Distribution of the types

[Jarno]

I was curious how you think about the distribution of the types.

[Creepy-bg]

I don't get out enough to know

[Phaedrus]

You know what I think about this, and I guess that our views are similar, Jarno.

[BLauritson]

Until I see proof otherwise, I'm just going to assume that the type distributions are more or less equal since I don't really have any reason to suggest otherwise. Not that I have any particular reason to suggest equal distribution either, but I like to think that it's more or less equal.

[Expat]

My impression is that they are not evenly distributed, based merely on my personal observations and trying to make sense out of them.

[Phaedrus]

Until I see proof otherwise, I'm just going to assume that the type distributions are more or less equal since I don't really have any reason to suggest otherwise.

Equal distributions are much less common in general than unequal distributions, and that means that it is unwarranted to assume that the type distributions are equal -- unless you have a really strong argument for that hypothesis. Rick has an argument for equal distribution, but that is based on the theory of Socionics itself and to a lesser degree on his personal experiences.

Not that I have any particular reason to suggest equal distribution either, but I like to think that it's more or less equal.

We have a very strong reason to believe that they types are unequally distributed, and that is the MBTI statistics based on millions of MBTI tests results. Such statistics does not tell us the exact percentages of each socionic type, but since the four scales are so similar in both Socionics and MBTT (those who still question what I am saying here are also questioning what for example Dmitri Lytov, Sergei Ganin, and other socionists say about the relations between the four dimensions in both theories, and to every person that can think for him- or herself the four scales in MBTT and Socionics obviously measure roughly the same traits) there is one thing that is not possible to doubt: that the types are not evenly distributed. Some types are definitely more common than others.

[BLauritson]

*sigh* I really can't be bothered to argue my position on this, so I shall say one thing and then leave. MBTI statistics mean nothing to me - I have nothing against the system but I see no reason for me to try and correlate MBTI and Socionics, considering that I learn more or less what I need to know from Socionics alone. I'm personally more comfortable assuming an equal distribution of types anyway and I don't feel strongly enough about the matter to enforce my viewpoint. If that offends you, that's your problem and not mine. Feel free to write a lecture on why you think I'm wrong for not marching to your beat, but know that I will not read it. This conversation is over.

[Phaedrus]

*sigh* I really can't be bothered to argue my position on this, so I shall say one thing and then leave.

That's your privilege of course. But I don't understand why you state an opinion in the first place if you are not interested in discussing it. And isn't the main purpose of participating in forums like this to welcome constructive criticism and arguments?

MBTI statistics mean nothing to me - I have nothing against the system but I see no reason for me to try and correlate MBTI and Socionics, considering that I learn more or less what I need to know from Socionics alone.

You seem to be missing my point. Contrary to you I have very much against Myers-Briggs Type Theory, but as a socionist you should have much less against the MBTI tests, since they are very similar to Socionics tests, and they can be used to test your socionic type as well as your MBTT type. That simple fact is acknowledged among many socionists.

I'm personally more comfortable assuming an equal distribution of types anyway and I don't feel strongly enough about the matter to enforce my viewpoint.

Your personal feelings of comfort are rather irrelevant in this case, don't you think? If that is a typical attitude of yours, maybe SG had a point when he was questioning the correctness of your self-typing.

If that offends you, that's your problem and not mine. Feel free to write a lecture on why you think I'm wrong for not marching to your beat, but know that I will not read it. This conversation is over.

Why should I be offended by your unwillingness to discuss your views or argue for them? If you choose to have false beliefs that's a waste of your intelligence -- but it's your life. You will do as you will anyway.

[Joy]

Proof? When did the lab test that determines one's type (with no margin of error) come out?

[Creepy-bg]

Proof? When did the lab test that determines one's type (with no margin of error) come out?

or bad brain chemestry for that matter

[Phaedrus]

Proof? When did the lab test that determines one's type (with no margin of error) come out?

You too seem to miss my point. I have never claimed that there is a full proof test that determines your type in individual cases. I have only claimed that the statistical pattern that emerges from millions of test results proves (and it definitely does so) that there is not an equal distribution of the types.

