Interesting Statistics

[Iximi]

From: http://asm.sagepub.com/content/3/3/225.full.pdf+html


I'm just interested in what sort of conclusions you guys think can be drawn from this data. It is not too often that I stumble upon a study like this.

 

[dala]

I wonder if the assumption that personal type is set is wrong, or whether people answer differently due to age-based cultural/social norms, different interpretations of the questions, or falsification of answers.

I find it interesting that there are more T's and J's amongst people with lower IQ's than higher ones. Low IQ is often conflated with laziness and inability to process data, but maybe it has more to do with the (in)ability to make intuitive leaps and understand symbolic information. This suggests to me (without a smidgen of proof) that these preferences are learned behaviour in order to keep up with those who do make intuitive leaps easily. For instance, the high IQ kid picks up on the concept of multiplication in relation to addition and can visualize how it relates to what he already knows (and then takes a nap), while the low IQ kid has to study hard to make that leap.

 

[a detached retina]

I'm assuming this is cross sectional, meaning there very well could have been more ISTJs in previous generations than this one.

 

[Iximi]

I'm assuming this is cross sectional, meaning there very well could have been more ISTJs in previous generations than this one.

Here is the abstract:

Abstract

To better understand the putative relationship between personality style and cognition, we assessed adolescents and adults on the Myers-Briggs Type Indicator (MBTI)and the Kaufman Adolescent and Adult intelligence Test (KAIT). The MBTI provides scores on four indexes: Extraversion-Introversion, Sensing-Intuition, Thinking-Feeling, and Judging-Perceiving. The KAIT provides Fluid, Crystallized, and Composite Intelligence Quotients.

Participants consisted of 1,297 individuals, aged 14 to 94 years, who were tested throughout the United States during the nationwide standardization of the KAIT. It was hypothesized that individuals favoring Intuition and Thinking would be more intelligent and would favor fluid over crystallized intelligence relative to those favoring Sensing and Feeling, respectively. Several multiple analyses of variance and covariance (MANOVAs and MANCOVAs) were conducted with sex, race, KAIT IQ, and KAIT Fluid and Crystallized IQ discrepancy serving as independent variables and continuous scores on the MBTI used as dependent variables.

Consistent with hypothesized relationships, people classified as Intuitive earned higher KAIT Composite IQs than those classified as Sensing. However, most other hypotheses were not supported, as the Fluid-Crystallized discrepancy was not meaningfully related to any MBTI dimension.

 

[a detached retina]

This is quite interesting. What I meant is that rather than tracking a specific group of people as they age, and giving them repeated MBTI tests, it seems like they simply gave some MBTI to old people and some to young people thereby failing to factor out generational cultural differences between, say, the 94 year olds and the 14 year olds. It may well be true that there were simply more sensors in the old days and people are going to be more intuitive in younger generations or it could be true that a single person tends to develop more sensing preference with age.

Good post. I hope I can grow out of my E and P eventually.

 

[Cogwulf]

The most interesting thing is that S/N correlates more with IQ than anything else, whereas the belief I've seen more commonly discussed is that F/T is related to IQ.

Of course, performing MBTI tests on people with low IQs could be problematic given the subjective nature of the questions.

 

[Iximi]

It may well be true that there were simply more sensors in the old days and people are going to be more intuitive in younger generations or it could be true that a single person tends to develop more sensing preference with age.

That or (N) types tend to die before they get too old

 

[BigApplePi]

It would be easy to sit here a type a critique of this test as I don't know what went on, but will say something anyway:

It is one thing to set up a test for a thousand people and look at statistics. Were there enough people tested to check this out? I'd like to see another independent test. Also there have been complaints by the Pod'Lair people that MBTI doesn't test what it's supposed to. I'd like to see them do this test. I'm sure that is not going to happen any time soon.

 

[Da Blob]

I think that this study was focused on the question of whether Intelligence was actually something that could be measured. If so, what personality factors, determine the level of intelligence (or perhaps vice versa). Apparently no causal link was found between Jungian personality traits and intelligence except a correlation between N and intelligence.

It is commonly believed that there are two types of intelligence (Fluid and Crystallized)

Fluid intelligence as measure by traditional IQ tests usually begins to decline at the age of 25 as IQ scores begin to decline. However, intellectual abilities do not, so it has been postulated that there is a second type of intelligence, crystalized. This makes sense because the IQ really has little meaning except for children, as it is a ratio of cognitive age over physical age.

For example, my IQ is 145 and I am 56 years old, so my cognitive abilities are supposedly that of a 84 year old individual (?) (I feel as though I am 84 sometimes... lol). So again, IQ scores are rather meaningless beyond adolescence...

 

[BigApplePi]

I think that this study was focused on the question of whether Intelligence was actually something that could be measured. If so, what personality factors, determine the level of intelligence (or perhaps vice versa). Apparently no causal link was found between Jungian personality traits and intelligence except a correlation between N and intelligence.

