Rishon Rishon

Myers Briggs

There's been some talk lately on Gene Expression about Empathizing-Systematizing. I haven't read the sources, but I must say it seems like just a rehash of a piece of a much more well-developed theory of personality that has been around for quite some time, and successfully employed in business and government: the Myers Briggs Type Indicator (MBTI). (Interesting aside: Myers and Briggs were a mother-daughter team.) Since I don't think anyone can claim to intelligently discuss personality without relating to it, if for no other reason than to contest it, I will endeavor to introduce it now. Warning: this is my own slightly idiosyncratic view of the subject.

I, personally have used Myers Briggs productively in both my personal life, and on the job. I was first introduced to Myers Briggs about ten years ago, and it was a transformative experience. (Actually, I knew about it for many years without paying much attention, until one day I saw a book on someone's shelf and began to read...) I can't think of anything else, that can be learned in a couple of hours from a book, that can so utterly change the way a well-educated person sees the world. It was like suddenly being able to see a new color, and with a little reflection and experience it has thoroughly informed the way I understand people and interpersonal relations.

Myers Briggs describes personality types according to four pairs of traits. While often these are treated as binary choices: you are either on thing or the other; I think of them as endpoints of axes: you are somewhere on the continuum between them. An added complication is that personality types describe preferred modes of behavior, while well-rounded people are often quite skilled in behaving in their non-preferred mode, as this site says:

This is analagous to handedness, where you sometimes use your preferred hand (eg: when using a pen to write) and sometimes use your non-preferred hand (eg: the hand you use to change gear whilst driving a car is determined by the design of the car, not your preferences). [It's a British site. In the UK you shift with your left hand - DB]

OK, the first thing people always want to know is, "what's my type". Here are a couple of tests. I didn't take either one of them, so I can't vouch for them. But here's my test:

Extrovert/Introvert (E/I) - If you like to have lots of social relationships, if you enjoy meeting new people, if you often talk to strangers when you encounter them, you are probably an extrovert. If you prefer to concentrate on a few special relationships, if you don't like meeting new people, if you rarely talk to strangers when you encounter them, you are probably an introvert.

Sensing/Intuitive (S/N) - If you like to learn examples first, theory second, if you think in words, if you like details, you are probably sensing. If you like to learn theory first, examples second, if you think visually, if you are impatient with "irrelevant" details (though they may be essential to getting the job done), you are probably intuitive.

Thinking/Feeling (T/F) - If you like thinking about things or ideas, if you enjoy sparring (physical or verbal), if you prefer truth to peace, you are probably thinking. If you like thinking about people, if you especially enjoy making people feel good (or bad, in pathological cases), if you prefer peace (or war, in pathological cases) to truth, you are probably feeling.

Judging/Perceiving (J/P) - If you prefer to make a decision now rather than wait for more data, if you think that there's usually a right way to do things, if you like achieving goals whether or not the goal has any objective value, you are probably judging. If you prefer to wait for more data rather than make a decision, if you think there are usually many right ways to do things, or it usually doesn't matter too much how you do things, if you are comfortable with vaguely defined objectives, you are probably perceiving.

Follow the links above for a description of each of the axes. I think the hardest one to explain, and the most interesting is the S/N axis. (At least to me, for it characterizes my personality more than any of the others - I am an extreme N.) Sensing people tend to relate directly to inputs from their environment, while intuitive people tend to use these inputs to construct complex inner models, and relate to them. A lot of people have trouble differentiating between thinking and judging. If you're having trouble, look at their opposites, for some reason they're easier to distinguish.

So which type are you? (I'm an INTP.) Here are links to descriptions of each type.

Read the description of your type. Does it sound like you? Try varying one letter at a time, especially if you are not sure about one of the answers. Do these types seem somewhat like you? Now switch ALL the letters, how much does this seem like you? (These descriptions are short, and so much less impressive than the descriptions that appear in the book. The first version of this book was my introduction to Myers Briggs. There was another book that I liked better, at the time, but I can't seem to locate it.)

