Log in

No account? Create an account
.:::. ..::.: .:.::..:.::. .::::
corvi [userpic]
'cause what good's a paid account if you don't annoy people with polls?

So I was poking around and noticed there are a lot of livejournal people who have their Meyers-Briggs type listed in their userinfo. I'm curious about the distribution, though.

Poll #173111 Meyers-Briggs

Do you know your Meyers-Briggs type? (If not, and you're bored, there seem to be plenty of online tests findable via Google. I just went and looked at one myself)Check two or more if you're between them.


On a scale of 1 (utter bullshit) to 5 (absolute truth-with-a-capital-T), how useful do you think this particular personality classification is?

Mean: 2.97 Median: 3 Std. Dev 1.06

If you know your "type", how accurate do you think the description is? From 1 (not you at all) to 5 (this guy must know me).

Mean: 3.87 Median: 4 Std. Dev 0.92


(reading you on damiana_swan's friends page)

From an entry in my journal:

I did IPIP-NEO awhile ago. I think the model behind it ("Five-Factor") is far more credible than that of the Myers-Briggs Type Indicator (old Jungian theory, mostly). Myers-Briggs may be falling out of favour among psychologists (I hope), but its popularity with the public remains high. Perhaps these are for the same reasons. In particular, I can't help noticing Myers-Briggs shares certain things in common with astrology:

  • Myers-Briggs is categorical, assigning to everyone one of sixteen concrete identities (mine is "INTP", by the way). Identities appeal to people more than linear ratings. This has been the first to fall, actually, since many people now take a more linear approach to it, particularly in such things as introversion/extraversion where it's pretty clear that there's a spectrum.

  • Myers-Briggs is non-judgemental. Each "type" has a part to play in the world. Compare that to, say, IPIP-NEO, which can report things such as "your level of altruism is low" -- something that may be useful information, but not something anyone wants to hear about themself.

  • The Myers-Briggs "types" form a nice symmetrical 2x2x2x2 system, just as the twelve signs of astrology are given a 3x4 structure. Symmetry appeals to people.

Myers-Briggs makes strong claims about the longitudinal stability of its assessment -- your "type" is not supposed to switch over your lifetime. I wonder about IPIP-NEO, to what degree its assessments of your personality in its Five-Factor Model might vary with life situation.

Thank you; this was exactly the sort of thing I'm interested in. I'll go look at those URLs.

From the five factor tests I've seen, I think the concepts that they are trying to measure are more useful in most ways than what the Myers-Briggs measures. But I'm very dubious so far as to whether any of the tests yet devised come anywhere near to measuring them. Good concepts, poor tests, IMO. (I believe I've taken the IPIP-NEO, but I don't have time to check the link right now.)

For all the I think the Myers-Briggs is conceptually flawed and a little weird in terms of what it's trying to measure, I think it has the advantages of being a well written test, and a test that has been administered very broadly, and the results them self looked into in many ways over a period of time. The latter is useful because it becomes a way of validating the test -- not only do we know what the test is supposed to be measuring, but we have data about what (if anything) it does actually measure.

Psych girl totally agrees with this. I've done extensive work with personality testing, and although the MBTI is a poor measure of overall personality, because it misses too many areas, it's a good measure of the aspects it does reach.

And I have yet to see a simple five-factor test that's decently valid- although the five-factor model is far superior as a description of personality.

Five-Factor has six sub-factors in each factor, so it's trying to measure 30 different scales. The long version of IPIP-NEO has 300 questions and takes about an hour. So even if self-assessment is a good idea, and even if the questions are good, it's kind of caught between trying to collect enough data to make an assessment and avoiding "test fatigue".

But I'm more interested in five-factor longitudinal stability. Does it change with my mood? Am I always going to be this neurotic or are there things I can do to change that?

I think nearly any sort of personality test is destined to be inaccurate, because mood would effect the overall outcome, likely moreso than actual personality traits. If I have a headache when I take the test, then I'm likely to be scored as very introverted. If I just got home from a fun night out, it'd be the exact opposite. The same applies to seasonal issues. I'm far less active in the extremes of winter and summer than during the autumn and spring, because I don't particularly enjoy leaving the comfort of my home to wander out in the elements. I'd much rather sit at home and curl up with a book or veg out in front of the television. But during pleasant weather, I'd rather be out at a festival, street fair, or market.

So, which is the real me? All of the above, of course. The problem with psychological tests like this are that they categorize something which, by its very nature, can't be categorized. Some people do fit nicely into the little round holes, but the vast majority have ever-changing personality traits, depending on mood and circumstance.

That's what I think, anyway.

Bah, weird. I'm sure I picked 4 as my answer to question 3, but it registered me picking 5. Oh well.

Haven't tested in years, so I don't know if it still fits me. Hrm.

I think the problem with the online tests is that they're a) short (I think the actual version of the MBTI is a lot longer, and administered under different conditions), and b) the descriptions invite a lot of subjective validation. Kinda like astrology, as ashley_y was saying.

So if a psychologist trained in Jungian archetype theory was administering the actual version of the test, and then was actually discussing the pattern that my results fell into with me in terms of how I see the world, and allowing me to respond to ways in which the box was not fitting (like being able to say "I would have answered 'both' or 'none of the above' to this question," instead of handing me a pat profile, I'd probably find it more reliable and accurate.

Just as a side note - I've taken the long test, and I've taken various short tests. Most of them give me the same answers (ENTX), though I think some short versions are better than others.

That's not how it's done, though. The real MBTI, which I've taken a few times and even administered (psych degree here), is multiple choice by nature. It is better than the short ones for accuracy, though.

Seconds to the comments of lunameow and eeyorerin. I've taken it a couple of times and I find the same problems with it mentioned above. (Many answers which vary with mood and season and such, many answers which would have been all or none of the above, etc.) Which is why I think such tests are pretty much always utter bullshit, which is why I don't know my type (and thus didn't answer the other two questions). Yeah, they can be fun as lj memes, but I don't want to spend THAT much time just for a meme. M-B has just about that much validity for me.

This particular classification is pretty useless for me, because I'm almost dead center on all four axes. I can't be classified by it. I've taken a lot of Meyers-Briggs tests (okay, four), and all of them were consistent in their failure to classify me. I'd be an XXXX according to their classification scheme.

I suppose that's kinda cool.

I'm not sure what you mean by useful. I'm not particularly sure it was ever intended to be "useful" just data.

Data is interesting for its own sake. The knowledge has caused me to be somewhat more tolerant of other people. Yes, somewhat.