Hacker News Reader

Comment by smokey_circles | Hacker News Reader

I don't know how/why people defend this test.

1. It's inaccurate. And that's not just "technically true", it's very rare that the test reproduces results.

2. Jung's theories (the ones used in the test) aren't just "controversial". They're wrong. He says it himself, in the same document they used to create the test.

It is just a cash cow now.

The _biggest_ problem with the test is it is harmful, not helpful. People believe the results are accurate _and permanent_ and so division is created where in reality there is more overlap than difference.

It's also not good to tell people "these are good" followed up by "you have none of them and never will".

By the way: This is true of a _lot_ of IQ tests too: We're still debating what IQ is, nevermind how to measure it. So we definitely shouldn't be telling people the results of their IQ tests, most of which are also not reproducible.

But the data these tests collect is useful. We just shouldn't wield such aouthority about a theory with a lot of apparent leaks we can't explain. There's a better way

Psychology is a young field. Lots of kinks to work out still

View on HN

Anecdotally I’ve taken an ‘official’ MBTI a few times over a couple decades, and always came out with the same results. I don’t think it’s magic or anything, but I also don’t think it hurts to reflect on how you interact with the world and those around you, and how those around you interact with you in return, and then try to identify some themes in those interactions to think about. I wouldn’t go making any grandiose proclamations from it, but as a conduit to some potentially beneficial reflection I see no harm in it either.

Getting different results is pretty common. But getting the same results isn't very rare. The article points to studies showing MBTI and Big 5 repeatability are comparable.

It was inspired by Jung's theories. It doesn't rely on them.

MBTI related literature stresses it measures preferences not abilities. It says no type is better than any other. And it tends to encourage developing what doesn't come naturally.

You can bin people in any number of arbitrary ways. Just the fact you can bin people does not make the binning usefull.

Meyers-Squid is equally valid as any other fiction driven binning process. Like dividing people into Hogswarts houses.

Stability alone is useless if the predictive qualities do not extend beyond the binning process. Being INTJ only predicts the next time you do the test (if the scoring is stable) you might be more likely to score INTJ again. It does not predict the tested persons’ life outcomes or performance at work, in personal OR statistical sense.

The scores don’t predict anything of value outside of the test framework. This is what it means to be useless pseudoscience. It can be a fun party game and thus be usable as a facilitation tool, but beyond that it’s useless (or harmfull if someone tries to use the scores for something like ’creating balanced teams’).

Are you really saying that testing as intj (for example) doesn't correlate with anything? Because that seems incredibly unlikely. E.g. three years ago Slate Star Codex readers were surveyed and found to be 10 times more likely to be INTJ/INTP than the general population.

Not particularly having a horse in this race, if a theory can't reliably make useful predictions, it's not a very useful theory. There's certainly something to be said about identifying correlations, but it isn't yet clear to me whether we can draw inferences based on those correlations to make useful predictions.

In the SSC example, I wouldn't be especially shocked if there are no useful inferences in either direction, but rather a confounding variable that makes it more likely for SSC readers to answer MBTI answers in a particular way. MBTI itself wouldn't necessarily have a lot to say about this factor. (And given the intellectual, self-aware content of SSC, it seems pretty likely to me that this is the case.)

There are tons of correlations, for example INTP/INTJ subreddits on Reddit being most popular while these types generally are one of the rarest. There's tons of things you can predict with this. INTPs are often going to face similar challenges and obstacles in life, there's similar things that will help them grow as a person that would not help other types.

Sure. Now can we do the same with the other fourteen types, or is the primary value here that there are two kinds of people: INTx and non-INTx?

There's probably some bias here, but I've only ever heard INTx specifically singled out when talking about MBTI. That suggests that MBTI may make too many clear distinctions where reality draws a much softer line. A coarser-grained theory is likely to have more predictive power and be more applicable. Alternatively, a fine-grained theory with the boundaries drawn differently, such as the Big Five, may better account for these softer distinctions.

I'm not saying this is definitively the case. I'm saying that a complex theory that makes a small set of predictions doesn't justify its weight. If MBTI does fall into that category, we should feel no qualms about finding something that works better for us.

(Personally, I get a lot more out of the Bartle taxonomy. Only four broad classifications, and more focused on what value one seeks from certain kinds of games. Not without its flaws, either.)

> INTPs are often going to face similar challenges and obstacles in life, there's similar things that will help them grow as a person that would not help other types.

I really am going to need a citation for this. Here's my new horse: there cannot be 16 broad (even with overlap) classes of obstacles and growth opportunities in life that are specific to each personality type. Humans are simultaneously too subtle for a personality test to provide this level of insight, and too similar for tactics to be inapplicable to all but one type of person. It smells like horoscopes, and there are only twelve of those.

We all have issues. Let's categorize those, not the personality types that we think correlate with them. It borders on claiming that the problems I have in life are because of something wrong with me, and not just the circumstances of life in general.

Have you looked into the "Cognitive Functions" that MBTI theory uses? If not, I'd recommend reading through the 101, 201, and 301 articles here https://personalityjunkie.com/typology-101/

To me I see "INTx" and think that makes no sense because INTJ and INTP have completely different functional stacks. INTJ's are dominant "Introverted Intuition" (Ni) followed by "Extraverted Feeling" (Fe), whereas INTP's are dominant "Introverted Thinking" (Ti) followed by "Extravert Inuition" (Ne). So they're basically completely different.

Once you have a functional stack, then the MBTI theory becomes a lot more subtle: not only does everyone have 4 our of the 8 functions in their stack covering both sides of each of the distinctions, but the relative strength of each of these functions can vary continuously.

And that's before you even get to variations based on experience, which the MBTI doesn't even attempt to describe (you'd typically use someone's MBTI type in conjunction with their life experience to explain their personality as a whole).