[Expat]

Yes -- to see a "proof" would be nice.

I started one little personal project on this, for my own amusement -- since established European monarchies are purely hereditary, the monarchs' types playing no role whatsoever in their position, and at the same time they are public figures, I thought that tracking the types of royal dynasties might give an idea. Of course then we have the issue of the influence of heredity, and also (as always) whether the typings are correct.

Starting with Britain:

- Prince Charles - INTj
- Elizabeth II - ISFj
- George VI - INFj
- Edward VIII - INFp
- George V - ESTj
- Edward VII - ISFp
- Victoria - ISFj
- William IV -?
- George IV - ENTp
- George III - ISFj

I do notice the vast majority of introverts, but those are the types I get from reading about them; perhaps some are extroverts and the position of constitutional monarch tends to make them seem like introverts. But I do think that at least the quadras are correct.

So: 3 Alphas - 1 Beta - 3 Gammas - 1 Delta

Which I don't think at all it's the "universal" distribution -- just illustrating the exercise. The sampling is too small.

[Blaze]

expat, what about diana? also i kinda thought the british monarchy seemed more delta.

[Expat]

expat, what about diana? also i kinda thought the british monarchy seemed more delta.

Diana's type is irrelevant for the exercise since she chose to get into that position, she wasn't born into it, so it's not random anymore. That's also why I left Prince Philip out. And the type of the monarchy as an institution, if that's what you mean, is again totally irrelevant for this exercise.

Having said that, I think Diana was ISFp.

[Blaze]

expat, what about diana? also i kinda thought the british monarchy seemed more delta.

Diana's type is irrelevant for the exercise since she chose to get into that position, she wasn't born into it, so it's not random anymore. That's also why I left Prince Philip out. And the type of the monarchy as an institution, if that's what you mean, is again totally irrelevant for this exercise.

Having said that, I think Diana was ISFp.

ah so relation of activity with the prince? hmmmm

maybe it's irrelevant but so what? lol

[Expat]

ah so relation of activity with the prince? hmmmm

Yeah, but I think it's more complicated than that.

maybe it's irrelevant but so what? lol

Well, I was just emphasizing the point of my little exercise

[ScarlettLux]

This question is impossible to answer

[Phaedrus]

This question is impossible to answer

It would be nice to hear your arguments for that opinion.

[olduser]

lack of suitable data.

but for the sake of theorizing I think there is a balance in some societies and imbalance in others. My university seems about 60/40 intuitive to sensors. The trailerpark seems about 80/20 sensors-intuitives.

[FDG]

From what I've seen with my eyes so far, they are very unevenly distributed. However, my sample is clearly very limited.

[Jarno]

There are a couple of people who voted to have proof for the balanced distribution.

I would like to hear what kind of proof.

( My own proof of the opposite choice is that the dichotomy's seem unbalanced in every report that i've read so far.)

[Rick]

Rick has an argument for equal distribution, but that is based on the theory of Socionics itself and to a lesser degree on his personal experiences.

Based on my next-door neighbor, types are distributed extremely unevenly . Actually, my conclusions were based on a compilation of available statistics from socionists (no, I didn't count MBTI or Keirsey statistics, or statistics from any other automated tests, whether socionic or non-socionic), including my own. I also gave some theoretical arguments, but those are always hypothetical anyway; someone else could "prove" theoretically that any other distribution is more "suitable" for society. Ultimately, this is one of those things that everyone has his own opinion on and doesn't really make a difference unless people start putting together statistical studies.

[Rick]

For fun I decided to compile the type lists used for the benchmark project and see what statistics we get (a total of over 800 people):

extraverts: 51%
introverts: 49%
intuitives: 53%
sensers: 47%
logical: 55%
ethical: 45%
irrational: 52%
rational: 48%

Alpha: 28%
Beta: 29%
Gamma: 24%
Delta: 20%

NT: 30%
NF: 23%
ST: 25%
SF: 22%

[Rick]

I suspect that if you compile celebrity lists by MBTT people, you will actually get similar results. The results I've shown here look pretty much the same as all other statistics I've gathered, plus or minus a few percent.

