[MBTI General] Scientific Validity of Personality Classifications and Tests

[ChildoftheProphets]

As a former NT myself, scientific validity still has a very meaningful place in my heart, and I wonder what you NTs have to say about the theories of Myers-Briggs, Keirsey, and most recently Helen Fisher (who independently recreated Keirsey's temperments on top of several nuerochemical premises).

Many say their theories lack validity when compared to the Five Factor Model, and perhaps the biggest problem is that there hasn't been enough research done using the other theories, especially Fisher's.

Any thoughts?

I personally find Fisher's model to be much more useful and explanatory than the Big 5, but I'm not entirely sure why.

 

[peterk]

Keirsey's theory of temperments only has his book explaining it as far as I know. Some other authors have jumped on his bandwagon. In "please understand me" by Keirsey there is a type test but he doesn't give any psychometric data(statistics) on it's validity or how it compares with the MBTI. The MBTI has Jung's theory behind it and a lot of exerimental data. The Five Factor Model has no theory behind it, just empirical evidence. By the way, some time ago the authors of the Big Five set out to disprove the MBTI using the Big Five test but to their suprise the Big Five test validated some of the MBTI's results. Can you give a reference to any works by Helen Fisher?

 

[ObliviousExistence]

they are all bunk, all pathetic failed attempts to create a logically coherent model out of Jungs work on psychological types. Pseudoscience under the guise of science.

 

[Orangey]

The Five Factor Model has no theory behind it, just empirical evidence.

Which is why it's one of the only ones (if not THE only one) generally considered acceptable in the academic field of psychology.

I think that, even with the FFM, there is significant controversy over what can be extrapolated from their findings. That is, I think it's not even a sure thing among psychologists that the FFM can predict behavior. The correspondence between the model and behavior is a sticky issue, and (so far as I know) personality psychologists using the FFM tend to ignore the importance of behavioral correlation, which has left them open to all kinds of criticism from those using other psychological approaches (especially those who place heavy importance on behavioral measures.) I would think that complicating things by introducing highly theoretical typologies of personality (with roots in psychoanalysis, no less) would produce no "better" results in terms of empirical validation.

I personally find Fisher's model to be much more useful and explanatory than the Big 5, but I'm not entirely sure why.

That's because the FFM doesn't really explain anything. It is simply a descriptive taxonomy that breaks human personality down into five broad traits. It's not its own, overarching theory; it doesn't explain things like a theory does. Empirical evidence gained using the FFM (for example, studies showing up to .3 correlation between traits and trait-behavior in job-performance) is sometimes instantiated in arguments for trait theory. But trait theory is separate from the FFM.

 

[peterk]

It would seem to me that a psychological test which has a theory behind it, MBTI, would generate more confidance in it than a test with no theory behind it, FFM. Doesn't the MBTI validate Jung's theory of psychological types just a little bit?

 

[Orangey]

It would seem to me that a psychological test which has a theory behind it, MBTI, would generate more confidance in it than a test with no theory behind it, FFM. Doesn't the MBTI validate Jung's theory of psychological types just a little bit?

I think "confidence" in a personality measure in psychology probably has more to do with its empirical substantiation than its theoretical background (or lack thereof.) The only reason a "theory-less," descriptive taxonomic system like the FFM is successful where stuff like MBTI is not is because, as a broad description of traits that provides no causal explanations, on its own (though keep in mind that people have tried to use FFM experiments as evidence in arguments for general trait-theory), as to the existence of traits and their relationships to one another and/or to behavior, it does not overstep its empirical bounds. It doesn't attempt to explain more than it legitimately can, and is thus okay. Of course, as such it's usefulness is limited, but that's a whole other avenue of criticism.

A commonplace MBTI claim to the effect that, for example, INTPs are less in tune with their emotions because their Fe is in the inferior position, and this causes them to appear cold or obnoxious to others is a theoretical explanation of behavior that cannot be substantiated empirically. It's like psychoanalysis; it may have some modicum of descriptive accuracy, and even potentially some predictive ability, but science, which is concerned with establishing theories that refer to an objective, measurable reality, can have nothing to say about it one way or another. We can't set up an experiment that will determine, in some way, whether it's true that Fe is really inferior for INTPs or not. It's classically unfalsifiable (though the jump from unfalsifiability to pseudo-science is not immediately warranted, as any good philosopher of science knows.) That is, unless someone comes along and finds some way of linking Jung's traits to some biological part of human brain-functioning, and offers some way of explaining their organization in individuals in a way that mirrors MBTI type organization, then MBTI is not a part of science. That's not to say that others can't use MBTI to do certain psychometric studies (i.e., correlations to job-performance, SAT performance, whatever); it's just that MBTI theory, which explains and justifies its own typological taxonomy, cannot be verified via empirical means. At least not yet. We can't "verify" whether it's really true or not that function order determines facility. Or even that functions exist or that, if they do, they go in any particular order.

