MBTI - Where Is The Proof?

[highlander]

I hear time after time that MBTI is only a set of logical groupings, it's not scientifically proven, you can't use it to dogmatically, etc. I then hear you need to consider the nuances of how the functions may be ordered, how strong they are etc.

My question is this - and I speak from the standpoint with no experience in research - why can't it be proven? At least in some form, fashion, or respect. Has nobody ever tried? Are there difficulties in running surveys or experiments to determine accuracy of the profiles? Do the studies show flaws and that it doesn't work (that we don't hear about)? There seems to be a large body of information out there, but always "no proof". Is there anything in the field of psychology that can be proven?

If there is no proof that it works, and people don't believe in it, then why are we all here?

 

[Snow Turtle]

MBTI can't be proven because it can't be empirically tested. There are many things within psychology, that while can't be 'proven' can be tested scientifically and thus become valid measuring instruments/theories.

MBTI tries to predict human behaviour based on these functions, yet there's no way to really test whether these functions exist or not (at least in a strict environment setting)... regardless I'm pretty happy with functions as descriptors

 

[Halla74]

With the right amount of electrical shocks, sleep deprivation, and sodium pentathol anything can be proven.

Do I have any volunteers?

This won't take long, I promise.

 

[Mole]

I hear time after time that MBTI is only a set of logical groupings, it's not scientifically proven, you can't use it to dogmatically, etc. I then hear you need to consider the nuances of how the functions may be ordered, how strong they are etc.

My question is this - and I speak from the standpoint with no experience in research - why can't it be proven? At least in some form, fashion, or respect. Has nobody ever tried? Are there difficulties in running surveys or experiments to determine accuracy of the profiles? Do the studies show flaws and that it doesn't work (that we don't hear about)? There seems to be a large body of information out there, but always "no proof". Is there anything in the field of psychology that can be proven?

If there is no proof that it works, and people don't believe in it, then why are we all here?

The problem is that MBTI was plagiarised from Carl Jung's book, "Personality Types".

But, as Jung himself says, his book, "Personality Types", was based on no empirical evidence.

And we are here because MBTI was taken up by the USA military during WW II. And then used by the USA military and USA business ever since to create a popular, world-wide cult.

So MBTI is essentially another business cult from the USA.

But thankfully this forum is no longer called, "MBTI Central" because the business of MBTI issued us with an Order to Desist from using their name. And so we are now called, "Typology Central".

So we remain here because this forum is tolerant and well run by the moderators, with a stable core of long term members and new members arriving all the time to keep us alive.

So we are here to enjoy one another and critique MBTI.

At least I am.

 

[ajblaise]

Actually, there have been academic studies on MBTI that show support for it. I'm not suggesting every study on it has been positive, but simply that positive scholarly reviews of MBTI exist, maybe more so than negative ones.


Study showing strong support for the construct validity of MBTI:
Construct Validity of the Myers-Briggs Type Indicator -- Thompson and Borrello 46 (3): 745 -- Educational and Psychological Measurement

Study showing strong internal consistency and test-retest reliability of MBTI:
Myers-Briggs Type Indicator Score Reliability Across: Studies a Meta-Analytic Reliability Generalization Study -- Capraro and Capraro 62 (4): 590 -- Educational and Psychological Measurement

Study concluding MBTI meets or exceeds the reliability of other psychological instruments:
bmj.com Rapid Responses for Schuwirth and Cantillon, 328 (7450) 1244

There are more, but I can't find them at the moment.

 

[Mole]

Actually, there have been academic studies on MBTI that show support for it. I'm not suggesting every study on it has been positive, but simply that positive scholarly reviews of MBTI exist, maybe more so than negative ones.


Study showing strong support for the construct validity of MBTI:
Construct Validity of the Myers-Briggs Type Indicator -- Thompson and Borrello 46 (3): 745 -- Educational and Psychological Measurement

Study showing strong internal consistency and test-retest reliability of MBTI:
Myers-Briggs Type Indicator Score Reliability Across: Studies a Meta-Analytic Reliability Generalization Study -- Capraro and Capraro 62 (4): 590 -- Educational and Psychological Measurement

Study concluding MBTI meets or exceeds the reliability of other psychological instruments:
bmj.com Rapid Responses for Schuwirth and Cantillon, 328 (7450) 1244

There are more, but I can't find them at the moment.

