Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[silverspawn]

Agree this is conceivable. I don't think it's what people generally believe, but I don't actually know that for sure.

But it still contradicts the laws of physics so I don't it's all that much better. I mean, it's a little better because it doesn't require humans being special, as you said.

[silverspawn]

The Top 10 Most annoying Things in the World, v3

10. people describing Elon Musk as a typical selfish billionaire
9. websites demanding special characters in passwords
8. people calling MBTI pseudoscience
7. bureaucracy
6. having to beg high status people for things
5. background music in videos that shouldn't have music
4. people calling crypto a ponzi scheme
3. programs that don't let you choose their installation path
2. people not changing their minds
1. insects

[silverspawn]

Now that I'm working on writing about consciousness and morality, this is no longer annoying, it's just the big problem.

[silverspawn]

This "MBTI is psudosciecne" thing just drives me insane. Every time someone does an MBTI survey in a niche community, the results are skewed away from the general distribution with something like p=0.0000000001. In my prismata team, we were like 4/5 INTJs or something. There are few things so obviously meaningful as MBTI.

[silverspawn]

And the "crypto is a ponzi scheme" is just so bafflingly stupid. Why say things that are so trivially seen to be untrue. I guess Awaclus explanation is the best I have.

[silverspawn]

One problem with Don't Look Up is that I'm think the comet would not, in fact, be an extinction event. Its impact would kill people in a very large radius, but nowhere near the entire earth, and then I would guess the average temperature drops by 30 degrees or something because sunlight is blocked out.

I don't think humanity goes extinct in that scenario. Rather, I expect 99.9%+ of people to die while a few manage to adapt to the new world.

Not at all a big deal, though (for the movie I mean); the comet is a metaphor anyway.

[silverspawn]

But could be wrong idk

[Awaclus]

This "MBTI is psudosciecne" thing just drives me insane. Every time someone does an MBTI survey in a niche community, the results are skewed away from the general distribution with something like p=0.0000000001. In my prismata team, we were like 4/5 INTJs or something. There are few things so obviously meaningful as MBTI.

I have probably gotten all of the letters at some point in my life. In 2013, I had apparently taken a test with an ISTP result recently:

I'm ISTP.

However, I just did a test again and got these results:



I don't think it's surprising that members of a tightly-knit community would tend to answer a set of questions about their personalities similarly, but I think this says less about their actual personalities and more about how they want to perceive themselves. People criticize MBTI because the axes are arbitrary, the hard binary categorization of each axis doesn't represent the reality where most people are not that far from the middle on any axis, and the results are not very useful in the long term because people keep getting different results all the time. And perhaps most importantly, if you strongly believe you fundamentally are a specific way because of an MBTI test result you have gotten, and you find yourself in a situation where you would benefit from being different, that belief itself can become a bit of a self-fulfilling prophecy and hinder your brain's ability to adjust to the circumstances.

[silverspawn]

One of the test results I remember distinctly was among waitbutwhy readers. Very likely a majority of them have never taken MBTI, and the results were extremely skewed. I think that itself disproves everything you just said. If the results were just random, you would get approximately even distributions on every community due to the central limit theorem. (A distribution becomes tight around its mean.)

And perhaps most importantly, if you strongly believe you fundamentally are a specific way because of an MBTI test result you have gotten, and you find yourself in a situation where you would benefit from being different, that belief itself can become a bit of a self-fulfilling prophecy and hinder your brain's ability to adjust to the circumstances.

This is a fully general counterargument against any and all personality tests. If you argue like this, and I see people arguing like this all the time, you can refuse to believe in every test no matter what results you get. This is part of the thing that annoys me.

Since you are close to the middle, it's not very surprising that your results change. This is not an argument against the test. The test tells you that you're close the middle. It may be an argument against how the results are communicated. The ENFP label is silly with E and F so close; it should be ?N?P or ???P. I am not close on any axis and have gotten the exact same result every time I've done the test. (Itself p = 1/256 if I remember correctly that I did it thrice.)

By the way, I wish you would have asked me to predict your results before posting.

[silverspawn]

I'll go even further and flip it around. If you're close to the middle every time you do the test, this is just more evidence that the test is measuring something real.

