2024 (UTC)

My grandfather used to say: It ain't the stuff you don't know that gets you into trouble. It's the stuff you're damn sure of that just ain't so.

This is a case in point. I explained the hereditarian fallacy above, and quoted a gold-standard source saying that yes, serious genetics professionals understand this to be a fallacy, and yet you're repeating it. To reiterate: the heritability of individual-level differences tells us nothing about group-level differences. If you think this is all based on Gould, you have clearly not looked at the literature that was so thoroughly discussed in the RfC.

I'm sorry to say, but at this point there's nothing more to be said. If you have source-based suggestions for improving the article, you are free to suggest them. But this talk page is not a place for complaining about the WP:THETRUTH. Generalrelative (talk) 22:12, 20 October 2024 (UTC)

Don't feed the trolls. The following discussion has been closed. Please do not modify it.

Many sources expressing agnosticism or hereditarian views were provided below. But you dismiss these in favor of your favorite blog post. Tarantaloid (talk) 08:36, 21 October 2024 (UTC)

Account with 1 edit, created today, after the IP sock was discovered and banned. @Generalrelative, does this warrant admin investigation? Brusquedandelion (talk) 08:38, 21 October 2024 (UTC)

Not worth wasting admin time unless it continues. Just expand the hat to cover the troll. Generalrelative (talk) 16:47, 21 October 2024 (UTC)

Done. Generalrelative (talk) 16:48, 21 October 2024 (UTC)

Generalrelative, that source you're linking to was published in American Psychologist so it's of good quality. But of its authors, one (Kevin Bird) is a botanical geneticist, one (John P. Jackson) is a professor of communication studies and Systems Training for Emotional Predictability and Problem Solving (STEPPS), and one (Andrew S. Winston) is a historian of psychology. In other discussions you've linked to this source which has the opposite issue: it is written by experts in human genetics but is a blog post.

For statements about scientific consensus, the best possible sources are those from prominent journals and academic publishers that are written by experts in human genetics or aspects of psychology that relate directly to intelligence. And we actually do have such recent sources available, such as these four: The Harden source more or less agrees with the current article, but the other three don't. Can you explain why, in terms of policy, we would privilege academic sources written by people whose expertise lies outside this field over those written by subject-matter experts? 64.127.212.41 (talk) 18:24, 13 October 2024 (UTC)

I think we should stick with peer-reviewed scholarship, not books (especially not without quotes and page numbers backing up these extraordinary claims), for this topic Superb Owl (talk) 04:33, 14 October 2024 (UTC)

There are quotes and page numbers for the relevant parts of all these sources in the off-wiki discussion linked to in my earlier comment. I also don't see a reason for regarding books from reputable academic publishers as either more or less reliable than reputable academic journals, as academic presses' editorial boards use a review process that's very similar to peer review, and the relevant sourcing guideline WP:SCHOLARSHIP makes no distinction between the two. 64.127.212.41 (talk) 05:09, 14 October 2024 (UTC)

64.127 – Likely block evasion from Captain Occam and stop linking to Justapedia an unreliable source. The only reason you are linking to that is because you have posted on that very page you are linking to. Most of the sources you are citing over there are not reliable: Russell T. Warne, Heiner Rindermann, Richard Haier all hold far-right political views and have appeared on alt-right podcasts. Academia has ignored their publications. Definitely not WP:SCHOLARSHIP. Most of what you are citing is WP:Fringe. Psychologist Guy (talk) 10:17, 14 October 2024 (UTC)

Rindermann has a poor academic record. Here he is only last month, a paper of his retracted by the editor in chief, "The Editor-in-Chief has retracted this article. After publication, concerns were raised about the methodology and dataset used in this research. Independent post-publication peer review has confirmed fundamental flaws in the use of student assessment studies as a measure of IQ or cognitive ability, and in the prominence of individual examples taken from the author's life". . Rindermann is the same person who spends his time attending far-right conferences making anti-immigration talks several of which can be found on YouTube. Obviously not a neutral or reliable source for this topic area. Psychologist Guy (talk) 10:24, 14 October 2024 (UTC)

