Ha, you are right. Every time there's a discussion on MBTI in an online forum, there's a small horde of people who proudly and loudly say INTJ, letting us know how logical and rational they are.
It's a model. Like all models, it's wrong as soon as you delve into the details. Under certain circumstances it can be useful, though.
The reason INTJs seem to be disproportionately drawn to it (actually, like most statements, that's inaccurate - in my experience INTJs and INFJs are most drawn to the MBTI, followed by other N types...which is what the theory would predict) is because INTJs like to make systems and models of how reality works, and MBTI is one model of how psychology works. Any halfway-mature INTJ (or person, really) will recognize that the model is not reality, and that once you dive into the details, you'll find numerous ways where the facts don't line up with theory. But as a tool for orienting yourself, it's a handy first approximation.
Astrology's a fun model to think about too. I wouldn't necessarily want to make predictions about an individual's behavior from it, but then, if I could actually predict an individual's behavior I'd rule the world.
So, where is the evidence that it is a handy first approximation? That is, that it makes any kind of testable predictions? Even if they don't always hold, there should be some correlations with some kind of quantifiable metric. So what quantifiable metrics to these personality categories correlate with?
In the same breath developers will espouse the benefits of TDD, Agile, and the abject failure of Waterfall, that language X is on the up and up and that technology Y improved their team's productivity with absolutely no evidence to back it up, and we all go along with it.
Yet when this topic comes up, we get all the 'pah, not science' comments, some with significant amounts of vitriol.
When everyone is a special snowflake and anyone can make it big if they work hard enough (both of which are false), and programming is a pop culture, I'm not surprised.
> When everyone is a special snowflake and anyone can make it big if they work hard enough
This is my observation of why personality theory upsets people, I should have spelled it out more.
If life success correlates with certain personality traits and IQ, (which are mostly a roll of the dice and genetically influenced, i.e. not egalitarian, cannot be changed), then the reality is that some people are going to achieve more and some are going to have to work a lot less to achieve certain levels of success.
This flies in the face of what society tells itself, that if you have no success it's because you did not work hard enough, that if you work really hard you can do as well as the other guy. This is not true, despite how nice it sounds and how good it makes people feel.
It doesn't matter how hard you try, you are not going to be the next Bill Gates, you are not going to be the next Elon Musk. Applying to Y Combinator and flipping to Google is not within the reach of everyone no matter how many hours you put in.
Modern personality theory brings this to light, and it's bloody uncomfortable for most people. The US Army, for example, during wartime, considered that people with an IQ of less than 85 were not trainable, at all, and were more of a liability than an asset. If I remember right that is roughly 10% of the population. 10% of the population cannot be trained to do anything useful in capitalist society, not even by the Army during wartime.
Now, can everyone just "get a job" or "work hard and make a living"? Science seems to be saying no. The biggest US government institution looking for bodies said no. There is no saftey net built in to society for these people in some of the biggest first world countries. Now imagine being below average... aren't you just lucky the dice rolled in your favour. This is what I meant.
If you don't find this unsettling then you may just be a psychopath!
EDIT: It's not ironic at all that you make a sweeping assumption about me based on something I wrote, not at all :)
You're getting IQ and personality type all wrapped up in each other.
Does personality type have any predictive power at all as to what jobs a person is good at? 29% of programmers are INTJ - OK, great, but that doesn't tell you if they're actually any good at it.
IQ is certainly a better predictor of performance, in as much as every job has a lower bound on IQ.
And you're straw-manning the "hard work leads to success" argument. Probably no one - even the staunchest believer in bootstraps - thinks a sub-85 IQ can be the next Bill Gates no matter how hard they work.
> It's not ironic at all that you make a sweeping assumption about me [...] If you don't find this unsettling then you may just be a psychopath!
tit for tat, I guess.
Look, don't take this the wrong way, but my observation is that most INxxs (especially INTJs) put a lot of stock in the MBTI because they believe their type indicates that they're really smart.
Don't know why this is seen as legitimate. MBTI makes the bad assumption that people that fall on either side of a bell curve (say in terms of extraversion) are categorically different people. More self-fulfilling than anything.
The way it was explained to me is that the letters are vastly less useful alone than with the scores. The score you get says where on that bell curve (approximately, depending on how well-posed the questions were) you lie in each of those scales. Not predictive by any means, but possibly insightful.