Now comes the fun part. Myers Briggs doesn't just give you a way to describe yourself, it gives you a way to think and talk about personality. For example, the 16 types can be grouped in various ways in order to make more general statements, the most common is: SP, SJ, NT, NF. I often use this particular breakdown when interviewing candidates for a job. Usually I can figure out pretty quickly what a person's personality type is (and when I can't it says something too, that they're probably near the middle of the spectrum, or they're good at using their non-preference). It's my experience that the best predictor of success in a job is not ability but enthusiasm - so I want to know what motivates a person:

SP - Action: These people like activity. All the best athletes are SPs. Soldiers are usually SPs (but officers are usually SJs). The best salesmen are SPs. Lots of really good programmers are SPs - they're they guys that just love programming, I call them computer jocks. To be really good at something, you have to love to do it over and over again, only SPs are capable of this.

SJ - Order: These people love to make order out of chaos. They love directing things, planning things, organizing things. A lot of bosses are SJs. Good secretaries are SJs. Most schoolteachers are SJs. Lots of good programmers are SJs - they're the ones that will research and plan before methodically carrying out the task.

NT - Ideas: These people like thinking about ideas. They like solving problems (not the administrative kind), inventing algorithms, and architecting solutions. Most scientists and engineers are NTs (though a lot of engineers are SJs). Lots of good programmers are NTs (I'm one of those), but they're likely to view programming as a means to an end rather than an end in itself (in contrast to SPs).

NF - Empathy: These people like to help people and express themselves (to people). Naturally, they gravitate to the helping professions: teaching, medicine, social work, social advocacy. They also fill the ranks of artists, writers, journalists. I once saw a claim that they make the best salespeople, and I believe it, but few NFs are interested in sales. NFs are not likely to be interested in programming, but when they are they're motivated by the notion of helping people by what they write, or pleasing the boss.

The 16 types are not distributed equally in the population, by any means. Keirsey claims the following figures (I couldn't find figures for individual types):

SJ: 40% - 45%
SP: 35% - 40%
NF: 8% -10%
NT: 5% - 7%

Assuming that personality types are inherited (and I think they are - my mother is an INFP, my father is an INTJ, and my sister is an INFP), I think this is clearly a case of frequency dependent selection. My skills, for example, as an INTP, are in demand because they are extremely rare. But I don't think I would want to live in a world in which my type were common. I have trouble with a lot of everyday tasks that most people would consider extremely simple, and I'm glad that there are a lot of people around to help me out with them. A typical programming task (for example) can always use another good SP or SJ, but how many NTs does it need? Especially INTPs (NTJs can fake being SJs - their J side enables them to do what is called for at the moment). Ten thousand years ago, I'm not sure what we would do.

The other interesting skew in the percentages is on the T/F axis (and this brings us back to the origin of this post). The T/F axis is the only one which exhibits sexual dimorphism. About 75% of men are Ts, while about 75% of women are Fs. I am quite sure that this explains most of the differences in career choice that we see between men and women, plus a lot of other differences. Anyone ever notice that men and women tend to have different personalities? Does it come as a surprise that most women are feeling, while most men are thinking?

Okay, lets have some more fun. I claimed to be able to tell a person's personality type without much trouble. So let's pick one: Razib. (My estimate of his personality type tells me he won't mind.) First: E or I? Well, that's easy, he's one of the most extroverted people I know (and I don't even know him - I'm judging by the stories he tells, and the fact that he tells them at all), E. Second: S or N? That's a hard one. I would guess S because he writes so fast, and works through so much material. I don't think an N personality is capable of it. Also, his writing style can be very sensual, but that could be an F influence (we'll get to that). However, and this is why I said it was hard, he's clearly very good at building internal models. But I'll go with the preponderance of evidence: S. Third: T or F? Another easy one, he's clearly a thinker. However, I note that he's quite good at using his feeling side when he wants to, T. Fourth: J or P? I think it's a P, I just don't get a goal-directed feeling about him, nor do I see him express strong opinions about a lot of things, usually he keeps his options open: P. So there's my guess: ESTP. Is it right?

(Cross posted at Gene Expression.)

Posted by David Boxenhorn at January 19, 2005 08:45 AM