Well, so much for you having "no interest in race and intelligence". But it's valuable that you're citing those RationalWiki pages, because that provides some useful information about whether the Norwegian IP user was correct in his suspicion about you. There is only one person who cites RationalWiki articles to support his arguments here, although I respect your patience in building up a convincing contribution history in other topics before returning to that behavior. You apparently (incorrectly) think I'm someone evading a block, and with your recent comments you've made me suspect that you're evading one also, so there is no meaningful discussion we can have here. All we can do is wait for other editors to comment, unless this discussion has already become too sidetracked to go anywhere. 64.127.212.41 (talk) 11:38, 14 October 2024 (UTC)

Avoiding the valid points I made. Why are you citing far-right extremists? They fail WP:RS And no I am not particularly interested in this subject I have rarely commented on it, but I have been attacked off-site by yourself and your friends so sometimes I will take exception. Several users have agreed you are a block evader and it is obvious, you have been using IPs on here for months to evade your block. And lol at the conspiracy theory claims I am evading a block myself from your far-right group of friends. I have created 100s of articles here and improved 1000s of articles. BTW I am an admin at RationalWiki, I think I have created 200+ articles there as well. Who are you again and what have you ever done for this project?

Based on your Captain Occam account you had no constructive edits but just trolled talk-pages on race and intelligence. It seems you have been at this for decades. It seems your MO is to stir up drama on race articles. Pathetic. I won't waste time responding to you again. See you in a few weeks I guess on a new IP like you have been doing for years. Psychologist Guy (talk) 12:28, 14 October 2024 (UTC)

WP:AGF 24.126.11.219 (talk) 14:10, 14 October 2024 (UTC)

References

Intro paragraph

The concluding sentence, “Many of the proponents of intelligence tests and IQ scores were eugenicists who used pseudoscience to push now-debunked views of racial hierarchy” is inappropriate, given that it’s irrelevant to a discussion of empirical evidence. Furthermore, hierarchy has nothing to do with this conversation. IQ has been the most replicable aspect of all of psychology, and it is summarized as racist in the introductory paragraph? That’s advocacy, not science. RationalFactor (talk) 03:04, 17 October 2024 (UTC)

This is historical context, and historical context is relevant to an encyclopedia, provided it is well-sourced. Generalrelative (talk) 20:59, 19 October 2024 (UTC)

Brother. We don’t describe physics as Christian because Christians invented it. There was nobody else around. That is even assuming that the racism claim is even valid. That is your political spin on discussions of data. Historical context is necessary, but not political anti-science. There was just another study that found IQ to be 80% heritable. The spin is not appropriate. RationalFactor (talk) 06:31, 29 January 2025 (UTC)

Sister. Wikipedia presents what the reliable sources say in proportion to due weight, and the lead summarizes the article body. You'll get nowhere arguing we should go against policy by characterizing verifiable information you don't like as "political spin" or "anti-science". No one is fooled by this. Generalrelative (talk) 22:37, 29 January 2025 (UTC)

What is this study you're referring to? Harryhenry1 (talk) 08:25, 29 January 2025 (UTC)

Harryhenry1: That's called a red herring. Generalrelative (talk) 22:37, 29 January 2025 (UTC)

I've linked to IQ classfications in the headers

I'm curious what the opinion is on this. I'm new to wikipedia, and this is my first edit, just wondering if the link I put is alinging with the rules. Addi-bot (talk) 16:13, 6 January 2025 (UTC)

Average IQ by country (2025 update)

Reliability of this ranking: All participants completed the same IQ test on this website in 2024. In 83.18% of the countries, the IQ scores are similar to those from the previous year (< 2 points difference) Amane Ojakasa (talk) 12:47, 11 January 2025 (UTC)

It's unreliable because deriving the average IQ scores of nations through a website where the participants have to take an online test suffers from self selection bias (Only rich people can do the test if the country at question is poor to the point where internet and computers are too expensive. Also, people who are interested in puzzles do online tests, which is another bias) along with the fact that most online tests are not as accurate as a real one. A more reliable, though perhaps outdated source would be Richard Lynn's "The Intelligence of Nations". The figures cited for the sub-saharan african nations are generally considered unreliable as some samples consisted of malnutritioned or otherwise mentally disabled men and children.

tl;dr: No genuinely reliable study assessing the IQ's worldwide exist as of now. Skellyret (talk) 18:36, 20 April 2025 (UTC)

Race

DaltonCastle has twice attempted to remove longstanding language stating that race is a social construct. This is unquestionably the scientific consensus, as has been repeatedly affirmed over at Race (human categorization) and in top-tier reliable sources like this Consensus Study Report from the National Academies of Sciences, Engineering and Medicine:

In humans, race is a socially constructed designation, a misleading and harmful surrogate for population genetic differences, and has a long history of being incorrectly identified as the major genetic reason for phenotypic differences between groups.