I think they just meant INTJ or INTP.

The functions are part of the controversial Jungian ideas the article mentioned. Most people ignore them entirely and treat the preferences as axes.

> The functions are part of the controversial Jungian ideas the article mentioned. Most people ignore them entirely and treat the preferences as axes.

They do. And IMO this is exactly why the results never reproduce. The functions actually make sense and anecdotally seem to correspond to something in the real world (they've rarely if ever been subject to scientific testing as far as I'm aware). The preferences by themselves are just a somewhat less reproducible variant of the Big 5 as criticisms claim.

> I think they just meant INTJ or INTP.

I'm sure they did. My point being that if you look at the functions (which most anyone who takes Myers-Briggs seriously does) then INTJ and INTP aren't similar at all. So talking about them together makes little sense. Although reading up the thread, it looks like they were just suggesting that they both have subreddits. Which I guess isn't really talking about their personalities at all.

MBTI reproduces much better if you treat the preferences as axes.

Most people who take MBTI seriously don't use the functions.

They are similar since they correlate similarly with lots of real world aspects. E.G. most simple is they are both very actively represented on Reddit and second one being top 2 highest average IQs. They also share very many similar interests.

What 'pseudalopex mentioned -- and also note that I was responding to someboody who grouped INTJ/INTP together. I don't have much of a vested interest in MBTI; I was just trying to work from the ancestor post's foundation to think critically about what we're getting from that interpretation.

I know even less about "congnitive functions", but it definitely sounds more subtle than the axes most people interpret MBTI as. My specific criticisms don't immediately transfer.

Thank you for your comment. How apt it is that I found it via synchronicity of reading the /newcomments page :)

I think those Cognitive Functions are just Jungian pseudoscience. They are based on work that is closer to fiction than science. Not any better or worse than any psychologically motivated work of fiction - but fiction never the less.

I like to see it as an as-yet poorly evidenced scientific hypothesis rather than an pseudo-scientific "fact". Which indeed may need to go through further revisions before the correct version is found. But I see lots of circumstantial evidence which makes me think it is worthy of further investigation.

Lots of good fiction feels profound. Good fiction feels right and true. But feeling right does not turn the fiction into science.

Sure, but neither does a lack of evidence mean that something isn't true. It's not science currently. But that doesn't mean it won't be (nor does it mean it will be).

I think you'd be surprised at how many scientific breakthroughs have come from individuals being bloody minded about things that everyone else thought were ridiculous. Science isn't just about testing hypothesis. The other part of it is about generating hypothesis to test. That part tends to be a lot more fluffy and subjective, but it's just as critical to the scientific process.

The main issue I have with Jungian analysis is that it takes Jungs ideas as given and then implements no work to test or validate them. Any authoritarian system that does not evolve is more like religious or political dogma than science.

The difference between Jung and for example Alfred Wegener who was one of the early proponents of plate tectonics is that Wegener's ideas were validated.

"Lots of strange ideas have been true" is not incorrect, but I think there are a lot more strange incorrect ideas out there than correct ones.

Talking about personality models seems to appeal to INTJs and INTPs more than anyone else. Never hearing about other types probably says something about your social circles too.

Are all psychometricians INTx's, then? This sounds like exactly the kind of over-conclusion that bothers people about personality testing and its applications. ;-)

Of course not. It's all probabilistic. I feel like the people who think MBTI and personality tests don't make sense are themselves the ones making those over conclusions and they can only think in binary, black and white way, when it really could be something like INTx is likely to be overrepresented in psychometricians fields by 4x, ESFJ might be underrepresented by 5x, etc.

So for instance maybe there's 6% INTXs in the world, but in psychometry field there happens to be 30% INTX representation. Nobody is making the claim that ALL psy

Or it could be these tests literally do not even discriminate 99% of the population if the measurement error is near the size of the mode width. See my other comment. https://news.ycombinator.com/item?id=26292915

A simpler way to think about it might be "How would I assign a p-value to the personality type?" Once you have that answer, all your "over-representation" stats should, naturally, trim out (or otherwise down-weight) the poor p-value personality assignments (any such over-representation stats that you have heard very likely do no such filtering).

Never hearing about other types probably says something about their social circles because amateur psychometricians come in other types too.

> Sure. Now can we do the same with the other fourteen types, or is the primary value here that there are two kinds of people: INTx and non-INTx?

I started with INTx to have at least one detailed, specific argument that there is correlation and MBTI is at least not completely meaningless. There's also a correlation between INTx and IQ, to go as scientific as we can here. Just search for IQ by personality type. Now you could say we are putting two completely meaningless metrics together. If they truly were, how could they have correlation?

Big Five is not the topic here. The argument is whether MBTI can have meaning and whether it can be useful. Let's say Big Five was better, it doesn't mean that MBTI is completely useless. And I'm not going to argue about Big Five vs MBTI.

> there cannot be 16 broad (even with overlap) classes of obstacles and growth opportunities in life that are specific to each personality type.

No one has made this claim. There are various obstacles and challenges in life, and the only claim is that some personality types are more LIKELY to face some obstacles and challenges than other personality types. E.g. INTX will LIKELY have to put in more conscious effort to improve social skills in order to better communicate with people. This already is useful information. It also gives insights into how a person thinks and perceives the world, if you go into how functions work and which personality types are more likely to use which functions to understand what is going on. If there are 16 categories you can get a quick understanding of what a person might be like if you know the MBTI type, which otherwise you would have to read tons of paper on that person to understand what he/she LIKELY is.

It's all a game of probabilities and spectrum. You could do statements like, this person is 65% I, 70% N, 75% T, 66% P, therefore there's likelihood of this person facing challenges with social situations being 80-90%. The way this person is likely to make decisions is to use internal thinking to make conclusions based on logical analysis. All of this can say a lot about the person just with 4 letters.