Guess which type there was most of on the list I compiled? LII. That's that type that's supposed to be least common according to MBTI statistics, I believe. And yet if you look at MBTT sites, you will find tons of famous INTJ.

[Blaze]

i'm behind the idea of equal distribution and i've become convinced about this esp with introvert vs extravert. the problem though with going from you're own experiences is that you'll probably surround yourself with people from your own and neighboring quadras rather than opposing quadras, so it becomes biased.

[FDG]

i'm behind the idea of equal distribution and i've become convinced about this esp with introvert vs extravert. the problem though with going from you're own experiences is that you'll probably surround yourself with people from your own and neighboring quadras rather than opposing quadras, so it becomes biased.

Not really - I know more ENFps than ENFjs for example


and ENTj women?? I know just two in a sample of thousands

[Blaze]

i'm behind the idea of equal distribution and i've become convinced about this esp with introvert vs extravert. the problem though with going from you're own experiences is that you'll probably surround yourself with people from your own and neighboring quadras rather than opposing quadras, so it becomes biased.

Not really - I know more ENFps than ENFjs for example


and ENTj women?? I know just two in a sample of thousands

providing you have typed them correctly, of course

[Exodus]

For fun I decided to compile the type lists used for the benchmark project and see what statistics we get (a total of over 800 people):

Ha, that's an interesting idea. Why didn't I think of that?

My impression is that they are not evenly distributed, based merely on my personal observations and trying to make sense out of them.

From what I've seen with my eyes so far, they are very unevenly distributed. However, my sample is clearly very limited.

Just out of curiosity, how do you compile the statistics? Who do you include?

I've been giving this a lot of thought lately. I used to think that the distribution was even, but I have compiled a lot of people's types, just about everyone I know, and the distribution is not very even at all. But then, there is no really compelling theoretical reason for it to be even, and if I change typings just to make it more so, then it's worse than assuming typing bias on my part. It seems like it's a important question just to verify whether socionics can, even theoretically, be objective. Socionists can't agree on individual types unless they agree on type distribution.

Guess which type there was most of on the list I compiled? LII. That's that type that's supposed to be least common according to MBTI statistics, I believe. And yet if you look at MBTT sites, you will find tons of famous INTJ.

But the average of individual socionists' distributions doesn't mean anything if the individual distributions are different. It just means that the biases are unbiased. (And besides, I know that my list was heavily biased towards INTj celebrities, and Alpha and Beta in general. INTj because I tend to always be on the look-out for famous INTjs, and I think I understand the type better than any other. Alpha and Beta because it was mostly actors, musicians, and people in the public eye, who are probably more likely to be Fe quadra types.)

Something tells me this is an issue that won't be resolved until, again, socionics becomes more scientific. I'm almost ready to give up on physics and switch to neurology at this point. But we'll see.

i'm behind the idea of equal distribution and i've become convinced about this esp with introvert vs extravert. the problem though with going from you're own experiences is that you'll probably surround yourself with people from your own and neighboring quadras rather than opposing quadras, so it becomes biased.

That sounds fine at first, but even if you have a single person in your life from Gamma, then it throws everything off, since their acquaintances become yours, etc. I have hardly any close friends, but I am in constant contact with people of all quadras, and not by choice. Heath's point is more sound; it does depend somewhat on the community you're in.

My method and results:

Many of my typings are people from my (former) high school class. Unfortunately, it's a magnet school, so many of the people chose to be there. But many of them did not, and if I was in a regular high school it would probably be far too big for me to get to know everybody well enough to type them.

The rest of the people are family, ex-coworkers, fellow fencers, teachers, etc.

(This is why I prefer MBTI notation.)

