News for hackers. Stuff that matters.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by mkn on Friday May 30, 2008 @02:48PM (#204730)
Between the Forer Effect (http://en.wikipedia.org/wiki/Forer_effect), confirmation bias (http://en.wikipedia.org/wiki/Confirmation_bias), and the fact that people routinely retest into different categories (http://en.wikipedia.org/wiki/Myers-Briggs#Study_of_scoring_c...), I don't place much stock in Myers-Briggs. Add to that the dubious way in which the theory was developed, and I'm completely mystified by the appeal of this system to so-called hackers.

It just seems like you could take any 4 either/or categories and have a test that's equally valid. Do you like apples or bananas? Wear corrective lenses or not? Like dogs or cats? Rent your home or own your home? (I'm ACDR, for the record. My type is particularly rare, yet makes up a majority of CEOs and start-up founders. Yay, me!)

Can we please put a stop to this nonsense?

Re:Poll: Hacker News (Myers-Briggs) Personality Types by bentoner on Friday May 30, 2008 @05:05PM (#204866)
You're missing the point. Myers-Brigg types are most useful for understanding other people, not for understanding yourself. Before I read about the other Myers-Brigg types, I just didn't understand why certain other people made decisions as they did, e.g., a member of a group insisting on a worse outcome for everyone to punish someone who'd offended them. I mean, it just wasn't logical!

You see, I was trying to understand/predict others' behaviour by projecting my own personality onto them. Much better is to understand/predict someone's behaviour based on how someone of their Myers-Brigg type would behave.

If understanding how others are different is what Myers-Brigg types are for, then the division into 16 types makes sense. It represents a trade-off between having too few types (and not having predictive power) and having too many (and having to learn too much information of less relevance).

Re:Poll: Hacker News (Myers-Briggs) Personality Types by asdflkj on Saturday May 31, 2008 @01:59PM (#205330)
How are you supposed to know a person's Myers-Briggs type, when you're trying to understand that person's behavior? It's not exactly listed on the nametag. So the best you can do is infer it from what you already know of that person. But if you already know so much as to be able to infer it, you can understand/predict that person's behavior without resorting to any pigeon-holing.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by bentoner on Saturday May 31, 2008 @04:24PM (#205401)
If you're a people-person and can instinctively understand the people you meet, then you're right. But not everyone is born with this skill -- certainly I wasn't. Learning about the other Myers-Briggs types helped me to develop it.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by tokipin on Friday May 30, 2008 @07:57PM (#204963)
those issues have to do with testing, not the theory itself. it's not the theory's fault that humans are so... kaleidoscopic when they try to look at themselves

as expected, people here are overwhelmingly INTJ/INTP. you'll find that's the case in most scientific disciplines

it's not easy to see how the theory is good without having noticed how particularly similar people can be

these type combinations are not just 'any four categories', their effect on a type is greater than the sum of the parts. a bit more descriptive is the underlying Jungian functions. for example, INTJ is syntactic sugar for Ni-Te, introverted intution primary (the part that makes them so confident) and extraverted thinking secondary (the part that makes them logical). INTP is sugar for Ti-Ne, introverted thinking primary (the part that makes them absent-minded), and extroverted intuition secondary (the part that makes them look at the world for answers)

so by flipping the J to a P, the whole type is different. not all letter flips are that significant. I/E is the closest. for example, Feynman (ENTP) and Einstein (INTP) were Ne-Ti and Ti-Ne respectively. Feynman was driven by extroverted intuition, while it served Einstein. the nature of their discoveries is similarly 'playful' and creative

also, these four categories are not poorly-chosen. the most significant category is the S-N axis. it's basically this simple: if you meet someone who shares this same letter (you are both S or both N,) you will click. otherwise you won't

here's some forums where the type differences stand out

http://infp.globalchatter.com/messageboard/ -- fluffy

http://www.intpcentral.com/ -- levitous

http://intjforum.com/ -- dry

one thing to keep in mind is these aren't 16 "boxes." it's a bit complicated. for example, i suspect pg is INTP, but not standard-issue. he might be an Enneagram 3 INTP, for lack of a formal label. it's something he would share with Maddox ^_^

Re:Poll: Hacker News (Myers-Briggs) Personality Types by mkn on Friday May 30, 2008 @09:37PM (#205010)
tokipin: those issues have to do with testing, not the theory itself.

I would argue that if you can't test a given theory, then what you have isn't a theory, but a belief. Otherwise, you're committing the No-True-Scotsman fallacy. The Wikipedia article I cited actually states that self-reporting is to be taken more seriously than the test results. Like, "Oh, I tested INTJ, but feel ESFP, therefore I'm ESFP." This is just rank silliness, akin to an astrologer justifying his failed predictions of your successful romance with, "Well, Mars was entering Gemini, but your love-lines were of different lengths."