> E.g. INTX will LIKELY have to put in more conscious effort to improve social skills in order to better communicate with people

I knew I needed to work on my social skills long before I tested as an INTP. My criticism is not that the correlations don't exist. My criticism is that they don't seem helpful.

> It also gives insights into how a person thinks and perceives the world [...] which otherwise you would have to read tons of paper on that person

Can't you just talk with them? Somebody once asked me for my MBTI class as a shortcut to trying to understand how to interact with me. It didn't work very well for them, and I didn't much appreciate being binned that way.

I've always been binned as a logical person in particular, but my internal landscape has been, subjectively, dominated by creative thought and expression. Logic is just a tool I use to impose some structure on that experience. People who stop at the surface -- and worse, who label me under MBTI -- manifestly do not understand me, my motivations, my goals, or my challenges. (If you go point to some MBTI page on Logicians and say "it mentions creativity right here!" I will be very cross with you.)

To say this is a sore spot may be an understatement.

Helpful is subjective. I don't feel like sharing something personal here though.

Do you take the time to understand everyone you meet, their motivations, their goals, and their challenges? Even the ones who don't want you to?

> Do you take the time to understand everyone you meet, their motivations, their goals, and their challenges? Even the ones who don't want you to?

To the degree our relationship warrants. Don't you? We all form mental models of the people we interact with over time. I don't see how the MBTI helps that process. As mentioned, for some people who have interacted with me, it has hindered that process by establishing a totem to which I am compared.

Is the argument that MBTI is useful when interacting with people you don't spend time with, who have a brief and transient presence in your life? Or is it more akin to birdwatching? I don't see either of those options as especially useful, but I won't fault anybody for it.

In other words, there are a lot of people you don't get to know as individuals really.

You don't have to understand how something helps other people to understand they find it helpful. Any tool can be misused of course.

You seem more interested in belittling other points of view than understanding. I'm done.

The population of people who get an INTP result might have quite many asperger spectrum non-neurotypicals among them. If MB contains an accidental asperger scoring system within itself it does not make it relevant in the _general case_ for other personality types.

That said if INTP can be used as a lodestar for creating asperger communities and peer-support groups it's not a bad thing. But it does not really validate MB in any way as a general purpose psychometric tool.

I personally filled an online MB questionnaire or two in adolescence and certainly found the INTP score as something that helped me acknowledge a few things about myself. But it was not really something I would call fundamentally important.

The more straightforward explanation is that the INTP/INTJ descriptions are much more exciting to someone with introverted traits than any other MBTI result. They are much more likely to be discussed for that reason alone. The glamor of the result affects its distribution, when in reality people who meet the description in a plausible way are quite rare and exceptional.

INTJ and INTP sound awful to other introverts in my experience.

The actual influence seems to be N - I - T - P.

If you look at raw subscriber counts, but it may still be I - N if there are that many more extraverts.

https://www.reddit.com/r/entp/comments/56tfmi/of_subscribers...

Statistically it would be very surprising if Meyers-Briggs score would not have correlations with some other samplings - but that does not make it any more relevant. The fact that you can bin answers does not make your questionnaire in any way relevant unless you have some theory how the binned statistic can be used outside of the scoring process. Meyers-Briggs is all about the scoring, and has no usable theory how that scoring could be used.

A correlation alone does not mean it "means anything". Accidental correlations are quite common. We also notice them really well and tend to give them more meaning beyond them being random events.

Picking only positive correlations from an arbitrary data set is also known as p-hacking - https://en.wikipedia.org/wiki/Data_dredging

INTJ and INTP are basically the "coolest" introvert result of the Myers-Briggs. For a supposedly rare result, it's one of the most trumpeted online and one with which people identify the strongest. The description of the result is essentially what those introverted readers imagine themselves to be when reading rationalist literature.

Thinking INTJ and INTP descriptions sound cool says more about you than the descriptions.

> predictive qualities do not extend beyond the binning process

I agree that these tests should not be used for the purpose of evaluating/predicting the "fit" of any individual in an organization from an HR or leadership perspective. They are helpful, imho, for understanding some characteristics of one's own personality and how it relates to others.

Personality is a real thing. It's also very complex so having some "bins" to structure how you (as an individual) think about it can be super useful. But yeah, in the hands of HR and other amateur psychologists it's not a useful tool.

The article covers reproducibility. The test does reliably reproduce results, at least if you don't quantize to boolean. Jung doesn't matter. Being a cash cow doesn't matter.

There is nothing in the field of psychology with better support than IQ. If you aren't going to accept IQ, then you've entirely tossed out the field of psychology. There is nothing left.

> There is nothing in the field of psychology with better support than IQ. If you aren't going to accept IQ, then you've entirely tossed out the field of psychology. There is nothing left.

Huh. That's quite a bold and incorrect claim. Are you a psychologist?

Perhaps the word "psychometry" would have been better. There isn't any psychometric test more stable than IQ.

That's a bit better, but confusing the statements "IQ is the most temporally stable psychometric construct" and "IQ is the only valid psychometric construct" is still egregious.

I don't read it that way, I think it's meant as 'if you reject IQ, you have raised your standards of evidence so high you must logically reject everything else in psychology/psychometry as well'. It's meant to highlight a contradiction.

Yes, if interpreted like that, I agree with it. What threw me off was "There's nothing left" part. I see now that I might've misinterpreted the meaning.

How do you define/consider “stable” this specific context?

Let's hear it from a psychologist: "If you don’t buy IQ research, you might as well throw out all the rest of psychology." (Jordan Peterson)

https://youtu.be/D7Kn5p7TP_Y?t=2938

For anybody interested in personality research, I warmly recommend this lecture series, by the way. These are the basics, which psychology students learn at university.

I do not understand this statement as claiming that IQ is the entire content of the field of psychology. Do you?

