MBTI has decent-ish correlations with career choice. IIRC something like >90% of architects are INT_s (introverted thinkers) and > 80% of counselors/therapist are INF_s (introverted feelers).

But the correlations are not strong enough, nor do they imply causality. And there are other personality metrics which have more predictive power.

That's not to say MBTI has no predictive power - it has some but they are weak. If all you know about someone is their MBTI type, you have more information on their personality than, say, knowing their hair color.

This isn't MBTI. I got confused as well..

https://journals.sagepub.com/doi/abs/10.1177/105960112211214... talks about an "intuitive cognitive style" which looks to tie to https://www.researchgate.net/publication/256450858_Not_All_G... as the easiest place to start, but I'm seeing research back to Cognitive Continuum Theory.

I'm still digging in (this looks like a productive rabbit hole) but this looks like a more rigorous paper based on a different foundation.

It's based on analytical psychology from the 1920s as Jung laid out in "Psychological Types." The funny thing is MBTI was developed for carbon copy paper and as an instrument to quickly type people. So they created the 4th column because in their scoring and layout, they couldn't indicate whether the person was more perceptive (N or S) or reasoning (T or F) focused by putting one in front of the other as Jung did, and also, in their test I guess they found it easier to figure out which function type you extraverted rather than what the attitude (introversion or extroversion) of your primary function was directly.

This leads to a lot of confusion. Jung for example might call a type "Introverted Intuitive with Feeling", and in MBTI that is INFJ, where the J means they extrovert the Feeling, but are primarily actually perceivers! Then there's Introverted Feeling with Intuition, which in MBTI is INFP, the P here meaning they extrovert intuition, but since they're primarily introverted they are "introverted feelers" foremost. I think this MBTI formulation has really made Jung's ideas unnecessarily confusing.

Also, Jung himself was not fond of people typing others, and thought people ought to learn the ideas and type themselves and of course allowed and discussed that some are not differentiated strongly in some dimensions, although he did view that as being a sort of lack of development.

So is it science? I guess I'd call it an interesting analytical model and leave it at that.

What does MBTI have to do with this article? This is a genuine question, since I can only read the abstract. If it's just the word introversion, my understanding is that introversion is still considered a meaningful concept, and its validity is not related to Meyers-Briggs at all.

It is. I find Big 5/OCEAN to be both more scientifically validated and more intuitive to use, as all Big 5 traits exist on a spectrum. It's much easier to understand what it means when someone says they're in the 25th percentile for extraversion (that is, 75% of people are more extraverted than you), and tease out what implications that trait has, than for them to say, "I'm an INTJ" and you need to know not only those 4 traits, but the "cognitive functions" beneath them and how they interact with each other.

I looked into this a bit a while ago, and what I've learned basically boils down to:

The MBTI does measure aspects that roughly correlate with other, scientifically accepted measures (such as the Big Five, where the classic MBTI basically is only missing the 'neuroticism' aspect, which is added as another dimension in some variants of the MBTI). As such, I believe it does have some validity on its own.

But the big problem is that the MBTI simply dissects every one of its dimension into two halves and your 'type' is based on which respective halves you fall into. But if your personality dimensions are well-calibrated, you would expect most people to fall exactly into the middle of the scale, and so it's pretty silly to then fully assign them to the camp they just barely end up in. This still kind of works for people that very clearly are at one end of the spectrum in all the dimensions, but it doesn't really work well for the majority of people.

All that is to say, I think if there was something like the MBTI that also had a neuroticism axis, and would use three intervals (of roughly equal 'expected size' among the general population) per dimension, it would be a lot more accurate. But at that point, you'd basically just have a clustering of the Big Five into 243 'types' -- which makes it a lot less handy than the neat '16 personalities' the MBTI offers.

Doesn't mean there isn't some meaningful signal in its results. Signal/noise ratio might be lower than with actual science, but there is so much data using the framework that the signal part is still useful.

Given 99% of people have never pondered their work behavior or personality impact on work, it's a 0 to 1 thing.

It's useful. It's not science. Those two statements don't have any contradiction. "Pseudo-science" sounds like a dismissal without further evaluation beyond "is it science?" In practice, it's more useful than the scientific big 5 model of personality.

It's absolutely pseudo science. People love it, and it's useful as a tool to recognize "different humans problem solve differently", but like, taking it more seriously than that is a problem.

Grouping humans into 16 categories is pseudo science. The idea that we have traits and tendencies is not, and these things help us understand our own biases if we choose.