Additionally, personality taxonomies or trait-measures (FFM included) are plagued with methodological problems in general. In the first place, self-report and factor analysis, the methodological staples of personality psychology, have been criticized; the inherent subjectivity of personality terms in natural language pretty much disallows for any "real-world" or "universal" reference, as people will answer self-report tests in certain ways depending upon their personal understanding of the terms (which is subject to a variety of cultural/societal/situational influences that cannot be weeded out experimentally.)

In the second place, it's logistically impossible to carry out the kind of experiements that would be necessary to measure an individual's behavior over time and across situations, which is basically the only way any meaningful relationship (or lack thereof) between (supposedly) stable character traits and behavior could be ascertained. And unless a significant correlation can be demonstrated, then ANY personality taxonomy will be limited to describing broad traits that don't correspond in trait-specific ways (or correspond only in a limited, unexplained way) to individual behaviors in individual situations. Such limitation is also a limitation to the usefulness of the model (clinical or otherwise.)

I don't know what you mean when you ask "doesn't MBTI validate Jung's theory of psychological types?" MBTI theory is different from Jung's; how could it "validate" Jung's theory if it is pretty much its own, different (albeit related, or derivative) theory? And if by "validate" you mean empirically, then MBTI isn't going to be "validating" anything anytime soon.

 

[peterk]

MBTI theory is an extension of Jung's theory: the 16 types are implicit in his theory. The fact that his theory came before the test is a powerful indication that that part of Jung's theory has some validity. The goals of the MBTI are more diffuse than most other tests,i.e., it is not as explicit about things as say the MMPI-2 and that is one of its strengths. It is a mistake to divorce MBTI theory from Analytical Psychology because they could enrich eachother. Why theorizing about types is more prevalent outside Analytical Psychology is a mystery to me. It is treated like a stepchild by the Jungians. Psychology is nowhere near to being a science. It only uses the scientific method and not until there is a mathematical theory of the mind, (and I believe there is a mind distinct from the brain but generated by it) that can make predictions and have them verified by experiment will it become a true science. In some sense Jung's theory of psychological types was a prediction and the MBTI test(and it is a test because it tests you to see what type you are) is the experiment which
has plausible results for the theory.

 

[INTP]

As a former NT myself

type doesent change, so you either never was and nt or you still are nt. cba to read more of this topic

 

[Totenkindly]

As a former NT myself, scientific validity still has a very meaningful place in my heart, and I wonder what you NTs have to say about the theories of Myers-Briggs, Keirsey, and most recently Helen Fisher (who independently recreated Keirsey's temperments on top of several nuerochemical premises).

Many say their theories lack validity when compared to the Five Factor Model, and perhaps the biggest problem is that there hasn't been enough research done using the other theories, especially Fisher's.

From what I know, Big Five / OCEAN has more corroboration statistically.

But there's tons of papers out about various tests and validation, just Google the topic and they'll pop up. I think an issue with the other tests is that they were derived more from a systematic approach to theory than derived straight from data.

 

[ChildoftheProphets]

Thank you all very much for this input--and please keep it coming!

@ Jennifer:

A week or two ago I did google this subject, but I didn't find as much as I would have liked. Still, it did lead me to Typology Central, and these replies have been way more helpful than google was!

@ the INTP from Finland:

Why can't personality type change? No matter where a person stands on the nature-nurture debate, either environmental conditions or chemical-related gene-expressions could produce changes in behavior, and thus theoretically change their personality.

(My theory on what happened to me in high school, which I cobbled together from two evolutionary psychologists' descriptions of stress-response as well as Helen Fisher's temperement theory, is that I adapted to the overpowering coercion of a stronger social group by increasing my reliance on emotional negotiation while decreasing my reliance on intellectual attack. This adoption of "tend-and-befriend" behavior lowered my testosterone levels, switching my personality from that of the NT Rational to that of the estrogen-drenched NF Idealist.)

@ Orangey and Perterk:

I hadn't really thought about that before; there's no theory at the bottom of the OCEAN! The dimensions it uses to describe people may be the most accurate, but not most useful.