It is commonly believed that there are two types of intelligence (Fluid and Crystallized)

Fluid intelligence as measure by traditional IQ tests usually begins to decline at the age of 25 as IQ scores begin to decline. However, intellectual abilities do not, so it has been postulated that there is a second type of intelligence, crystalized. This makes sense because the IQ really has little meaning except for children, as it is a ratio of cognitive age over physical age.

For example, my IQ is 145 and I am 56 years old, so my cognitive abilities are supposedly that of a 84 year old individual (?) (I feel as though I am 84 sometimes... lol). So again, IQ scores are rather meaningless beyond adolescence...

I'm going to guess that a strong reason why I.Q. works in youth is that intelligence is highly correlated with experience. One's "intelligence experience" in youth typically increases linearly with age though it doesn't have to. After the early 20's, the linearity breaks down (because one is busy applying, not increasing intelligence?) and intelligence can go any which way but probably doesn't due to established personality.

Note one category I'm suspicious of is the black-white one. The N's and F's are roughly double for whites. If the races are innately about the same, wouldn't this be due to cultural differences? I thought that the Pod'Lair people claim MBTI is innate. Am I misreading something?

 

[Yet]

For example, my IQ is 145 and I am 56 years old, so my cognitive abilities are supposedly that of a 84 year old individual (?) (I feel as though I am 84 sometimes... lol). So again, IQ scores are rather meaningless beyond adolescence...

hi Blob ... it does not mean that your abilities are supposedly that of a 84 year old ... that is a mistake. Somehow it has became a popular explanation but is is truely nonsense.
It is not meaningles beyond adolescence either.
145 is a score which (depending on the test) can be related to a percentile of other scorers ... (a position in a normal distribution)
http://en.wikipedia.org/wiki/Normal_distribution

It is true that the percentile says a lot more than the score itself.

 

[Trebuchet]

I would have bet money against Introverts outnumbering Extraverts. I guess a lot of those apparently extraverted people are faking it.

And while I knew there was a strong gender difference in T/F, I didn't expect such a large difference in N/S or J/P.

 

[Iximi]

Note one category I'm suspicious of is the black-white one. The N's and F's are roughly double for whites. If the races are innately about the same, wouldn't this be due to cultural differences? I thought that the Pod'Lair people claim MBTI is innate. Am I misreading something?

This is what stands out to me as well. IF the data can be trusted that nugget of data is probably the most interesting part of the study. It is hard to believe that pure cultural differences would result in such a large difference in typology- if it is genetic it would suggest that African Americans and Caucasians have inherently different types of brains... never really a popular line of thought in academia.

I think to clear up this fundamental question of brain function we would need to see cross-sections of populations from around the world from a variety of cultures and ethnic groupings. Quite frankly, we can't even be sure that MBTI types even hold up in other parts of the world seeing as there isn't much information on it outside of Europe/North America.

 

[RobdoR]

To me this raises questions about the validity of the MBTI test. The age and race differences especially raise concern. If the questions have a cultural bias, then I can see how differently cultured people could answer them differently.

I've also noticed how the test is not a very good measure of one's type anyway. It's a good starting point, but most of the people I know who have gone deeper into temperament theory are not what they originally thought they were.

 

[BigApplePi]

Abstract

To better understand the putative relationship between personality style and cognition, we assessed adolescents and adults on the Myers-Briggs Type Indicator (MBTI)and the Kaufman Adolescent and Adult intelligence Test (KAIT). The MBTI provides scores on four indexes: Extraversion-Introversion, Sensing-Intuition, Thinking-Feeling, and Judging-Perceiving. The KAIT provides Fluid, Crystallized, and Composite Intelligence Quotients.

Participants consisted of 1,297 individuals, aged 14 to 94 years, who were tested throughout the United States during the nationwide standardization of the KAIT.

Oh yeah? Just how were these participants chosen? Volunteers? Don't tell me volunteers are random. If they weren't volunteers just how were they chosen? Were they forced to take the test? Paid? Are you going to tell me personality type would have no bearing on those who chose to take the test? Gimme a break.

 

[Dimensional Transition]

Oh yeah? Just how were these participants chosen? Volunteers? Don't tell me volunteers are random. If they weren't volunteers just how were they chosen? Were they forced to take the test? Paid? Are you going to tell me personality type would have no bearing on those who chose to take the test? Gimme a break.

I agree with this.

 

[BigApplePi]

Notice something happens here. Those ARE interesting statistics. They seduce us. We go ahead and speculate and create theories. We wonder why the contradictions. It's something like religious faith making me wonder why some people have a contempt for those of religious faith. I was seduced by those statistics making me no different than others. I didn't think to question how they were taken until later ... and I still don't know how they were taken, lol.

 

[Iximi]

The "Method" section of the Journal Article here: http://asm.sagepub.com/content/3/3/225.full.pdf+html

Should clear up some of your concerns.

 

[walfin]

Interesting to see S/N change with age; another validation of the Flynn effect?