No, of couse not, that would be a very, very strange claim. I agree with rcxdude's reading of the statement: 'if you reject IQ, you have raised your standards of evidence so high you must logically reject everything else in psychology/psychometry as well'

Jordan Peterson is a great example of the failing of Psychology in the academic world. I'll list some quotes by him below and you can decided weather or not he should be listened to, or if he is simply selling right wing propaganda on Youtube.

"If men are pushed too hard to feminize they will become more and more interested in harsh, fascist political ideology."

"the current hierarchy might be predicated on competence." [referring to men being placed higher than women in society]

""Peterson says that "disciplines like women's studies should be defunded", advising freshman students to avoid subjects like sociology, anthropology, English literature, ethnic studies, and racial studies, as well as other fields of study that he believes are corrupted by "post-modern neo-Marxists".""

He also has a bible series on YouTube that asserts that god created the appropriate hierarchy of man, and that it should not be tampered with. We should mold society on the archetype of the book of Genesis. Which in my humble opinion is simply Calvinism in the extreme.

Most of this can be found on YouTube or his Wikipedia page.

This is not to say everything he says is wrong, that would be silly. But I believe it should be approached with a heavy dose of skepticism.

Your conclusion ("the failing of Psychology in the academic world") is not supported by your evidence.

You have demonstrated that there is one prominent psychologist who holds some fairly-obviously terrible and inaccurate beliefs, and is willing to trumpet them to anyone who will listen.

This is no more an indictment of the field of psychology than the occasional right-wing "climate scientist" Fox News brings out to "prove" that climate change is not real is an indictment of the field of climate science.

Many scientific fields likely have a handful of far-right kooks who would be happy to rack up YouTube views promulgating alt-right talking points dressed up in the trappings of science. That says very little about the state of their fields, and a great deal about the state of our nation. And YouTube.

I'm sorry. I was very murky with my wording. I was not trying to say that the field of Psychology had failed or was a failure. What I meant to say was that, our current state of academia has in this instance demonstrated "a" failure. That failure to me is highlighted by the comment I responded to when it was implied that Peterson's view was more "bonafide" because it was from a "real" psychologist. By all academic standards the commenter would be correct. Peterson is not only a Psychologists but also a professor, A widely published and cited one at that. One who has been afforded debates at collages with Sam Harris and Zizek. This to me is a failure of the academic world to distinguish between leaders of the field of psychology and crackpot YouTube personalities. By credentials alone he is both and this is unnerving to me.

That said, of course, I do not believe that it speaks to psychology as a whole. But again I poorly worded that sentence and take responsibility for the misunderstanding. I would only add that I think, and I may very well be wrong, that if a physicist were to parrot "free-energy/perpetual-motion" to anyone that would listen, that person would be promptly removed from teaching at any legitimate university. That is not to say physics is better than psychology, if anything it implies that it is easier than the still very important soft sciences.

>Most of this can be found on YouTube or his Wikipedia page.

Not a Peterson acolyte, a Jungian, or even right wing for that matter, but I'm a political philosophy junkie so I have some familiarity with Peterson's claims, mostly from watching debates and listening to arguments. And I will say this, your random out of context quoting, and hiding behind "check his Wikipedia page and YouTube yourself" is the same sophistry you're accusing him of. To give just one example, the claims around Western Civilization you bring up imply neo fascism, while they are in defense of Enlightenment ideals. Forget Peterson, there has been a debate between idealists and materialists about the value of Enlightenment ideals since the horrors of WWII (Frankfurt School being prominent critics), and a defense in the 90s by Idealists and classical liberals (The End of History) and an attacks by Postmodernists since the 70s (Foucault et. al).

Today there still is that debate, because the Postmodernist epistemology dominates and both other dominant alternatives, let's call them free market Western liberals and Marxists, don't agree. For you to summarize the debate Peterson weighs in on as a Western superiority is straight up manipulation on your part. You hide behind the manipulation by telling readers to go to his Wikipedia page and YouTube videos as if you've just cited something.

I do not believe I mentioned a single thing that you are talking about in your comment. I'm not entirely sure what you are talking about in your comment. You seem to have quite the bone to pick with postmodernism, something I left completely left unaddressed about Peterson.

I only implied that saying men will become fascists if they are feminized is ludicrous. Does "postmodernism" mean feminizing men? And I suggest that the idea of men having rightfully taken their god given place higher up the hierarchy of society, is Calvinistic, by definition this is true. Is this what you think is "western superiority"? because I don't think I've ever said those two words together in my life. I also don't think I quoted out of context. Though it seems you feel the context reaches as far back as the enlightenment, in this way I have not given a history of the world before judging Peterson's arguments as something that should be met with skepticism.

I really do mean to go check it out for yourself. All the quotes I grabbed were from the Wikipedia page, I have not hidden anything. I've seen very little of what Peterson has to say, but all of it strikes me as very crackpot like. If I were a professor and publishing psychology papers I do agree my "citations" would not be up to snuff. But I am not either of those things and was not attempting to cite him in that manner. I think there should be more proof to the legitimacy of Peterson being a scientist rather than my HN comment needing more proof that he is not.

We're not going to debate the ins and outs of whether his views or crazy or correct, not least because I don't agree with many of them, but also because I'm not interested in playing devil's advocate.

I just think it's not fair to quote - anyone - out of context, referencing a Wikipedia collection of a few quotations, and finger that as proof of insane or bad or incorrect views. Your explanation of his views is about as long or longer than the quotes you're referring to. These are extremely nuanced, complex issues, and to summarize with outrageous highlights (feminizing men, which you've brought up) is sophistry.

Peterson can be completely wrong, a bad intellectual, whatever you want to call him, but the correct thing is to understand the ideas that you disagree with without relying on an authority (The New Yorker, among others, which were quoted a few times in your Wikipedia article references) before calling them out, because they're not being portrayed accurately.