Lol, maybe the Big 5 needs an abbreviation system like MBTI does: something like "OCEAN xxxxx," where each x equals a number between 1 and 3, to indicate the degree of each trait. Still, this does seem kind of cumbersome . . .

Anyway, Helen Fisher's last book, as far as I know, is Why Him? Why Her?, and she also has her own website:

Dr Helen Fisher - Biological Anthropologist - Home Page

Also, the basic premise of her personality theory is that the behaviors of Artisans, Guardians, Rationals, and Idealists (whom she calls Explorers, Builders, Directors, and Negotiators) are based on the dopamine, serotonin, testosterone, and estrogen systems, respectively.

Based on what you guys have said, she may have more empirical evidence than Kiersey does, although still not enough to make SciAm Mind happy (I read an article last year that critiqued her findings).

@ ObliviousExistence (or anyone who wants to comment):

What would you suggest be done to make MBTI and other temperment theories more scientific?

 

[Halla74]

@ the INTP from Finland:

Why can't personality type change? No matter where a person stands on the nature-nurture debate, either environmental conditions or chemical-related gene-expressions could produce changes in behavior, and thus theoretically change their personality.

(My theory on what happened to me in high school, which I cobbled together from two evolutionary psychologists' descriptions of stress-response as well as Helen Fisher's temperement theory, is that I adapted to the overpowering coercion of a stronger social group by increasing my reliance on emotional negotiation while decreasing my reliance on intellectual attack. This adoption of "tend-and-befriend" behavior lowered my testosterone levels, switching my personality from that of the NT Rational to that of the estrogen-drenched NF Idealist.)

I'm no endocrinologist, but it seems to me it would be very difficult to change the amount of testosterone your body is predisposed to manufacture, just as it would be to change the amount of insulin, T3, T4, or growth hormone produced by your body. Those events are genetically coded.

You can take medications to change your hormone levels, obviously, but I don't know that I've ever heard of psychological events changing endocrine system activity, especially in consideration of changing one's personality type. Hmmmmm.... <Thinking Out Loud> Prolonged depression might so long as it exists, maybe eh?

 

[INTP]

Why can't personality type change? No matter where a person stands on the nature-nurture debate, either environmental conditions or chemical-related gene-expressions could produce changes in behavior, and thus theoretically change their personality.

(My theory on what happened to me in high school, which I cobbled together from two evolutionary psychologists' descriptions of stress-response as well as Helen Fisher's temperement theory, is that I adapted to the overpowering coercion of a stronger social group by increasing my reliance on emotional negotiation while decreasing my reliance on intellectual attack. This adoption of "tend-and-befriend" behavior lowered my testosterone levels, switching my personality from that of the NT Rational to that of the estrogen-drenched NF Idealist.)

even if your behavior changes that doesent change your type, you can learn to use different functions on situations that doesent come naturally from you. this can develop your type in a way that you can use some functions easier than you used to. for example intj can learn to use Fe if he uses it enough and even if he would notice that most situations involving other people go better with Fe and Te and he would start using Fe more when handling with people, that wouldnt mean that he would change to infj, it would just mean that he has developed his intj personality.

maybe some heavy realization with hallucinogens or long time depression or something like that could change your type, but personally i think that you need to lose your mind big time to change your type

 

[Halla74]

maybe some heavy realization with hallucinogens or long time depression or something like that could change your type, but personally i think that you need to lose your mind to change your type

I think its necessary to lose your mind in order to survive all the shit life throws at you. Seriously.

Lose your mind to save it! Regardless of your type!

 

[tcda]

even if your behavior changes that doesent change your type, you can learn to use different functions on situations that doesent come naturally from you. this can develop your type in a way that you can use some functions easier than you used to. for example intj can learn to use Fe if he uses it enough and even if he would notice that most situations involving other people go better with Fe and Te and he would start using Fe more when handling with people, that wouldnt mean that he would change to infj, it would just mean that he has developed his intj personality.

maybe some heavy realization with hallucinogens or long time depression or something like that could change your type, but personally i think that you need to lose your mind big time to change your type

I enjoy the irony of this thread: it was started to point out the fact that there's no hard scientific evidence for MBTI as a whole, and yet it's ended with the assertion of undisputable veracity of obscure aspects of MBTI which are disputed even within MBTI.

 

[ChildoftheProphets]

I enjoy the irony of this thread: it was started to point out the fact that there's no hard scientific evidence for MBTI as a whole, and yet it's ended with the assertion of undisputable veracity of obscure aspects of MBTI which are disputed even within MBTI.