Everybody, N = 424

I: 0.56
E: 0.46
N: 0.49
S: 0.53
T: 0.46
F: 0.56
J: 0.5
P: 0.52

ENTp: 0.054
ISFp: 0.09
ESFj: 0.09
INTj: 0.057
ESTp: 0.047
INFp: 0.059
ENFj: 0.059
ISTj: 0.064
ESFp: 0.054
INTp: 0.071
ENTj: 0.045
ISFj: 0.068
ENFp: 0.059
ISTp: 0.059
ESTj: 0.035
INFj: 0.061

alpha: 0.29
beta: 0.229
gamma: 0.238
delta: 0.215



Males, N = 202

I: 0.5
E: 0.5
N: 0.57
S: 0.44
T: 0.55
F: 0.46
J: 0.43
P: 0.58

ENTp: 0.074
ISFp: 0.054
ESFj: 0.054
INTj: 0.074
ESTp: 0.064
INFp: 0.054
ENFj: 0.069
ISTj: 0.054
ESFp: 0.079
INTp: 0.109
ENTj: 0.059
ISFj: 0.02
ENFp: 0.059
ISTp: 0.069
ESTj: 0.035
INFj: 0.054

alpha: 0.257
beta: 0.243
gamma: 0.267
delta: 0.218



Females, N = 221

I: 0.61
E: 0.42
N: 0.42
S: 0.62
T: 0.37
F: 0.66
J: 0.57
P: 0.47

ENTp: 0.036
ISFp: 0.122
ESFj: 0.122
INTj: 0.041
ESTp: 0.032
INFp: 0.063
ENFj: 0.045
ISTj: 0.072
ESFp: 0.032
INTp: 0.036
ENTj: 0.032
ISFj: 0.113
ENFp: 0.059
ISTp: 0.05
ESTj: 0.036
INFj: 0.068

alpha: 0.321
beta: 0.213
gamma: 0.213
delta: 0.213


That's rather striking. Females tend much more to be SFs, though not so much ESFp. Men, and people in general, are slightly more likely to be Alpha or Gamma, though women are very much more likely to be Alpha, with all other quadras being exactly even among women. Maybe I see other males more as competition?...or maybe I'm too focused on finding my dual, hehe. But this typing sample is fairly small, and it probably is biased towards Alpha because of voluntary associations, despite my response to blaze.

One interesting thing is that the idea that ethical types are more common among women, and logical types among men holds up in general, but not for ISTj, ENFj, ESFp, ESTj, and ENFp, which is even.

Here is the sample of all my classmates, probably the most random subset of the data:

Class of 2007, N = 145

I: 0.59
E: 0.41
N: 0.56
S: 0.46
T: 0.39
F: 0.62
J: 0.46
P: 0.55

ENTp: 0.034
ISFp: 0.055
ESFj: 0.028
INTj: 0.062
ESTp: 0.048
INFp: 0.09
ENFj: 0.09
ISTj: 0.041
ESFp: 0.069
INTp: 0.076
ENTj: 0.014
ISFj: 0.097
ENFp: 0.097
ISTp: 0.069
ESTj: 0.034
INFj: 0.083

alpha: 0.179
beta: 0.269
gamma: 0.255
delta: 0.283


This is after many changed typings. At first the distribution was highly Alpha-centered, and my stats program was down for a long time, so this change was a big surprise. I can see that the integral type of my grade was probably Gamma (or maybe Delta). We were well known for not having much group cohesion, as opposed to the class of '06, which, I kid you not, is usually described as "Very team-spirited. Nice individually, but real jerks in a group."

Actually, I think the system of education our school uses, the International Baccalaureate, is INTj. It puts a lot of emphasis on the philosophical side of learning, and has very rigorous grading standards, which they analyze statistically to make sure it's objective and fair for everyone. The highest-ranked IB "officer" at the school is INTj. I figure earlier classes at my school, the ones where it was just starting out, were probably alpha as well. Currently the school is going through a giant upheaval; 7 teachers are leaving. It's crazy. But come to think of it, I have a theory about that too...

[FDG]

i'm behind the idea of equal distribution and i've become convinced about this esp with introvert vs extravert. the problem though with going from you're own experiences is that you'll probably surround yourself with people from your own and neighboring quadras rather than opposing quadras, so it becomes biased.