You have accused me of saying many things thus far that I have emphatically not said. Which I would point out is worse than not providing "full context". I specifically stated that it would be silly to think everything Peterson said was false. I made almost no explanation of his views that were my own words other than accusing a portion of it to be Calvinistic thinking. If you think there are finer points in believing in hierarchical superiority amongst genders because of "Genesis" feel free to let me know as I must be too unsophisticated to recognize the nuance.

In fact I have not much of a clue as to anything of the things your referring too and lack the imagination to see a context larger than what those quotes establish. I have watched a bit of his videos, they are full of outrageous and unsubstantiated claims, often self defeating, and because of that, to your point, are hard to summarize. There seems to be beliefs that Peterson holds that are dear to you in some way, and I'm sure a great many of them have merit. Feel free to debate Marx, Foucault, WWII, etc. but if you claim the pecking order is the way it should be due to "nature" and "god" expect some kick back. Such as I would have expected a person, who believes those quotes to not be reflective of Peterson's views, would have said exactly what they believe them to have meant instead. Did he not imply that feminism leads to fascism, does he in a more complex way than I can see supporting gender equality? What are these enlightenment ideas that I missed? Not sarcasm, here, I really do not know what you mean.

I'm sorry if quoting the New Yorker, Wikipedia or the man himself is sub-par to your standards. If you have some better sources I'd be glad to hear you out.

Peterson is a jungian hack

Can you elaborate? I’m aware of his controversies in pop culture but I’d be interested in any disagreements within his academic field. I’ve heard him disagree with Angela Duckworth regarding her “grit” attribute but that’s the extent to which I’ve heard

Thank you for the reactionary downvotes, but how about making an actual argument?

(self-reply because I can no longer edit the comment)

> There is nothing in the field of psychology with better support than IQ.

This is not true. See Wikipedia[0][1][2].

IQ actually holds little support, compared with (e.g.) bilateral sound recognition, or somatotopic arrangement on the postcentral gyrus. (Among psychologists, that is; IQ probably has more support amongst the general population, because it's pop-pseudoscience.)

[0]: https://en.wikipedia.org/wiki/History_of_the_race_and_intell...

[1]: https://en.wikipedia.org/wiki/Intelligence_quotient#IQ_testi...

[2]: https://en.wikipedia.org/wiki/Intelligence_quotient#Validity...

What we can say about iq tests is that they test something, and those results remain consistent through life. The results are positively correlated to many things (such as academic success. The people scoring a 9, the maximum result, on the tests used in the Swedish conscription live longer, etc...).

Had it been completely pop-pseudoscience it would not be used by, say, the Swedish school system as a tool to gauge whether someone has special needs that warrant another form of school with extra support. The tests used are however not the same as most people have done online, like WAIS (but for kids. Can't remember the name of it).

> The results are positively correlated to many things (such as academic success.

They would be so correlated even if they were a bad metric. Imagine a score that multiplies academic success and the highest Tetris score obtained within 100 miles of your birthplace, then subtracts 50 points if you're a girl. That would also be positively correlated with academic success.

IQ tests measure something, yes, but they're not magic. (See also: https://en.wikipedia.org/wiki/Goodhart%27s_law) I received a high-quality education including what amounts to IQ-test tuition (to pass an entrance exam); do you know of any studies that attempt to factor out this particular confounder?

IQ tests measure the g factor and also happen to be the best way to measure it. So those opposed to IQ should either come up with a better way to measure g or be open about their rejection of the existence of differences in cognitive abilities.

There's a pretty obvious false dichotomy between "rejection of the existence of differences in cognitive abilities" and assumption that such differences are as shallow as a single factor called g stably related to scores on a collection of pattern recognition tests.

If the best way we have to measure something is very bad, perhaps it's best to say we just can't measure it and keep looking, rather than using a very bad measure. It's like instead of trying to invent scales, we kept trying to measure objects' mass by their volume, which happens to work decently for many common objects, bud fails spectacularly for others. And instead of admitting we just can't measure mass, we'd fill our bags based on volume, and be consistently annoyed when we found we can pick up the "heavier" bag of balloons, but can't pick up the "lighter" bag of bowling balls.

Don't you first have to define what you want to measure and demonstrate that it exists?

Only if you're doing science. It's possible to find things out without science slowing you down, but I wouldn't ordinarily recommend it; science is a great way to avoid the common traps that humans invariably fall into when they try to bypass the scientific method.

> IQ tests measure the g factor

Ah, yes. The g factor: “that which is measured by IQ tests”.

The main arguments for IQ tests, as far as I understand, that they correlate with an assortment of cognitive metrics, which all correlate with each other. This is a question. “It's the g factor” is not an answer; you've just stuck a circularly-defined label on the question (or, rather, on a particular frequentist analysis of the question) and declared it solved.

The best evidence for the existence of the g factor are IQ tests, and IQ tests are optimized to measure the g factor.

Do you see a problem there?

The g factor—or more descriptively—the theory of general intelligence has little support outside of the field of psychometrics. And the field of psychometrics is not known for it’s scientific rigor. Needless to say, there is little reason to believe that the theory of general intelligence has any grounds in reality, nor that the g factor exists outside of an operationally defined concept that might not hold any descriptive power for human behavior outside of a narrow setting of pen and paper tests.

Raw IQ scores aren’t very stable over time. At 90+ peoples raw scores have dropped significantly vs age 30. Trauma, Alcoholism, and other diseases can also lower scores dramatically. You can correlate adult IQ’s with tests taken by 2 year olds, but the individual raw scores again vary significantly.