Unfortunately, the inconvenient truth is that every Psychology Department on any reputable University regards MBTI as having the same truth value as astrology.

And as you know, every Astronomy Department in any University regards astrology as having no truth value whatsoever.

 

[nozflubber]

There's been research done on maze performance on rats with an audience/spectators (ala the social facilitation experiments) . The resulting distribution was distingtively bi-modal which implies a natural, genetic introversion/extraversion population difference.

So as far as I'm concerned, I/E is valid enough. I would say that the rest of the scales can't be validated due to the complexity of relations in the human mind and communication issues. Not to mention if you put any human in any psych lab they often respond categorically different than they would in the "real world".

I dunno though, part of me can't ignore the possibility its all bunk other than I/E. But even if it is, its a useful communication tool. If I work as a server and you ask me "how did you fare today?", and I said "Mercury did not smile upon me today", you can tell what I mean. Even if mercury is not a valid god or a valid concept, I can still mention him in a meaningful manner.

So MBTI might just end there. It could simply be about tools useful for communicating to others. And I'm fine with that. You should be too.

 

[Mole]

It [MBTI] could simply be about tools useful for communicating to others.

A good workman needs good tools.

And in particular a good personality test needs to be valid and reliable.

But the inconvenient truth is MBTI is neither valid nor reliable.

 

[nozflubber]

Mercury and Poseidon are neither valid nor reliable(anything but), but i can still incite them in my communication to others.

 

[FemMecha]

I started a thread a while ago which went tits up that attempted to explore research for and against MBTI. I only found research against it in which the National Academy of Sciences examined the system and found it lacking in validity - the testing process was not consistent except for the E/I axis.

I am glad to see ajblaise's post in this thread. I've had trouble finding links to support MBTI in research, including on the main MBTI webpage. I look forward to reading over the links.

 

[Mole]

Mercury and Poseidon are neither valid nor reliable(anything but), but i can still incite them in my communication to others.

Not so long ago we worshipped Mercury and Poseidon as Gods.

But today no one believes in Mercury or Poseidon, and no one worships them.

Just as no one today believes in astrology or alchemy, for astrology has been replaced by astronomy and alchemy by chemistry.

Both Mercury and Poseidon, and both astrology and alchemy came to us prior to the Enlightenment and MBTI comes from the same stable.

 

[nozflubber]

True. Yet I don't have to believe in INTPs in order to use the type description to explain myself to another person familiar with what that means. It's not about belief , worship or even valid statistics, its about communication. And communication based upon a smoke and mirror system such as MBTI is better than nothing.

 

[ajblaise]

Unfortunately, the inconvenient truth is that every Psychology Department on any reputable University regards MBTI as having the same truth value as astrology.

And as you know every Astronomy Department in any University regards astrology as having no truth value whatsoever.

Concerning universities which study MBTI seriously; you are talking about the Psychology Department of places like Stanford, Texas A&M University, Oxford, Oregon State University, Rutgers. Everywhere.

And this can all be verified with a few simple Google Scholar searches.

In fact, there are whole academic journals dedicated to psychometrics, which is what MBTI falls under, like the journal of Educational and Psychological Measurement and Journal of Psychological Type. This is empirical evaluation.

Are you actually against psychometrics all together?

 

[Mole]

MBTI is a cult because it is clung to despite evidence and reason.

 

[tinkerbell]

I hear time after time that MBTI is only a set of logical groupings, it's not scientifically proven, you can't use it to dogmatically, etc. I then hear you need to consider the nuances of how the functions may be ordered, how strong they are etc.

My question is this - and I speak from the standpoint with no experience in research - why can't it be proven? At least in some form, fashion, or respect. Has nobody ever tried? Are there difficulties in running surveys or experiments to determine accuracy of the profiles? Do the studies show flaws and that it doesn't work (that we don't hear about)? There seems to be a large body of information out there, but always "no proof". Is there anything in the field of psychology that can be proven?

If there is no proof that it works, and people don't believe in it, then why are we all here?

MBTI is a segmentation - ie cuts the population up - ie classify people.