As far as the theory itself, the original "Please Understand Me" book is filled with wild extrapolations from anecdata and borderline mystical thinking.

tokipin: also, these four categories are not poorly-chosen. the most significant category is the S-N axis. it's basically this simple: if you meet someone who shares this same letter (you are both S or both N,) you will click. otherwise you won't

This is precisely why this kind of pigeonholing is so maddeningly counterproductive. "Clicking" has just as much to do with personal goals as it does personality similarities, no matter how (mis)measured. What if one is aware that there are benefits to meeting with minds that have perspectives different from one's own? Perhaps that person would most "click" with people wildly differently than his/herself.

tokipin: one thing to keep in mind is these aren't 16 "boxes."

It would appear that this is at least in conflict with the idea that clicking is determined (at least partially) by compatibility on a personality axis. At most, this is an absolute refutation of MB personality theory in that it denies it any predictive power. What's the point of having the categories if you can always plead out of them when the evidence is uncomfortable?

Re:Poll: Hacker News (Myers-Briggs) Personality Types by tokipin on Friday May 30, 2008 @10:54PM (#205044)
> I would argue that if you can't test a given theory, then what you have isn't a theory, but a belief.

most psychology is like this

> The Wikipedia article I cited actually states that self-reporting is to be taken more seriously than the test results.

currently doctors look at pictures to determine if a patient has cancer or some other condition. why can't the computer do that instead? neural networks are used for this sometimes, but it remains that the subject is not so easy to capture with a pixel-wise algorithm. hence, why is it strange that a point-by-point questionnaire is inadequate to ascertain someone's type?

> As far as the theory itself, the original "Please Understand Me" book is filled with wild extrapolations from anecdata and borderline mystical thinking.

the original theory isn't Kiersey's or Myers-Briggs'. it's Jung's. he didn't think of it as a "put me in a box" type of thing. that was how Myers-Briggs sold it. also Kiersey originally used four general archetypes, which he did not notice until later corresponded nicely with NT (rational) / NF (idealist) / SJ (guardian) / SP (artisan)

invention/reinvention of these types occurs often. i remember finding a book published in Einstein's time titled something like "on the two types of intelligence in mathematical discoveries." Jung's theory may not have even existed at that point, but the book was clearly talking about Ni/Ne (by extension, mostly INTJ/INTP)

another reinvention is the programming characters Mort, Elvis, and Einstein. Elvis = INTJ, Einstein = INTP, Mort = probably ISTJ

and again, a lot of psychology is hand-wavy. you're not going to be able to prove any of these things any time soon. that doesn't mean they aren't sound. take a look at those forums i linked, and tell me if you think there is no difference between them, especially no difference having precisely to do with their types

> This is precisely why this kind of pigeonholing is so maddeningly counterproductive.

indeed, it can suck, and some people use typology defensively and offensively, which is certainly counterproductive

> "Clicking" has just as much to do with personal goals as it does personality similarities, no matter how (mis)measured. What if one is aware that there are benefits to meeting with minds that have perspectives different from one's own? Perhaps that person would most "click" with people wildly differently than his/herself.

try it. tell me if you find someone whose second letter isn't the same and you click with them. i guarantee you you won't. i don't mean that in an assholish dismissing sort of way, but the S-N axis is by far the most "categorizing"

it's almost like speaking a completely different language. you could flip every other letter and keep that axis the same, and you will be able to understand eachother easily. but flip just that letter and you're at eachother's throats

is this a generalization? despite the fact that my best friend doesn't share my letter on that axis; no, it isn't. it's 800% proven fact, and will be the case in 800% -- maybe up to as high as 1200% -- of situations. even videogame characters, and the letters S and N themselves, hate eachother due to this rigid dichotomy

> It would appear that this is at least in conflict with the idea that clicking is determined (at least partially) by compatibility on a personality axis. At most, this is an absolute refutation of MB personality theory in that it denies it any predictive power. What's the point of having the categories if you can always plead out of them when the evidence is uncomfortable?

could you think of them as boxes? yes. but they aren't boxes in the way most people would think. genaralizations is more accurate, and gets across what i was trying to say -- you may be introverted, but that doesn't mean you can't go hit corners in your lo-lo wit da gangstaz whilst spewin out madd lyrics, or lead a country as president, etc. in fact interesting things happen when you put types in places they wouldn't normally be. you could think of it as a mutation effect that in some cases is very beneficial

Feynman (ENTP) is a great example. most scientists are INTJ/INTP, but science was lucky and grateful to have Feynman as a member

your type is you, but what you do with you is up to you :)

Re:Poll: Hacker News (Myers-Briggs) Personality Types by hawk on Friday May 30, 2008 @08:47PM (#204990)
you are an entp presumably?
Re:Poll: Hacker News (Myers-Briggs) Personality Types by tokipin on Friday May 30, 2008 @11:27PM (#205061)
intp actually, though my Ne is strong, so you are correct in that sense. this is what i was getting at about how it can get complicated. my 'function composition' is Ti-Ne, but that's talking about the dominance (perhaps conscious/subconscious) of the functions, not their strengths

Jung said everyone has all 8 of the functions in varying degrees of strength. i believe it's possible (and this is stepping out of my bounds) that you may be, say, Ni-Te (INTJ) but have very strong Fe

Re:Poll: Hacker News (Myers-Briggs) Personality Types by Alex3917 on Friday May 30, 2008 @04:03PM (#204811)
"It just seems like you could take any 4 either/or categories and have a test that's equally valid."