Thanks TCDA--I'm laughing at the irony too, but I hope this thread is far from over!

even if your behavior changes that doesent change your type

But isn't type determined by asking people about their behavior? It can only follow then, that if a person's normal pattern of behavior completely changes, by definition his type changes too.

I'm no endocrinologist, but it seems to me it would be very difficult to change the amount of testosterone your body is predisposed to manufacture, just as it would be to change the amount of insulin, T3, T4, or growth hormone produced by your body. Those events are genetically coded.

You can take medications to change your hormone levels, obviously, but I don't know that I've ever heard of psychological events changing endocrine system activity, especially in consideration of changing one's personality type. Hmmmmm.... <Thinking Out Loud> Prolonged depression might so long as it exists, maybe eh?

You should read The Brain that Changes Itself by Norman Doidge; he talks a lot about research in neuroscience, and how every change in behavior is linked to a change in neuronal cell networks in the brain. And of course, the brain controls a lot of our hormone levels, while the electrical transmissions between neurons themselves are only possible due to nuerotransmitters such as dopamine and serotonin.

I once read an article about physical abuse in early childhood creating staggering increases in adrenaline, which in turn poisoned their fight-or-flight response by permanently activitating the genes which trigger adrenaline. Soldiers experiencing the trauma of war have the same epidemiology.

PTSD can be treated with beta blockers that negate adrenaline's influence on the body, or with other medications that increase serotonin levels, which are also supposed to have a calming effect. Perhaps in the future doctors will just repair the damaged genes themselves.

I think its necessary to lose your mind in order to survive all the shit life throws at you. Seriously.

Lose your mind to save it! Regardless of your type!

Wow, lol, I'm not sure what to make of that!

 

[ChildoftheProphets]

Also, it's interesting that Fisher says Negotiators are linked to estrogen, and past knowledge says estrogen is linked to femininity, since from biology we know that testosterone and estrogen levels affect vocal chord thickness. I've sung in choirs on and off throughout my life, and over the years I've noticed that sopranos tend to be more feminine in their dress and mannerisms than altos.

I myself am a tenor, and last year I read that the average male voice speaks at around 110 Hertz--or A2 on the standard musical scale--yet my voice seems to speak around D3, which is about 147 Hz. The average female voice speaks around 220 Hz, which is A3.

This is biological proof that I have more estrogen inside me then the average male, and good inductive evidence to support Fisher's hypothesis, since on personality tests I come out as an NF, Idealist, or Negotiator (which all roughly share the same traits).

 

[Orangey]

But isn't type determined by asking people about their behavior? It can only follow then, that if a person's normal pattern of behavior completely changes, by definition his type changes too.

Type is most definitely not determined by behavior. That's why it's not an empirically adequate theory; MBTI has very questionable bearing on behavior, either as a way to explain observed behaviors or to predict future behavior. Questions on an MBTI test that ask about your perception of your behavior hardly qualify as a real way of measuring your actual behavior. In any case, I don't think MBTI even purports to be a mostly (much less strictly) behavioral theory.

 

[SolitaryWalker]

Type is most definitely not determined by behavior. That's why it's not an empirically adequate theory; MBTI has very questionable bearing on behavior, either as a way to explain observed behaviors or to predict future behavior. Questions on an MBTI test that ask about your perception of your behavior hardly qualify as a real way of measuring your actual behavior. In any case, I don't think MBTI even purports to be a mostly (much less strictly) behavioral theory.

MBTI is indeed an empirical theory, though an incoherent one. As framed by Keirsey and popular folk typology authors, it is merely a collection of behavioral traits people routinely seem to exhibit. Jungian typology or the study of temperament better describe the enterprise that you seem to have in mind.

 

[Orangey]

MBTI is indeed an empirical theory, though an incoherent one. As framed by Keirsey and popular folk typology authors, it is merely a collection of behavioral traits people routinely seem to exhibit. Jungian typology or the study of temperament better describe the enterprise that you seem to have in mind.

What do you mean when you say that MBTI is an "empirical theory?" I'm pretty sure that we're both aware that MBTI doesn't qualify as a scientific theory under any interpretation of the phrase, so I assume that you did not mean to equate "empirical theory" with scientific theory. But if MBTI is not a scientific theory, then in what sense can we call it an "empirical theory?" Certainly it has no empirical validity. I'll continue on my assumption streak and, well, assume that by calling MBTI "empirical" and a "theory" simultaneously, you did not mean to make reference to its (miniscule) level of empirical support.