Psychologists don't like this test because it's unscientific, and I agree. But something is up in that developers seem to be overwhelmingly INTJ and the rest of the population is not.
Could be that we answer ambiguous questions differently, not that we have different personality traits.
Still, I've noticed that there is a bias in the general workplace against people who are perceived to be introverts, and so I answer ENTJ if asked, which has only happened a couple of times.
Ugh. This comes up so often on Hacker News that I've prepared a FAQ paragraph about how unvalidated the Myers-Briggs test is. Read the articles you'll find at the links for much more information than I have time to type here with arthritic fingers. The Skeptic's Dictionary discussion[1] is a good place to start, but there are other critiques of the test too. Don't miss the discussion of the National Academy of Sciences review of the test.[2] "Overall, the review committee concluded that the MBTI has not demonstrated adequate validity although its popularity and use has been steadily increasing. The National Academy of Sciences review committee concluded that: 'at this time, there is not sufficient, well-designed research to justify the use of the MBTI in career counseling programs,' the very thing that it is most often used for." Psychologists who really know about personality testing don't regard the Myers-Briggs test as a worthwhile test.[3]
An article from the Washington Post newspaper sums up the matter pretty well:
"Now, 50 years after the first time anyone paid money for the test, the Myers-Briggs legacy is reaching the end of the family line. The youngest heirs don’t want it. And it’s not clear whether organizations should, either.
. . . .
"Yet despite its widespread use and vast financial success, and although it was derived from the work of Carl Jung, one of the most famous psychologists of the 20th century, the test is highly questioned by the scientific community."[4]
So, we have a marketing company that helps others advertise to developers based on Myers-Briggs results showing 3 main results? Guess I'm not supposed to be a developer since I keep scoring ENTP (why I've had to take the damn thing multiple times starting at 13 is a bar discussion on various jobs / government grants).
The only model of personality taken seriously by academics at present is the five factor model [1].
It's well researched, the traits were not 'designed', rather the traits popped out of linguistic analysis and statistics, and it's stable across different populations and over most of a person's lifetime.
There is an open test called the IPIP Neo which has either 120 or 300 questions, which you can take online [2].
If you're really interested in what is being taught these days about the theory, this is a good set of lectures take takes you from the early theory of personality through to what is considered science today [3].
It's based on solid science you sniping twerp. Feel free to not 'believe' in statistics if you like. There's a club down the road for anti-vaxxers while you're at it.
By the way, the MBTI has been shown to correlate strongly with the NEO/big5 so your very negative comment is exposed as one of those typical HN negative comments.
MBTI is still the most used personality system even though it isn't empirically keyed. The main issue with this research, however, is that the MBTI tests are often subpar. I won't trust these results unless they tell which test they administered.
You're getting beat up and i feel at least partly responsible. I agree with you wholeheartedly MBTI is as much psychology as Mentos in Diet Coke is chemistry. It's a child's toy, a fun afternoon pastime. Perhaps it will inspire a few to actually go learn things, but this use is bordering on dangerous.
For a long time I had no respect for economics. Or psychology for that matter. The more I learn, the more i see how fabulously difficult it is to tease information out of data. Psychology has got to be the worst. There are a zillion reasons, but right off the bat, people lie. People lie for so many reasons. With computers, I can almost always tease out a crisp yes or no answer to any question i wish. I'm sure the vast majority of HN users default to this kind of thinking, that if i just look hard enough, I'll get that crisp yes or no. And obviously, you don't have crisp yes or no's because you're not looking hard enough.
There's an xkcd about everything is just physics. I believe that's true, all of chemistry is just physics, all of biology is just chemistry, all of psychology is just biology. People fail to appreciate psychology is a better indicator of what can be known. I can't know physics. Quantum Mechanics is famously hard and non intuitive. In every field it's harder and harder to push that edge of knowledge out, because reality at the edges isn't what you think it is. no one "knows" the millions of lines of code that compose the linux kernel. It's too complex. We can dig in and poke around, and some people have fantastically deep understandings, but, 15M is too much. we just can't do it.