And you're allowed to change from 2013 to 2021 -- in fact, I would argue you did.

[silverspawn]

I think I would have guessed IN?P for you, but of course this is not that helpful anymore with hindsight.

[Awaclus]

One of the test results I remember distinctly was among waitbutwhy readers. Very likely a majority of them have never taken MBTI, and the results were extremely skewed. I think that itself disproves everything you just said. If the results were just random, you would get approximately even distributions on every community due to the central limit theorem. (A distribution becomes tight around its mean.)

I bet if you asked waitbutwhy readers to do a Which Moomin character are you? quiz, you would also get extremely skewed results. That doesn't mean it's scientific.

And perhaps most importantly, if you strongly believe you fundamentally are a specific way because of an MBTI test result you have gotten, and you find yourself in a situation where you would benefit from being different, that belief itself can become a bit of a self-fulfilling prophecy and hinder your brain's ability to adjust to the circumstances.

This is a fully general counterargument against any and all personality tests. If you argue like this, and I see people arguing like this all the time, you can refuse to believe in every test no matter what results you get. This is part of the thing that annoys me.

It's not a counterargument against personality tests themselves, it's a counterargument against how some people understand the results. I'm not worried that Moomin quizzes will prevent people from overcoming their difficulties IRL. Maybe I'm wrong, but I get the impression that people treating MBTI results in particular as more significant and permanent than they should is a real phenomenon that exists (and is apparently based on the theory behind the MBTI itself?), and the more scientific attempts at personality tests don't seem to have the same kind of following.

Since you are close to the middle, it's not very surprising that your results change. This is not an argument against the test. The test tells you that you're close the middle. It may be an argument against how the results are communicated. The ENFP label is silly with E and F so close; it should be ?N?P or ???P. I am not close on any axis and have gotten the exact same result every time I've done the test. (Itself p = 1/256 if I remember correctly that I did it thrice.)

I'll go even further and flip it around. If you're close to the middle every time you do the test, this is just more evidence that the test is measuring something real.

A ton of people are close to the middle. 50% of people will get a different result after just five weeks. If the test was legit, it would give up trying to map the spectra into binary values so that it can sort people into neat categories, because what results people get IRL is the opposite of a clear dichotomy, it's a bell curve. But, because it is a pseudoscientific test, it's more important for it to uphold the tradition than to reflect the world.

And you're allowed to change from 2013 to 2021 -- in fact, I would argue you did.

it's 2022 now

Sure. But it's not just that I have changed since 2013, I have changed multiple times back and forth, and right now even the way I visualize a lot of the questions is influenced by the fact that I just had an absolute blast drinking calvados and watching Sora no Woto with a bunch of /a/nons over IRC last Saturday, so if it asks me whether I enjoy group activities, of course I'm going to completely agree with that. Which might not be the case if I was instead visualizing a school project with difficult people.

By the way, I wish you would have asked me to predict your results before posting.

Eh, good point. Well, you can predict my Moomin character if you want.

[faust]

This "MBTI is psudosciecne" thing just drives me insane. Every time someone does an MBTI survey in a niche community, the results are skewed away from the general distribution with something like p=0.0000000001. In my prismata team, we were like 4/5 INTJs or something. There are few things so obviously meaningful as MBTI.

I am not in any way an expert, but it was always my impression that MBTI is pseudoscience. There seem to be a lot of papers debating its value.

Your argument doesn't really work. It says that MBTI measures something, and that seems clear; MBTI indicators are correlated to other more widely accepted indicators. I think the problem is that beyond that, MBTI makes far-reaching claims about the personalities of an INTJ that seem to not be supported by data.

[faust]

But, you can predict my type if you wish. That's a fun one as people always seem to get it wrong.

[silverspawn]

I've already thought about this and found it extremely difficult. I weakly predict N over S, moderately predict F over T, and weakly predict J over P. And on the first axis I have no idea. I think this one requires the kind of information that you get from knowing people in person. (And I would have gotten Awaclus wrong there, I think.)

I assign much less than 50% on getting all three right.