The categories used by the Big Five were actually created using empirical research.

http://en.wikipedia.org/wiki/Big_Five_personality_traits#Con...

Re:Poll: Hacker News (Myers-Briggs) Personality Types by nostrademons on Friday May 30, 2008 @06:21PM (#204907)
Okay, then on that note, what's everyone's Big Five personality test score? There's an online inventory at:

http://www.personalitytest.net/ipip/ipipneo120.htm

Mine is 10 (3-4-4-70-17-57) / 73 (70-61-33-80-74-63) / 43 (52-33-41-35-26-80) / 41 (12-20-53-89-48-28) / 67 (81-50-5-38-87-87) in test order (that's Extraversion/Agreeableness/Conscientiousness/Neuroticism/Openness). Interestingly, that seems like it's about the worst possible personality for an entrepreneur, with super low scores in all the extraversion categories, high cautiousness, high self-consciousness, low self-discipline and adventurousness.

Re:Poll: Hacker News (Myers-Briggs) Personality Types by Alex3917 on Friday May 30, 2008 @08:10PM (#204971)
I don't remember my scores offhand, but I remember getting weird results the last time I took an online version. I think it said I had low openness to experience, which is weird, because that factor should be pretty much off the charts for me. It's important to remember that while the categories themselves have empirical support, the specific test you took might not have high reliability for assigning scores within any given category.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by epi0Bauqu on Friday May 30, 2008 @05:17PM (#204881)
Despite the drawbacks in MB, the poll results show an interesting and compelling deviation from expected (i.e. average) results.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by timr on Friday May 30, 2008 @09:51PM (#205016)
Well, the results are certainly skewed, but that's not necessarily an indication of anything interesting.

For example, you can't rule out the possibility that the test is particularly appealing to people with a certain personality (i.e. "hackers"), and that the test happens to label these people identically (i.e. "NT"). The fact that the test was posted here, and that the results are heavily skewed toward a small number of types, could be one big example of selection bias.

Re:Poll: Hacker News (Myers-Briggs) Personality Types by randallsquared on Saturday May 31, 2008 @08:55AM (#205232)
"the test is particularly appealing to people with a certain personality (i.e. "hackers"), and that the test happens to label these people identically (i.e. "NT")."

Isn't this the only claim that proponents of the test make? It's the whole point: people with different personalities are supposed to show up differently on the test. It's not clear to me why you seem to think that the claims for the test are actually claims against it. :)

Re:Poll: Hacker News (Myers-Briggs) Personality Types by timr on Saturday May 31, 2008 @05:10PM (#205421)
Well, the Meyers-Briggs test claims to be able to discriminate many different types, not just NT variants. But NT seems to be a type that is particularly well-identified by the test. You don't often hear SJs (for example) raving about the accuracy of their type classification.

Simpler example: if I created a test that appealed primarily to cooks, written using language and questions that related to food and gastronomy, you probably wouldn't be surprised if that test "profiled" the characteristics of a large number of professional chefs. You could argue that such a test is therefore a good discriminator of professional chefs, but that doesn't necessarily make it an interesting assessment of personality.

Point is, maybe there's something about the analytical nature of the people described as "NT" that leads them to find this test appealing/accurate/relevant. You can control for this phenomenon in an experimental setting, but you can't point to the results here, and suggest that it validates the test because the results are exceptionally skewed.

Re:Poll: Hacker News (Myers-Briggs) Personality Types by llimllib on Saturday May 31, 2008 @01:39AM (#205110)
But even if it is an enormous case of selection bias (which I won't argue against, since it's a position I've held myself), it may still be interesting. And to hold that there's such a wide selection bias among the type of people who frequent this site is, I would argue, an interesting phenomenon.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by timr on Saturday May 31, 2008 @05:21PM (#205426)
I don't disagree. "Interest" is a pretty ambiguous word. That said, if this site appeals to a certain type of person, and the test appeals to the same type of person, then the results aren't surprising. I guess that's what I'm getting at.
Re:Poll: Hacker News (Myers-Briggs) Personality Types by jrsims on Friday May 30, 2008 @08:26PM (#204977)
That's exactly what an INTJ would say. If you were an INTP, you wouldn't have posted that because you would've been second-guessing yourself. :)