Anyway, i missed the deleted comment, i'm sure it was horrible. It wasn't my intention to derail your point.
edit
Oh, it was flagged. Oh, it was you. Ah well, point still stands.
Granted, I have practically no background in academic psychology research. I am willing to believe that there is a kernel of science in psychology.
And I 'believe' in statistics, which is just math after all. This is notwithstanding the apparent fact that the social and life sciences seem to be in a state of panic right now due to issues regarding the use of statistics. At best, it means that a few reports should be discarded. At worst, I have to anticipate that there may be entire fields or subfields that are "science," but that have produced no reliable scientific knowledge base as of yet.
But for a layperson like myself, my only likely contact with psychology is at least a couple steps removed from academic science:
1. Does the application of scientific methodology actually produce useful scientific knowledge in all fields at the present state of the art?
2. Are people who claim to apply psychological knowledge in ways that affect me, likely to use that knowledge in an unbiased and ethical fashion?
In my life, we are talking about people such as educators (of my kids), HR managers, journalists, and little consultants. Should I trust these people to tell me anything that I don't already know about myself or through general common sense?
This is how it can come to pass that something is both "science" and "bullshit" when viewed from opposite ends of a pipeline from basic research to practical application.
Let's assume that when I say statistics I'm talking about the scientific application of statistics and not Fox New's use of statistics, now to your questions:
1. Not always, but that is the nature of scientific research and you should know this.
2. I would say that they would, if you found yourself or a member of your family and/or friends were affected by a mental illness. Which, by the way, is very, very likely.
In regards to common sense, you can't always rely on it, and there is research to prove this.
As for your utilitarian measurement of what is useful to you or bullshit, great! You sound like you'd make a great engineer, as opposed to a research scientist. Though don't hate on the researchers, they discovered some of the things you take for granted in your day to day application of what is 'not bullshit'.
My education is in physics, and I work in industry. I would at least admit that I would not be very happy trying to do research in the social sciences. As for being diagnosed with a mental illness, rumors that practitioners follow diagnostic fads and over-prescribe meds, would prevent me from even seeking a diagnosis unless my suffering were really profound, which it isn't.
The interesting thing to me is that people say that the MBTI is junk pseudoscience, that the Big-5 (OCEAN) model is the only scientifically valid personality test...but then take a look at the correlation coefficients between the MBTI and 4 out of the 5 dimensions of the OCEAN model:
A correlation of 0.7+ is huge in the social sciences - it's significantly more than the correlation between say intelligence and genetics. If I look at the numbers, my interpretation is that the MBTI is largely measuring 4 out of the 5 dimensions of the OCEAN model, but leaves Neuroticism out of the model.
It's pretty hard to market a test that's going to call you neurotic, nor could you write nice positive analyses for it like all the other letters have.
Are there MBTI tests accurate enough to report numbers along each scale instead of just picking a letter at each end? The way it only gives extremes when most people are presumably normal really makes it look bad.
"There's just no evidence behind it," says Adam Grant, an organizational psychologist at the University of Pennsylvania who's written about the shortcomings of the Myers-Briggs previously. "The characteristics measured by the test have almost no predictive power on how happy you'll be in a situation, how you'll perform at your job, or how happy you'll be in your marriage."
On the validity of MBTI, the main use I've found is to use the eight functions as part of my vocabulary. It's a pretty handy approximation to be able to say to someone in conversation "wow, Stephen Fry has a really developed Fe" and have them know what you mean. There an approximation is all you need, and in exchange you get neat symbols to hang entire categories of ideas onto.
On the flip side, it's all too easy to move symbols around in your head without investigating whether the model you've built corresponds to reality...
That chart -- Ow, my eyes! The multi-direction, multi-frequency cross hatching (and it's wrong in the legend). The bars showing the difference (but only absolute value, ignoring sign) between the two other bars. The cheesy 3D effect. A Tufte Hall of Shame candidate.
This survey was taken on CodeProject.com, a site which seems biased toward C#, C++, .NET, Java, and mobile devs.
Perhaps it's just that sample of developers who skew toward INTJ, ENTJ and INTP.
I suspect that the results would be different in a survey of developers from a more open source-focused community, e.g. if GitHub conducted that survey and a lot of idealistic open source developer types responded.