DaltonCastle's rationale for the removal is that this is only a consensus among SOME biologists but this is misleading; it is a scientific consensus even if some WP:FRINGE voices say otherwise. In any case, though this is not the place to relitigate the matter of whether race is a social construct, it is the place to more thoroughly substantiate the rationale for my reverts, and potentially discuss the way we phrase the point in the present article. Generalrelative (talk) 01:38, 23 March 2025 (UTC)

Validity

Generalrelative Hi I was just wondering why you removed my edit because I don't think it repeated any previous points. The section 'reliability and validity' gives good information on the reliability of IQ but fails to give the same treatment on its validity, instead mentioning the beliefs of people who are arguably in the minority (Stephen Gould). I think the edit is important because it both provides a more neutral point of view to the section and also provides accurate information on the validity of IQ. Skellyret (talk) 18:11, 20 April 2025 (UTC)

I suggest you read the section again. Gould is far from the only source, and the sources we do mention represent the mainstream view. See e.g. the mainstream textbook we quote by Wayne Weiten: "IQ tests are valid measures of the kind of intelligence necessary to do well in academic work. But if the purpose is to assess intelligence in a broader sense, the validity of IQ tests is questionable." That's what we say because it's a great distillation of the mainstream view, i.e. WP:NPOV. We do discuss Pearson and g in a separate section, and various positive correlates with IQ, such as education attainment, elsewhere. Further, your own sources don't really support your text. There's a large element of WP:SYNTH going on in that edit. Generalrelative (talk) 18:16, 20 April 2025 (UTC)

I think the quote is rather vague (doesn't define "assess intelligence in a broader sense" what does that mean exactly?) and gives the implication that IQ measures how book smart you are instead of your cognitive functions. Skellyret (talk) 18:22, 20 April 2025 (UTC)

IQ is a tricky subject, but that's precisely why we rely on reliable sources like Weiten's textbook to figure out the best way to phrase things. But yeah, he's essentially in agreement with the Yale business school source you wanted to cite: "Why a high IQ doesn't mean you're smart" . Regarding "book smarts" versus cognitive function, this is far from settled science. No one seriously doubts that an average primitive hunter-gatherer would score poorly on even the most "culture fair" IQ test, and no one seriously doubts that they would at the same time massively outperform an average modern city-dweller at the sorts of cognitive tasks required to survive without modern technology. In general, IQ is an important metric, especially because of its high reliability, but its content validity is still an open question. Generalrelative (talk) 18:38, 20 April 2025 (UTC)

I would agree with you however I still do think the section is giving IQ and IQ testing the short end of the stick. I doubt that primitive hunters would score equal to us even on Raven's matrice IQ tests, but that's because education and proper nutrition stimulates your intelligence which is why IQ is 30 - 50% environmental. When people train at a task, even if it engages with your cognition (like maths or the insight that primitive hunters needed i guess), your practice will eventually make you better at that specific task. This is why we renew IQ tests and test methods every decade, so we can minimize the effect that studying and schooling has on how well you can do at IQ tests. Take the inuits for example.

I think the fact that IQ has a good predictive power to various other fields (educational attainment, job performance, income, occupational attainment) is a testament to how IQ measures more than just how academically good you are. Skellyret (talk) 18:50, 20 April 2025 (UTC)

It's absolutely fair for you to have that opinion, but note how we state that IQ correlates with those other metrics (educational attainment, job performance, etc.) rather than causes them. That's because we follow the science, which is quite circumspect on the matter. Generalrelative (talk) 18:55, 20 April 2025 (UTC)

Yeah but it's your opinion that a single vague quote that somewhat supports IQ's validity is enough for the wiki article lol. Skellyret (talk) 19:18, 20 April 2025 (UTC)

The section we're discussing has been largely stable for years, and this is a page with 1200 watchers. I don't recall whether our use of Weiten has been questioned explicitly before, but at this point it enjoys WP:IMPLICITCONSENSUS. You will need to persuade others in order to change the POV of the section. Generalrelative (talk) 19:31, 20 April 2025 (UTC)

Haha alright, fair enough. Skellyret (talk) 19:43, 20 April 2025 (UTC)

IQ and populations with cultures very different

It would be interesting to explore in more detail the issue of measuring IQ in individuals from populations with cultures very different from that of the West, such as the Pirahã people. Could someone expand on this topic in the article, citing relevant sources? Myron Aub (talk) 08:16, 2 June 2026 (UTC)

Systemic bias

Sections "Social correlation", "Public policy", "Genetics and environment", and in "History" all but 1.1 and 1.2 draw almost exclusively from American sources and give the American perspective. There is no discussion or mention how IQ is viewed and used in other parts of the world. If IQ isn't as important in other countries around the world, then that should be reflected in the article.