Not really - I know more ENFps than ENFjs for example


and ENTj women?? I know just two in a sample of thousands

providing you have typed them correctly, of course

After 2 years of typings, I've become pretty sure of my skills when typing from a short psychological distance;

anyway, hotelambush, I don't compile them with precision, just an approximation - in order for me to declare the distribution very skewed, it's sufficient to find only a type whose ratio over the total population is much lower than any other - and as Walras' Law says, if at least one type is not in equilibrium, then there must be at least another one which isn't.

[Expat]

My impression is that they are not evenly distributed, based merely on my personal observations and trying to make sense out of them.

From what I've seen with my eyes so far, they are very unevenly distributed. However, my sample is clearly very limited.

Just out of curiosity, how do you compile the statistics? Who do you include?

Since I've got interested in socionics I sometimes think about the types of people I've met throughout my life, in several countries of the world (a nice distraction when on long drives) -- but I haven't really done any precise statistics because of the imprecise nature of the subject.

My family (parents and brother and his wife) are all Alpha, so obviously there is a huge overload of Alphas from that side and I try to compensate for that by focusing on people who meet more of less "randomly".

My not-so-precise-and-ever-so-changing impressions --

- Against the conventional wisdom, I tend to think that ethical types are more common among both men and women. I think Baby also had this impression some time ago.
- I think that NTs are somewhat less common and SFs somewhat more common than the other clubs.
- Rationals and Irrationals seem to be more or less evenly balanced.
- Somewhat more introverts than extroverts.
- some types I seem to meet all the time, without any effort (on business, etc): ISTj, ISFp, ESFj seem to be the most frequent. Others I have to make a huge effort to think of RL examples: ENTj, INTp, INTj. This may of course not be an objetive perception.

[Rick]

For fun I decided to compile the type lists used for the benchmark project and see what statistics we get (a total of over 800 people):

Ha, that's an interesting idea. Why didn't I think of that?

My impression is that they are not evenly distributed, based merely on my personal observations and trying to make sense out of them.

From what I've seen with my eyes so far, they are very unevenly distributed. However, my sample is clearly very limited.

Just out of curiosity, how do you compile the statistics? Who do you include?

So far I've only been taking statistics presented by publishing socionists or by studies of socionists (summarized at http://www.socionics.us/philosophy/t...ribution.shtml). Now, with our benchmark list, I can add a fifth set to the list.

So far I haven't taken into consideration lists presented on forums, although a quick review of them is enough to confirm the general picture obtained from the rest of my studies (i.e. that separate individuals may have their typing "biases," but that collectively they almost completely cancel each other out). I'm not disputing people's typing skills here; I'm just showing the end result if you mix them together.

I used to think that the distribution was even, but I have compiled a lot of people's types, just about everyone I know, and the distribution is not very even at all.

By my standards your samples are fairly even. You don't get, for example, that 70% of people end up on one pole of a dichotomy. Your greatest "lopsidedness" in both-sex samples is 62% ethicals in your graduating class.

Socionists can't agree on individual types unless they agree on type distribution.

I'm not sure about this, but I think widely different distributions is a sign of significant differences in theoretical understanding and typing paradigms.

I know that my list was heavily biased towards INTj celebrities, and Alpha and Beta in general.

And our participants were primarily intuiters, with more alphas and gammas, but our end distribution was still quite even.

Something tells me this is an issue that won't be resolved until, again, socionics becomes more scientific. I'm almost ready to give up on physics and switch to neurology at this point. But we'll see.

I agree. "Why not do both?" says the in me

My method and results:

Wow, I'm impressed.

At first the distribution was highly Alpha-centered...

Yeah, most people go through that. My teacher was extremely Delta-centric.

-----------

Another potential source of info is to take the Russians' typing data and compile it (from http://www.google.com/translate?u=ht...&hl=en&ie=UTF8) -- I mean not the end result (which I have already done), but the individual typings. Since we don't know the number of each person's typings, we can just use the percentages of each type in their list and assume that there will not be a "list length" does not correspond with "types preferred." I will get back to you in 15 minutes...