Google had (has?) an informal "just for fun" Meyers-Briggs type badge on the Intranet. Among people who participate, roughly 65% are either INTJ, INTP, or ENTJ, with a further 10% ENTP. INTJs alone made up about 30% of the company, vs. about 3% of the general population. Only about 10-15% were an S of any type (vs. about 75% of the population).
Yes, there's selection bias on multiple levels here. S-types tend to think that the MBTI is hogwash and so would be underrepresented in any voluntary poll. (The other majorly-overrepresented group was INFJs at ~7-8% vs. about 2% in the general population. In my experience, INFJs are not a significant portion of Google, but they are a significant portion of people interested in the MBTI.) Google also has an INTJ founder and a culture very friendly to INTJs, which skews the numbers inside the company.
But I suspect that if anything, the open-source community would be more biased toward the INTJ/ENTJ/INTP axis. The C#/C++/.NET/Java market has a large number of developers who are in programming because it makes for a good living, which is a more S-based trait.
Looking at this thread and not arguing for / against the Myers-Briggs, I do have a question.
Has anyone reading this who took the test scored an S (sensing) instead of the N (intuition)? I see variance in the other three slots but not that one.
Yes, I scored an S (and I am a developer). IMHO, The reason why most people get an "N" and often INTJ is because all the cool answers lead one to be typed as INTJ (I score INTJ sometimes, but I was pretty sure I am not that sharp and did not score straight A's in school) . If one attempts the test thinking he is logical and unemotional as Dr.House or Dexter, he would end up being typed as an INTJ.
A lot of the times questions are ambiguous and I choose an option as I just wanted to get done with the boring part. The interesting part is then later talking about what fiction character has what type and what celeb is of what type and then reading up /r/<MBTI>.
I think the rest of the stuff like correlation with being a developer (and good developer if you look at INTJ forums/descriptions) is generally wrong because 1) people take the test thinking about what they want to be and not what they really are 2) Most of it is plain stereotyping and since S types are generally typed in as less logical, most developers' will end up not getting that as their results. 3) I believe (and have observed in my 20-something life) that hard work trumps everything in the end
51 comments
[ 64.2 ms ] story [ 4592 ms ] threadAnd isn't it funny how just about everyone advertising their MBTI happens to be an INTJ?
If you encounter someone making hiring or dating decisions based on this, walk away.
The reason INTJs seem to be disproportionately drawn to it (actually, like most statements, that's inaccurate - in my experience INTJs and INFJs are most drawn to the MBTI, followed by other N types...which is what the theory would predict) is because INTJs like to make systems and models of how reality works, and MBTI is one model of how psychology works. Any halfway-mature INTJ (or person, really) will recognize that the model is not reality, and that once you dive into the details, you'll find numerous ways where the facts don't line up with theory. But as a tool for orienting yourself, it's a handy first approximation.
So is astrology.
https://news.ycombinator.com/item?id=9412144
The MBTI is reasonably well correlated with 4 of the "Big Five" personality traits, which are on good empirical footing.
https://en.wikipedia.org/wiki/Myers%E2%80%93Briggs_Type_Indi...
Yet when this topic comes up, we get all the 'pah, not science' comments, some with significant amounts of vitriol.
When everyone is a special snowflake and anyone can make it big if they work hard enough (both of which are false), and programming is a pop culture, I'm not surprised.
The whole industry is doing TDD and using the same programming language? Programmers disagree all the time about that stuff.
> When everyone is a special snowflake and anyone can make it big if they work hard enough
You've got some serious baggage.
This is my observation of why personality theory upsets people, I should have spelled it out more.
If life success correlates with certain personality traits and IQ, (which are mostly a roll of the dice and genetically influenced, i.e. not egalitarian, cannot be changed), then the reality is that some people are going to achieve more and some are going to have to work a lot less to achieve certain levels of success.
This flies in the face of what society tells itself, that if you have no success it's because you did not work hard enough, that if you work really hard you can do as well as the other guy. This is not true, despite how nice it sounds and how good it makes people feel.
It doesn't matter how hard you try, you are not going to be the next Bill Gates, you are not going to be the next Elon Musk. Applying to Y Combinator and flipping to Google is not within the reach of everyone no matter how many hours you put in.