The problem is that any article that includes the American perspective will necessarily suffer from WP:FALSEBALANCE. Since IQ played and still plays an important role in the United States, it makes sense that most of the sources on IQ are from the United States, which gives the global Wikipedia readership a skewed perspective on IQ.

For example:

• in the lede: IQ scores have been used for educational placement, assessment of intellectual ability, and evaluating job applicants. In research contexts, they have been studied as predictors of job performance[21] and income.[22] As far as I can tell, both of the cited papers are using research done in the United States, but the statement is universal.

A few suggestions:

• Split the article into "Intelligence Quotient in the United States" and "Intelligence Quotient", which would include international perspectives on IQ.
• Rename this article.
• Edit the lede and the body in a way that makes it clear that this article is mostly about IQ in America.

TurboSuperA+_[talk] 10:55, 9 June 2026 (UTC)

Sections on Factor Analysis and the mathematical Implications

I've added two paragraphs covering the fact that Intelligence Quotient tests are universally based on the statistical (mathematical) model of Factor Analysis. This is a relevant area to cover since it pertains to the theoretical underpinnings of Psychometrics in general and Intelligence Quotient research in particular, and the validity of IQ as a tool to probe the characteristics of human intelligence is directly connected to the validity of applied Factor Analysis on the underlying data.

With that background, since NightHeron (talk · contribs) undid my edits although they were thouroughly sourced and questioned the relevance of covering (the well-known methodological limits of) Factor Analysis and how those limits may yield false positives specifically within the domain of human intelligence, I hope we can have a discussion here to align on a way forward. (I will wait 24hrs to hear back and then provisionally restore my edits pending resolution here). Nasu.MusashiKosugi (talk) 04:27, 28 July 2026 (UTC)

First, your long paragraph on “factor analysis” is unreadable, because of the large number of phrases that convey no clear meaning:

• 1. “conforms to its assumptions” What assumptions are you talking about? This conveys nothing to the reader.
• 2. What’s a “positive factor extraction”?
• 3. What “boundaries” are you talking about? Boundaries of what?
• 4. In the context of IQ, what does it mean to say that “dataset is generated by mutually reinforcing network feedback loops”? What “network” are you talking about?
• 5. What “global system bottlenecks” are you referring to?
• 6. Later in the same sentence, what mathematical algorithm are you referring to?
• 7. What do you mean by “background system dynamics”?

Second, you don’t seem to be aware of how controversial the use of “factor analysis” for IQ has been. Many years ago Stephen Jay Gould in his “The Mismeasure of Man” devoted 90 pages (all of Chapter 6) to debunking the misuse of factor analysis by Charles Spearman, Cyril Burt, Arthur Jensen, and others to justify the interpretation of IQ as equivalent to general intelligence (Jensen’s “g factor”). Jensen is a fringe figure in the debate who is best known for racist pseudoscience (having written an article in 1969 in a journal aimed at educators in which he argued that low-scoring minority children are genetically inferior and uneducable). NightHeron (talk) 06:02, 28 July 2026 (UTC)

See below answers point by point for your questions in "First" section

1. Factor Analysis makes quite many and quite narrow mathematical assumptions about the nature of the "mechanism" it is supposed to identify. As soon as those are not met, Factor Analysis breaks one way or another. E.g.:

• Linearity & Additivity: Assumes strict linear relationships between latent factors and observed items, misinterpreting non-linearities as fake factors.
• Continuous Multivariate Normality: Assumes variables follow normal distributions, creating spurious "difficulty" factors from skewed, binary, or ordinal data.
• Local Independence: Assumes item correlations vanish given the factor, misinterpreting direct item-to-item interactions as overarching traits.
• Reflective Causality: Assumes latent variables top-down cause observed symptoms, misinterpreting downstream effects (funnels) as root causes.
• Latent Cause Architecture: Assumes a central underlying trait, mistaking dynamic, mutually reinforcing network interactions for a single entity.
• Etiological Homogeneity: Assumes identical factor scores imply identical causes, grouping structurally distinct upstream triggers into one entity.