[silverspawn]

Your argument doesn't really work. It says that MBTI measures something, and that seems clear

This is really the only point I was trying to make. Well, that and that this something varies strongly across different communities.

In social science, this is basically the standard: finding real correlations. If you accept this, how do you define whether or not it's science? (And why does this even matter?)

[faust]

I've already thought about this and found it extremely difficult. I weakly predict N over S, moderately predict F over T, and weakly predict J over P. And on the first axis I have no idea. I think this one requires the kind of information that you get from knowing people in person. (And I would have gotten Awaclus wrong there, I think.)

I assign much less than 50% on getting all three right.

It's 2 out of 3 at least. My result is INFP. P is actually the strongest characteristic it seems; I got a 92% score there in the test I just took.

[faust]

Your argument doesn't really work. It says that MBTI measures something, and that seems clear

This is really the only point I was trying to make. Well, that and that this something varies strongly across different communities.

In social science, this is basically the standard: finding real correlations. If you accept this, how do you define whether or not it's science? (And why does this even matter?)

Then I think you misunderstand the criticism of it.

MBTI is claiming much more than "our test measures something". You can use the noise a car makes to try and measure its speed, and there will be some correlation, but that doesn't mean decibel are a scientific measurement of velocity.

MBTI claims in particular that the types are meaningful and that you can make all kinds of predictions about a person based on knowing their type, yet this is not based in any evidence. From what I understand, this is what people talk about when they call MBTI pseudoscience.

(Now that anti-Muskianism is on the verge of dropping out of your most annoying list, I need to find something else to stay relevant. This will do.)

[silverspawn]

The concept of OpenAI is just so damaging that I feel myself losing good will for Elon, even though I don't think it really says much about his character

From what I understand, this is what people talk about when they call MBTI pseudoscience.

I don't know where you meet these people who are making these reasonable arguments.

I've already thought about this and found it extremely difficult. I weakly predict N over S, moderately predict F over T, and weakly predict J over P. And on the first axis I have no idea. I think this one requires the kind of information that you get from knowing people in person. (And I would have gotten Awaclus wrong there, I think.)

I assign much less than 50% on getting all three right.

It's 2 out of 3 at least. My result is INFP. P is actually the strongest characteristic it seems; I got a 92% score there in the test I just took.

I'll be a bit nicer to myself and score this 3/4 since F was the strongest prediction

[silverspawn]

MBTI claims in particular that the types are meaningful and that you can make all kinds of predictions about a person based on knowing their type, yet this is not based in any evidence. From what I understand, this is what people talk about when they call MBTI pseudoscience.

What about correlations between what friends you have/what people you find interesting and their similarity with your type? Do you doubt they exist or do they not count?

[Awaclus]

I don't know where you meet these people who are making these reasonable arguments.

You should know, because the waitbutwhy article about their readers' MBTI types I believe you were talking about mentions one and links to a bunch of posts he has written about it.

[silverspawn]

do you really think I read more than 50% of the articles linked in WBW articles?

[faust]

MBTI claims in particular that the types are meaningful and that you can make all kinds of predictions about a person based on knowing their type, yet this is not based in any evidence. From what I understand, this is what people talk about when they call MBTI pseudoscience.

What about correlations between what friends you have/what people you find interesting and their similarity with your type? Do you doubt they exist or do they not count?

Well, I think it's a function of personality how well you get on with people of a similar personality. If someone's personality makes it hard for them to initiate contact then 2 people with such a trait will probably have a hard time being friends. Or 2 people who both want to be the center of attention will likely clash if they're in the same group of friends.

But in general, yes, people who have similar personalities will understand one another better, and to an extent MBTI measures personality (though it seems there are better ways), so I would expect the correlation exists.

[silverspawn]

I mean, I think you already conceded so much about MBTI that arguing for more seems hard. Not that there still isn't a big difference between how good I think MBTI is and how good you probably do, but this is not the agonizingly overdetermined issue I was complaining about

[silverspawn]

It's pretty funny that even when fiction tries to depict ridiculously terrible president like in Don't Look Up, they just can't help make it seem plausible. Also thinking about the Republican in House of Cards.

And then reality is over there doing this Trump thing which isn't even slightly plausible