Modern personality theory brings this to light, and it's bloody uncomfortable for most people. The US Army, for example, during wartime, considered that people with an IQ of less than 85 were not trainable, at all, and were more of a liability than an asset. If I remember right that is roughly 10% of the population. 10% of the population cannot be trained to do anything useful in capitalist society, not even by the Army during wartime.
Now, can everyone just "get a job" or "work hard and make a living"? Science seems to be saying no. The biggest US government institution looking for bodies said no. There is no saftey net built in to society for these people in some of the biggest first world countries. Now imagine being below average... aren't you just lucky the dice rolled in your favour. This is what I meant.
If you don't find this unsettling then you may just be a psychopath!
EDIT: It's not ironic at all that you make a sweeping assumption about me based on something I wrote, not at all :)
Does personality type have any predictive power at all as to what jobs a person is good at? 29% of programmers are INTJ - OK, great, but that doesn't tell you if they're actually any good at it.
IQ is certainly a better predictor of performance, in as much as every job has a lower bound on IQ.
And you're straw-manning the "hard work leads to success" argument. Probably no one - even the staunchest believer in bootstraps - thinks a sub-85 IQ can be the next Bill Gates no matter how hard they work.
> It's not ironic at all that you make a sweeping assumption about me [...] If you don't find this unsettling then you may just be a psychopath!
tit for tat, I guess.
Look, don't take this the wrong way, but my observation is that most INxxs (especially INTJs) put a lot of stock in the MBTI because they believe their type indicates that they're really smart.
Please don't post things like this here.
I'm an ENTJ by the way, but these tests are really hard for me to take because I always end up being influenced by what I want the result to be.
Still, I've noticed that there is a bias in the general workplace against people who are perceived to be introverts, and so I answer ENTJ if asked, which has only happened a couple of times.
An article from the Washington Post newspaper sums up the matter pretty well:
"Now, 50 years after the first time anyone paid money for the test, the Myers-Briggs legacy is reaching the end of the family line. The youngest heirs don’t want it. And it’s not clear whether organizations should, either.
. . . .
"Yet despite its widespread use and vast financial success, and although it was derived from the work of Carl Jung, one of the most famous psychologists of the 20th century, the test is highly questioned by the scientific community."[4]
[1] http://www.skepdic.com/myersb.html
[2] http://www.psychometric-success.com/personality-tests/person...
[3] http://www.indiana.edu/~jobtalk/HRMWebsite/hrm/articles/deve...
[4] "Myers-Briggs: Does it pay to know your type?" 14 December 2012
http://www.washingtonpost.com/national/on-leadership/myers-b...
but I hope you can still appreciate the irony of your post Mr. Skeptical
A lot of Carl Jung's work is some of the most ridiculous pseudo-science that I've ever seen. MBTI is less credible due to its relationship with Jung.
It's well researched, the traits were not 'designed', rather the traits popped out of linguistic analysis and statistics, and it's stable across different populations and over most of a person's lifetime.
There is an open test called the IPIP Neo which has either 120 or 300 questions, which you can take online [2].
If you're really interested in what is being taught these days about the theory, this is a good set of lectures take takes you from the early theory of personality through to what is considered science today [3].
[1] http://en.wikipedia.org/wiki/Big_Five_personality_traits
[2] http://www.personal.psu.edu/~j5j/IPIP/
[3] https://www.youtube.com/playlist?list=PL22J3VaeABQAhrMCQUa6s...
By the way, the MBTI has been shown to correlate strongly with the NEO/big5 so your very negative comment is exposed as one of those typical HN negative comments.
MBTI is still the most used personality system even though it isn't empirically keyed. The main issue with this research, however, is that the MBTI tests are often subpar. I won't trust these results unless they tell which test they administered.
EDIT: By the way I wasn't talking about the MBTI, if you care to read the thread. Try reading instead of skimming next time.
For a long time I had no respect for economics. Or psychology for that matter. The more I learn, the more i see how fabulously difficult it is to tease information out of data. Psychology has got to be the worst. There are a zillion reasons, but right off the bat, people lie. People lie for so many reasons. With computers, I can almost always tease out a crisp yes or no answer to any question i wish. I'm sure the vast majority of HN users default to this kind of thinking, that if i just look hard enough, I'll get that crisp yes or no. And obviously, you don't have crisp yes or no's because you're not looking hard enough.