[Joy]

The beatings will continue until morale improves.

<3

(and now back to your regularly scheduled thread about the distribution of types)

[Rick]

OK, I have my results. (this was probably a big waste of time

Here's what I got with all the lists (36 -- I took out the Keirsey and MBTT list):

percentages by type (traditional list order, by quadra):

alpha: 22.0
6.8
5.0
4.6
5.6

beta: 31.0
12.2
6.3
7.5
5.0

gamma: 25.2
5.9
7.1
6.8
5.4

delta: 21.9
5.6
4.1
6.4
5.7

E 55.8
I 44.2
J 50.7
P 49.3
N 53.9
S 46.1
T 51.3
F 48.7

-----------

A few of the lists were very unbalanced (over 1/3 of the typings belonging to one type, or over half of the types receiving no typings), so I removed them (4) and got the new stats:

alpha: 23.1
6.4
5.6
5.1
6.0

beta: 28.3
9.1
6.6
7.7
5.0

gamma: 26.0
6.4
7.7
6.7
5.2

delta: 22.6
6.1
4.6
6.0
5.8

E 53.6
I 46.4
J 49.4
P 50.6
N 51.5
S 48.5
T 53.0
F 47.0

In the second sample, no type is twice or more as common than another, whereas the first list had disproportionately many EIEs -- nearly 3 times as many as EIIs (two of the wacko lists were Gulenko-2004 and Meged/Ovcharov-2004, who suddenly decided together around 2000 that EIEs rule the world).

[Blaze]

i'm behind the idea of equal distribution and i've become convinced about this esp with introvert vs extravert. the problem though with going from you're own experiences is that you'll probably surround yourself with people from your own and neighboring quadras rather than opposing quadras, so it becomes biased.

Not really - I know more ENFps than ENFjs for example


and ENTj women?? I know just two in a sample of thousands

providing you have typed them correctly, of course

After 2 years of typings, I've become pretty sure of my skills when typing from a short psychological distance

why don't you take a fucking chill pill fdg? hahaha what are you fully licensed socionist? you go to school for psychology? you have 40 years of life experience?

[FDG]

reading it back, it looks like i was pissed, but i was merely stating a fact (lol)

[Blaze]

reading it back, it looks like i was pissed, but i was merely stating a fact (lol)

hahaha i spose it is a fact that you believe you are good at typing people... j/k

[Jarno]

Well ESTP's are known to be masters of reading body language.

Although it doesn't mean you're great at typing, for that you need more then that and have a good interpretation too.

I don't know exactly when you have enough experience with typing, 2 years seems adequate, but it still can be improved.

Only if someone can VI all 16 types, male and female, he's probably a master.

[Animal]

My not-so-precise-and-ever-so-changing impressions --

- Against the conventional wisdom, I tend to think that ethical types are more common among both men and women. I think Baby also had this impression some time ago.
- I think that NTs are somewhat less common and SFs somewhat more common than the other clubs.
- Rationals and Irrationals seem to be more or less evenly balanced.
- Somewhat more introverts than extroverts.
- some types I seem to meet all the time, without any effort (on business, etc): ISTj, ISFp, ESFj seem to be the most frequent. Others I have to make a huge effort to think of RL examples: ENTj, INTp, INTj. This may of course not be an objetive perception.

Yes, IME Si and Fe tend to be the most common functions, followed by Se, Ti, Ne, and Fi. Ni and Te tend to be least common. I believe there are more sensors than intuitives and more ethical types than logical (and as Expat says, among both sexes). I do think, though, that introverts and extraverts tend to be more equally distributed, although possibly slightly more extraverts.

[olduser]

College communities have a very good cross-section of types, but I still find very few Gammas. But my conception of Gammas might be poor. I know a lot of betas/alphas.

[Expat]

Yes, IME Si and Fe tend to be the most common functions, followed by Se, Ti, Ne, and Fi. Ni and Te tend to be least common.

That perfectly agrees with my own experience.

Also, most of publicity and marketing is Si and Fe. Surely it's an indication that they are among the most common functions?