Or do you simply mean to say that MBTI is an "empirical theory" insofar as it is putatively (at least in the "folk typological" formulations) derived from someone's observations and subsequent classification of concrete behaviors? If that's the case (and it had better be, since I seem now to have exhausted my interpretive powers), I don't see how anything I said in my previous post contradicts what you're saying here.

Namely, I said that:

1. Your MBTI type code is not determined by behavior.
This is true. Your MBTI type code is determined by self-report of your own behavior, not your actual behavior. The difference is huge.
2. MBTI does not accurately explain the behavior that we observe.
This is true as well. MBTI has been found to fail common statistical tests of validity. And that's not to mention its almost complete lack of empirical support.
3. MBTI cannot accurately predict behavior across time and varying situations.
Again, MBTI is not valid. It has not been found to measure anything useful and is consequently piss poor at predicting real behavior.

My post was in response to ChildoftheProphets, who suggested that personality instability is the logical consequences of MBTI because MBTI tests determine people's types purely from behavior. In other words, if someone is an INTP because they behave as an INTP (per the behaviors prescribed by the theory), and they start acting like an ESFP one day, then they are no longer "really" INTP. They become ESFP. Personality is both unstable and indeterminate because people can change their behavior at any time.

Keep in mind that I am taking "behavior" here to mean "observable, measurable displays of behavior." I think it's appropriate, too, considering that the type of behavior ChildoftheProphets is talking about is the type that others would notice as un-true-to-type.

I decided to take this up by pointing out that a person's MBTI type is not in fact determined by behavior. It's determined by your subjective appreciation of your own behavior on a self-report personality inventory that you take on a particular day, under particular circumstances. As such, the type that you turn up with is in the most literal sense NOT determined by your behavior. Changing your behavior, then, will not result in a metamorphasis into a completely different personality type because the MBTI test did not determine your type by analysis of your behaviors.

 

[SolitaryWalker]

What do you mean when you say that MBTI is an "empirical theory?".

Generally, I take a theory to be any coherent set of propositions about the world. Since I've deemed MBTI incoherent, I cannot regard it as theoretical. As far as what empirical meant, I certainly had no intention of suggesting that MBTI can pass any standard of scientific investigation. One of your guesses regarding what I meant by that term was correct.

Or do you simply mean to say that MBTI is an "empirical theory" insofar as it is putatively (at least in the "folk typological" formulations) derived from someone's observations and subsequent classification of concrete behaviors??".

Precisely, MBTI is an arbitrary collection of extremely vague assertions about the behaviors of people and how they define their characters.

Or If that's the case (and it had better be, since I seem now to have exhausted my interpretive powers), I don't see how anything I said in my previous post contradicts what you're saying here. ??".

There is no contradiction between our views, my purpose was merely to make an addition to your post rather than to refute one of your claims. I intend to draw a sharper distinction between MBTI and a study of temperament. This distinction is relevant because your previous conversation was about how a person's identity may be influenced by MBTI. However, at first, I think I've misinterpreted your post as suggesting that the study of MBTI was non-empirical because it dealt with a study of temperament. Yet, in your last post you've clarified that it is non-empirical for a different reason, because it cannot be supported by empirical evidence.

My post was in response to ChildoftheProphets, who suggested that personality instability is the logical consequences of MBTI because MBTI tests determine people's types purely from behavior. In other words, if someone is an INTP because they behave as an INTP (per the behaviors prescribed by the theory), and they start acting like an ESFP one day, then they are no longer "really" INTP. They become ESFP. Personality is both unstable and indeterminate because people can change their behavior at any time. ??".

I am puzzled by just one thing: why you even bothered responding to that.

Keep in mind that I am taking "behavior" here to mean "observable, measurable displays of behavior." I think it's appropriate, too, considering that the type of behavior ChildoftheProphets is talking about is the type that others would notice as un-true-to-type.

I decided to take this up by pointing out that a person's MBTI type is not in fact determined by behavior. It's determined by your subjective appreciation of your own behavior on a self-report personality inventory that you take on a particular day, under particular circumstances.??".

That determines the four letter code you'll receive as a result of your MBTI test, but not the nature of your temperament. Altogether, I insist on a sharp separation between MBTI and the study of temperament. To be clear, I am not claiming that you do not recognize this difference, but merely that it has not been clearly expressed in your post. Since the above discussion was about how MBTI impacts a person's identity, it is quite relevant. By separating MBTI from the study of temperament, we effectively show that a person's four letter code has little to do with their identity, temperament or solidified habits of thought and action.