It seems stable because of how age is adjusting for and the huge sweet spot where 64.2% of the population is grouped between 85 and 115. Basically people with significant developmental disabilities at 5 don’t suddenly become normal, but an IQ of 180 at 5 isn’t stable long term.

At quantum scale or near the speed of light newtonian mechanics breaks down, so we know it a local approximation. However, for set circumstances it approximates it does it well. For many decades physists searched universally true theory to no avail.

My point is this, IQ seems to have signal. It probably doesn't hold up in extreme circumstances. Psychology in particular seems to have a difficult time holding large enough studies to find universal constants that we should accept that IQ might be the best local approxination for now.

That doesn’t support your original point. Color blindness is very persistent over time IQ isn’t. Sure +/- 7 IQ points sounds very accurate, but that can be more than a 30 percentile change from two different tests 1 month apart meaning the actual signal is very weak.

You can read papers saying IQ texts on 2 year olds are strongly correlated with adult IQ, but read the details and individual people’s scores varied wildly.

So sure in aggregate it seems consistent, but that’s largely an artifact of test design. Someone with mental handicaps so profound as to be non communicative will consistently score very low, it doesn’t mean much for the general population.

First of all, scores are normalized to age (you never compare 90 to 30 years olds. Only 90 to 90-ish year olds).

No serious IQ test claim to measure much more than 2 standard deviations up and down (meaning they all break down somewhere above 130-135 or below 70). I have done WISC/WAIS three times (as a part of a psychological assessment) and the conversion to IQ scores were within 3 points (the WAIS/WISC scale measures many different abilities, and can be converted to an "aggregate" IQ score). I asked about this and when people are healthy the tester said people tend to score very similarly on tests, even many years apart.

Yes, the score is computed relative to other 90 year olds, however people’s individual cognitive decline is heavily influenced by heath and behavior so individuals can move up or down significantly on a that relative scale. Thus the computed score at 3 doesn’t accurately predict someone’s score at 90 nearly as well as a test taken at 30 predicts someone’s score at 40.

The issue with the underlying raw score at 90+ is it doesn’t map very well to a bell curve. Many people are quite healthy but a significant number are drastically reduced from their younger self.

My high school transcripts show the results of IQ tests taken in the 4th and 6th grade. Between those two tests, my IQ had dropped by 10%.

What test was it? I have only taken one test (WAIS and WISC, which are more or less the same but for adults and children respectively). It takes about 1 to 1.5 hours and is done alone with the psychologist that administers it.

"as a tool to gauge whether someone has special needs that warrant another form of school with extra support"

Someone is using IQ tests for their intended purpose?!

To be fair the person you replied to wrote it’s the best accepted option in psychology.

Not that it’s actually useful.

Who really cares if psychology goes the way of the dodo?

If IQ is all the field has produced that’s of value and we’ve been doing grand engineering and science things since before it was around (grand in then contemporary context perhaps) what’s the field really getting us?

Does it build infrastructure, feed people, and heal wounds?

Given human propensity for shallow vanity and gatekeeping it often tends to be used along those lines.

Having a high IQ does not equal more political power over anyone.

> Does it build infrastructure, feed people, and heal wounds?

The answer to all these is yes.

> Build infrastructure

Maybe not build but affects design. Civil engineers often use models from cognitive psychologists during the planing phase. E.g. how tall and how luminescent should the traffic lights be to optimize energy savings without risking safety? Or close to a dense population center can we build an airport before the sound pollution starts posing public health risks? Or even: “Is it wise to build water pipes with lead?’

> Feed people

To the same extent that quantum mechanics feeds people, by granting jobs and building models, maybe optimize some aspects of food distribution. A psychologist certainly can affect what food people choose. Food companies have a bunch of psychologists working in their marketing. And governments might hire psychologists to help persuade people to eat healthier food.

> heal wounds

This one is easy. Depression is a deadly disease. Psychotherapy is our best way to treat mild to moderate depression. And even sever depression is best treated with drugs and psychotherapy. Psychologists are saving lives every day. You asked about feeding people above, many people have been cured of anorexia because of psychologists. Behavior therapy is the best way we know of to treat many phobias and obsessive compulsive behavior. PTSD can develop into deadly depression if left untreated. This is also prevented with psychotherapy.

Note that none of the above examples have anything to do with IQ. If IQ goes away, psychologists will still continue their work in improving and saving peoples lives with better infrastructure, better public health policies, and by literally healing (or at least treating) peoples wounds.

IQ—or more accurately the theory of general intelligence—is highly controversial inside the field of psychology. In fact a part from developmental psychologists—which might use IQ tests to determine if a child requires special needs—I bet by far most psychologists go their entire career without ever touching an IQ test or citing a paper because of IQ results.

It is true that the debate about general intelligence is huge. So you see a lot of papers written about it. But this is a subfield inside psychology (in fact a subfield of psychometrics which is a subfield of psychology; the whole field of psychometrics invoking skepticism among many scientist).

I’m gonna do you a favor and list sub-fields of psychology that work regardless of the theory of general intelligence and therefor work just fine without IQ:

* Cognitive psychology (and computational psychology)

* Neuropsychology

* Social psychology (and psychological sociology)

* Pathological Psychology and Psychiatry

* Developmental psychology (although they might use IQ tests; but for different reason as psychometricians do)

* Psychological first aid (not a science but a method; similar to paramedics but for trauma victims)

And here are some generic theories inside psychology which have nothing to do with—and might reject—the theory of general intelligence.

* Behaviorism

* Neurological determinism

* Evolutionary psychology

* Psychoanalysis

Even inside psychometrics (which is the believe that any psychological effect that exists is measurable; and that operational definition is a legitimate way to describe psychological effect) I bet there are people that reject the theory of general intelligence on the grounds that IQ is not measuring general intelligence, but something else.