There's an xkcd about everything is just physics. I believe that's true, all of chemistry is just physics, all of biology is just chemistry, all of psychology is just biology. People fail to appreciate psychology is a better indicator of what can be known. I can't know physics. Quantum Mechanics is famously hard and non intuitive. In every field it's harder and harder to push that edge of knowledge out, because reality at the edges isn't what you think it is. no one "knows" the millions of lines of code that compose the linux kernel. It's too complex. We can dig in and poke around, and some people have fantastically deep understandings, but, 15M is too much. we just can't do it.
Anyway, i missed the deleted comment, i'm sure it was horrible. It wasn't my intention to derail your point.
edit
Oh, it was flagged. Oh, it was you. Ah well, point still stands.
And I 'believe' in statistics, which is just math after all. This is notwithstanding the apparent fact that the social and life sciences seem to be in a state of panic right now due to issues regarding the use of statistics. At best, it means that a few reports should be discarded. At worst, I have to anticipate that there may be entire fields or subfields that are "science," but that have produced no reliable scientific knowledge base as of yet.
But for a layperson like myself, my only likely contact with psychology is at least a couple steps removed from academic science:
1. Does the application of scientific methodology actually produce useful scientific knowledge in all fields at the present state of the art?
2. Are people who claim to apply psychological knowledge in ways that affect me, likely to use that knowledge in an unbiased and ethical fashion?
In my life, we are talking about people such as educators (of my kids), HR managers, journalists, and little consultants. Should I trust these people to tell me anything that I don't already know about myself or through general common sense?
This is how it can come to pass that something is both "science" and "bullshit" when viewed from opposite ends of a pipeline from basic research to practical application.
1. Not always, but that is the nature of scientific research and you should know this.
2. I would say that they would, if you found yourself or a member of your family and/or friends were affected by a mental illness. Which, by the way, is very, very likely.
In regards to common sense, you can't always rely on it, and there is research to prove this.
As for your utilitarian measurement of what is useful to you or bullshit, great! You sound like you'd make a great engineer, as opposed to a research scientist. Though don't hate on the researchers, they discovered some of the things you take for granted in your day to day application of what is 'not bullshit'.
http://en.wikipedia.org/wiki/Myers%E2%80%93Briggs_Type_Indic...
A correlation of 0.7+ is huge in the social sciences - it's significantly more than the correlation between say intelligence and genetics. If I look at the numbers, my interpretation is that the MBTI is largely measuring 4 out of the 5 dimensions of the OCEAN model, but leaves Neuroticism out of the model.
Are there MBTI tests accurate enough to report numbers along each scale instead of just picking a letter at each end? The way it only gives extremes when most people are presumably normal really makes it look bad.
http://www.vox.com/2014/7/15/5881947/myers-briggs-personalit...
"There's just no evidence behind it," says Adam Grant, an organizational psychologist at the University of Pennsylvania who's written about the shortcomings of the Myers-Briggs previously. "The characteristics measured by the test have almost no predictive power on how happy you'll be in a situation, how you'll perform at your job, or how happy you'll be in your marriage."
On the flip side, it's all too easy to move symbols around in your head without investigating whether the model you've built corresponds to reality...
Perhaps it's just that sample of developers who skew toward INTJ, ENTJ and INTP.
I suspect that the results would be different in a survey of developers from a more open source-focused community, e.g. if GitHub conducted that survey and a lot of idealistic open source developer types responded.
Yes, there's selection bias on multiple levels here. S-types tend to think that the MBTI is hogwash and so would be underrepresented in any voluntary poll. (The other majorly-overrepresented group was INFJs at ~7-8% vs. about 2% in the general population. In my experience, INFJs are not a significant portion of Google, but they are a significant portion of people interested in the MBTI.) Google also has an INTJ founder and a culture very friendly to INTJs, which skews the numbers inside the company.
But I suspect that if anything, the open-source community would be more biased toward the INTJ/ENTJ/INTP axis. The C#/C++/.NET/Java market has a large number of developers who are in programming because it makes for a good living, which is a more S-based trait.
Has anyone reading this who took the test scored an S (sensing) instead of the N (intuition)? I see variance in the other three slots but not that one.