Because it's self measured people change their minds about how they deal with stuff from one day to another... hence it is less stable because of that. Self categorisation requires a person to understand dynamics of themselves that they may not really be tuned into... I know a screamingly ESTJ who doens't know if he is a J type... they guys writes war and peace instructions for travel (and we are talking short hop not major travel)..... he is just feels less constrained than the premis of Jness... (but he is SO J...)

Also I think a fair few scales are a bit wooly in terms of being truely discrinimating.... therefore you end up with a fair proportion around the boundaries of categories....

Personally speaking, I don't think the categorisation is tight enough... ie there are TOO many people wiggling around the edges...

The bit about it can't be tested/evaluated it total rubbish, it can be evaluated quantitatively, however it will shot the device is way far from perfect.

BUT a segmentation doens't need to be PERFECT to be useful..... Does it make sense, can you use it... if it works 80% of the time - it is better than a poke in the eye with a rusty nail... it is helpful...

Not everying needs to be evidenced....

 

[ajblaise]

MBTI is a cult because it is clung to despite evidence and reason.

... except all the evidence and reason I just referenced, which can all be verified. Directly contradicting your claims.

Victor, one who closes their eyes and covers their ears, doesn't see the light.

 

[Mole]

MBTI is a segmentation - ie cuts the population up - ie classify people.

Because it's self measured people change their minds about how they deal with stuff from one day to another... hence it is less stable because of that. Self categorisation requires a person to understand dynamics of themselves that they may not really be tuned into... I know a screamingly ESTJ who doens't know if he is a J type... they guys writes war and peace instructions for travel (and we are talking short hop not major travel)..... he is just feels less constrained than the premis of Jness... (but he is SO J...)

Also I think a fair few scales are a bit wooly in terms of being truely discrinimating.... therefore you end up with a fair proportion around the boundaries of categories....

Personally speaking, I don't think the categorisation is tight enough... ie there are TOO many people wiggling around the edges...

The bit about it can't be tested/evaluated it total rubbish, it can be evaluated quantitatively, however it will shot the device is way far from perfect.

BUT a segmentation doens't need to be PERFECT to be useful..... Does it make sense, can you use it... if it works 80% of the time - it is better than a poke in the eye with a rusty nail... it is helpful...

Not everying needs to be evidenced....

As lovely a person as you are, Tinker, I understand you are a self confessed astrologer and so, inadvertently, bring MBTI into well deserved disrepute.

 

[simulatedworld]

It can't be proven because there's no way to test it empirically.

The test only gives you a rough idea, but it runs into confirmation bias problems because it depends on self-report...so it only tells you how you see yourself, not how you actually are. And nobody else can take the test for you because, obviously, that can only tell you how someone else sees you.

Besides that, though, what is there to test? All it's really saying is, "People who think and behave this way, we're going to label xxxx, but people who think and behave that way, we'll call yyyy instead."

You just take the set of all possible human thoughts/opinions/behaviors and then categorize them arbitrarily. The idea that human thoughts/opinions/behaviors can be categorized is self-evident, but how we should categorize them is totally up to personal opinion and the way you conceptualize differences between people.

If we were going to prove any of this, it would require research into brain chemistry and the connections between different areas of the brain and their influence on behaviors and opinions. As far as I know, nobody has ever done any such research connecting Jung's cognitive functions to real neurochemistry.

How else might we go about "proving" typological categories? There's not even a falsifiable claim being made.

 

[tinkerbell]

As lovely a person as you are, Tinker, I understand you are a self confessed astrologer and so, inadvertently, bring MBTI into well deserved disrepute.

I'm a professional researcher of good standing and have a first class honors degree -

...........but I do understand that you tend to spout opinions without evidence at all, so far be it from me to remove you from your little fantasies...

 

[tinkerbell]

It can't be proven because there's no way to test it empirically.

.

That's not actually true......very little of this stuff can't be researched in a robust manner, however it would be very time consuming and expensive....

the research design would be a hgue sample size who are taking an MBTI test preiodically over a time span... this would provide definative meaurment and give and understanidn og the % of migration and relaibility.

From one data set you can also understand the likely reliability within that particualr sample....