I agree with your points re. Myers-Briggs, however modern IQ tests are generally regarded as pretty reproducible and an individual’s score remains quite stable over long periods of time. Confidence intervals are about 10 points for decent tests. [1]

They are not perfect of course, but it’s a little unfair to compare MB with IQ tests in this manner. I know you caveated your comment by saying that not all IQ Tests are like this, which is certainly true. However, I see no statistical reason not to tell people the result of a high-quality IQ test.

[1] https://en.wikipedia.org/wiki/Intelligence_quotient#Reliabil...

>2. Jung's theories (the ones used in the test) aren't just "controversial". They're wrong. He says it himself, in the same document they used to create the test.

Not a Jungian, but the "wrong" claim is a bold one, in reference to Jung's theories. Are they even falsifiable ? Psychology isn't a hard science. For example, how could you falsify something like the idea of archetypes?

Not trying to be an ass, but i think "hard science" is a bad term. Psychology is a young, very complex science. Real psychology really started with Freud, and yes he was inaccurate. Way better than the understanding of human behavior at the time (wandering womb, look it up), way off the mark these days. We're talking about less than 100 years of real research in the field. Electricity, physics, chemistry and even modern atomic theory (1803) have plenty more time in comparison. Freud had to escape the Nazis, just so you have a general timeline. We dont look back to the 1800s at all the cats that made atomic models and go, "what a bunch of fuck ups". They were building up on knowledge. In some respects, they weren't necessarily wrong, just insufficient.

I dont understand the pop culture enjoyment of trashing psychology turn. Every real psychologist in practice and in research are the first to admit how little we know, how incredibly complex it is (meaning we still dont even know all the moving parts involved) and the fact we are still at the starting line. The fact is, we're trying to understand the human mind with millions of broken examples. Learning to rebuild a car using old, junk, crashed, broken, pillaged and uncooperatively hostile junkyard cars isnt an easy feat especially if you dont know what a functional car looks like.

Though, Myers Briggs was just a cash grab with some cliff notes on Jung. Its an intelligentsia zodiac horoscope. Cute, fun, dont take it seriously.

> 1. It's inaccurate.

Reproducibility is most directly correlated against certain identifiers. For example no matter where or how I take a MB assessment I am always INTP. This indicates that for some persons, as associated to certain scored identifiers, this test is always reproducible. That is less a flaw in this test than the test subjects and further suggests MB becomes more valid when correlated with other tests that measure self-awareness and objectivity. The same persons that score poorly on objectivity are the ones less likely (less capable) to accept that suggestiveness and sensitivity are greater culprits in their measures.

Same here. I have taken the test multiple times throughout my life and I get INTP every single time. The description is completely accurate too, down to the individual nuances describing an INTP. It’s frighteningly accurate.

I wonder if less common MBTI types are more accurately defined.

I don’t think it’s a matter of either assessment description or population percentage but rather a measure of confidence on questions where a persons may, outside of testing, otherwise defer to social reinforcement.

> I have taken the test multiple times throughout my life and I get INTP every single time.

Are you aware of Hobson's Choice?

You are aware of the Barnum effect, I presume? (https://en.wikipedia.org/wiki/Barnum_effect)

I'm aware of this effect and I never cared for horoscopes, Harry Potter houses and what not.

But INTP description is very specific.

> Chief among Logicians’ interests is exploring and building models for underlying principles and ideas, even going so far as to find these concepts, in their own way, beautiful – this makes them natural mathematicians, systems analysts, and career scientists, especially in more abstract fields such as physics.

Maybe 5% of population would identify with this?

> Logicians are known for their brilliant theories and unrelenting logic – in fact, they are considered the most logically precise of all the personality types. They love patterns, and spotting discrepancies between statements could almost be described as a hobby, making it a bad idea to lie to a Logician.

Maybe 20% of population would identify with this?

[0] https://www.16personalities.com/intp-personality

Can you distinguish yourself from the INTJ description? This was my main problem with the personality types: I agreed wholeheartedly with the (rather long!) descriptions of maybe five of the sixteen.

Anecdotally, whenever I’ve read all 16 descriptions on a site, only one has ever “felt” like me. I’ve never done it blind (reading the descriptions without knowing the types), but each one does come across as distinctive enough that they can’t just be accused of being Barnum phrases.

Another quirk is that the only people who are going to advocate for these tests are those for who felt their own results were highly descriptive and meaningful to them. So they’re useful in that personal sense, one can point to a few paragraphs of text and say “look, this describes how I think and see the world, read this if you want to understand me better”. To me that undeniably has utility, even if it’s only for a minority of the population (assuming most people sit somewhere in the middle of the bell curves and therefore don’t identify that strong with any of that types).

I get the same Hogwarts house every time I take that test too.

You have good points there. From reading the article, the inaccuracy is by inappropriately classifying people into groups, and the groupings lack sensitivity. As you also point out the underlying theoretical framework is also wrong. The article does point out that the Myers Briggs does correlate reasonably well with four of the big 5 though. Having said that I would definitely discourage use of the Myers-Briggs.

The original author has a good point about the naming of the big 5 factors though.

Two things I take issue with though - IQ is reasonably well understood in that we know it measures something reliably and accurately. Exactly what that thing is though, is not really understood - and probably can't be. And how it relates to the real world is also somewhat poorly understood, and may not be possible to fully understand. But it does have good ecological and construct validity. I'm talking about good IQ tests here - any less validation than the huge amount that the Wescheler scales go through and the validation is questionable. I'm aware of the racial biases in IQ testing which is a significant problem.

Psychology isn't a particular young science - big chunks of modern statistical analysis have their roots in 19th and early 20th century psychology. So psychology is a young science in the same sense that modern statistics is. Back when I was studying psychology 20 to 30 years ago, a lot of what is now called "neuroscience" was then called "cognitive science" or more rarely "cognitive neuroscience". Neuroscience sounds sexier though.