2. Here it means that Factor Analysis will when its assumptions are not met frequently not identify that but instead report false positives (i.e. happily report factors and traits that are not real but downstream artifacts from mechanisms that work completely differently)

3. Whether the aforementioned assumptions are met (= within boundaries), or not met (=outside boundaries)

4. I.e. Han van der Maas’s Mutualism Model of Intelligence. Cognitive skills like memory, vocabulary, and spatial reasoning begin early in childhood as distinct, independent abilities. As a child grows, these skills actively reinforce and nurture one another through positive feedback loops, causing all cognitive domains to become strongly interconnected and grow together over time. Factor Analysis misinterprets this web of mutual growth as evidence of a single, overarching biological trait, extracting a false-positive general factor ($g$) from what is actually an emergent system.

5. Situations that violate the "Reflective Causality" assumption (above). Example: For the sake of argument, let's assume that the Prefrontal Cortex is an LLM and the Amygdala sets its "temperature" (LLM concept). In our toy brain, the Amygdala can be triggered to set the temperature by one million external inputs, most of them environmental, some of them hereditary. But whatever the reason, when the amygdala is triggered and the temperature is set, the brain always reacts with a highly covarying set of symptoms. High temperature: hallucinations, tangential drift, formatting breakdowns, and random gibberish, balanced by increased creativity and novel phrasing. Low temperature: accurate and logically consistent outputs that risk becoming repetitive, rigid, overly literal, and trapped in text-looping degeneracies. In this kind of scenario, Factor Analysis will identify huge statistical correlating factors and assume it's caused by a latent inherited trait (rather than the one million heterogenous causes that were set up as the real mechanism in our hypothetical).

6. The latent variable models family of statistical methods. For IQ testing, specifically Item Response Theory (IRT) combined with Confirmatory Factor Analysis (CFA)—operating under the Cattell-Horn-Carroll (CHC) hierarchical framework.

7. Here it means, in the aforementioned scenarios, the "true" mechanisms that produce the statistical data which FA misunderstands as latent traits.

I'm unsure where you're going by pointing out that the useage of factor analysis in IQ methodology is "controversial"? As a factual matter, virtually all modern IQ test batteries rely on Factor Analysis (or its direct mathematical descendants) at some stage of their design, structure, and validation. As such, it makes sense that the Wikipedia page on Intelligence Quotient ought to clearly inform that it is a concept intimately related to Factor Analysis and also be forthcoming with the well known shortcomings of Factor Analysis.

Nasu.MusashiKosugi (talk) 14:33, 1 August 2026 (UTC)

@Nasu.MusashiKosugi, I think you're missing the point of the first part of NightHeron's post. Their list (#1 through #7) consists of examples of unclear writing in the section that you added. A lengthy explanation of the vague jargon on the talk page doesn't make the text any clearer for readers. If it's determined that discussion of factor analysis is appropriate in this article, it needs to be completely rewritten to make sense to readers. Schazjmd (talk) 15:34, 1 August 2026 (UTC)

To clarify the second part of my post, I of course agree that factor analysis is a standard part of statistical methodology. Gould also agrees in his book. What's controversial is not the use but rather the misuse of factor analysis to "prove" that there's a single factor measurable on a linear scale (in practice, they mean IQ) that represents general intelligence. That's what Gould has termed the "reification" fallacy that's in the work of Pearson, Burt, and Jensen. NightHeron (talk) 18:21, 1 August 2026 (UTC)

@SchazjmdOk, fair point, then if we're all aligned I'll take a new stab at writing this up in a clearer way, adressing those unclarities. However, I am not sure it makes sense to go as far into details as I did in response to NightHeron here in the talk page. So I think I should aim to find some middle ground between e.g. "its[FAs] assumptions" and enumerating in detail exactly what those assumptions are. Are we in agreement on that approach?

@NightHeronOk, I think we're on the same page on that, and in fact my edits and my response here speaks to the same circumstance: that Factor Analysis has some serious methodological problems when used as a method that purports to both measure and prove intelligence as a latent human trait.Nasu.MusashiKosugi (talk) 04:10, 2 August 2026 (UTC)