I do have a massive issue how many social scientists abuse psychometric techniques in inappropriate ways. Management research is particularly guilty here, and I've been pressured into doing this kind of inappropriate analysis in the past.

So IQ measures something, but we don't know what and we don't know how it impacts the real world. How can we meaningfully say it measures something then?

Without having a horse in a race, it's probably the idea that the results persist between multiple tests and vary by who's taking the test. Persistence seems like a decent lower standard to say that there's _something_ present, even if you're not quite sure what it is.

Well, for example, you can definitely see variations in people's short term memory capacity, basic facilities with dealing with language, and logical problem solving skills. And importantly from a clinical perspective you can see when these are mismatched. However these are very reductive variables, and in the real world there are many more things in interacting with each other than these isolated components.

For many, many years we could not explain why gravity exists and how it works. We could only experience it.

We knew of gravity much before we knew of electricity. Try explain electricity to somebody in year 1300 and you'd be burned as a witch.

And yet we understood electricity much earlier than gravity. How electricity works is much simpler to explain than gravity.

We do know what it measures and we do know what impact it has on real life scenarios.

To some extent, yes. But I think it's better to emphasise the fact that we're ignorant rather than what we know because people really like abusing IQ test results.

> “intelligence is whatever my test measures”

— Alfred Binet, inventor of the IQ test

The main issue is we know IQ is meaningfully correlated with real world outcomes, but also that real world environments are meaningfully correlated with ranges of (non-normalised) IQ in a way we can't adequately explain or replicate. It's less obviously correlated with prior knowledge than most tests, but still very trainable (especially if you're a neural network with no generalised reasoning ability whatsoever...)

And like all types of performance test the "something" it measures is a vector of pattern matching ability (both the component generalisable to solving more difficult problems and the component that doesn't except in the domain of career-relevant tests) but also other things obviously correlated with future performance but not fitting most people's definition of intelligence (attitude towards test, comfort in guessing, time management in timed tests, and to a degree experience), so whilst we know what the scale of the apparent impact is, we don't know really know what's behind the cause

What surprises me about Myers Briggs is that people are placed on each dimension either on one side or on the other. Three no center for each dimension. Yet these traits are normally distributed so most people would fall in the center. Telling most people they are average would not sell well. So yeah, this test is unnecessarily polarizing.

I never thought about the lack of a neutral trait, but you are right.

However I don't think there's any reason to expect these traits to have a normal distribution. Humans have plenty of bimodal-ish traits.

The Central Limit Theorem is one big reason to expect that traits have a normal distribution. Wikipedia has a good explanation. [1] and [2]

[1] https://en.wikipedia.org/wiki/Central_limit_theorem

[2] https://en.wikipedia.org/wiki/De_Moivre–Laplace_theorem

How would the central limit theorem apply to seemingly bimodal distributions like left vs right handedness?

With respect to personality, I believe that there are likely a number of independent factors, perhaps hundreds, that affect in some way where one’s position lies on each of the axes of personality. These factors could be particular childhood traumas, birth order, parent’s behavior/addiction, childhood friends, and genetic factors. Even if each of these numerous random variable is binomial the central limit theorem applies.

This article [1] about what determines a persons handedness starts out:

> Like most aspects of human behavior, handedness is a complex trait that appears to be influenced by multiple factors, including genetics, environment, and chance.

It goes on to say that there are up to 40 genes that contribute. It seems like it's complex enough that the central limit theorem should apply yet handedness is definitely a bimodal distribution.

[1]: https://medlineplus.gov/genetics/understanding/traits/handed...

Handedness is not what people care about in discussing personality assessment.

Check out the big five model: https://en.m.wikipedia.org/wiki/Big_Five_personality_traits

Neuroticism, openness, agreeableness, conscientiousness, extraversion.

On each dimension, many people are average and a few are on each extreme. At least that's the world I've observed. It's not a world full of extreme introverts and extreme extroverts. Most people are in the middle of that scale.

When I did a Myers Briggs test almost 40 years ago, I think I came up INTP (or something like that). I never interpreted the I to be extreme introvert, although I'm definitely on that side of the scale.

I wonder how many people score exactly on the mid-point and when that happens, what do they do?

Central limit theorem does not apply unless each experiment is independent. This is not true of the concepts which psychometrician operationally define including personality factors and intelligence.

As an example: Human height is not normally distributed. It is multi-modal based on geographical location, socioeconomic status and sex. There are genetic and environmental factors that makes it so that each childbirth is not an independent event marking a new height.

The same holds for personality traits (if they exists) and intelligence (if it exists as described by IQ tests). The field of psychometrics has a problem with assigning normal distribution to everything even though there is reason to belief things are not normally distributed.

The traits aren't normally distributed. And the article mentions nearly every test treats them as axes.

Sort of the main point of the article, wouldn’t you agree?

At least that’s why I gathered it proposed a letter “x” for centered traits.

Do you have a link where Jung said his theories are wrong?

Agree, the test has been dismissed enough times. People are far too complex.

It has recently been renamed and represented by Jordan Peterson. Unfortunately, the JP readers I've met who mention his version were unaware of the MB version. The critical history of MB should serve as a warning. They are not eureka moments, but smaller steps.

>It has recently been renamed and represented by Jordan Peterson.

You've been downvoted because the OCEAN/Big Five model is not Myers Briggs.

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

https://en.wikipedia.org/wiki/Myers-Briggs_Type_Indicator

Ok. Thank you.

My reply would be...

By any other name.

Another way of how 'not' to understand the world, but a way to simplify it, for both benign and nefarious reasons. Control is in there somewhere. And it is more than mildly concerning that people here understand something that is considered deeply complicated by most, computers, and yet seem convinced humans can be pigeon-holed into a few 'types'.

Reply on news.ycombinator.com