The Myers-Briggs personality type ontology seems to be relatively common (as common as anything of this type is) but nobody ever seems concerned with the evidence behind it: It’s always presented as one of those little party games, which likely accounts for its relative popularity.
So, is there any evidence for it?
I have taken several reputable psychometrics classes in undergraduate and grad school. I have taken the real Meyers-Briggs as well as several on-line tests. They always come out the same for me (INTP) and most other people with little variation. That is an important validity measure for any psychometric test. INTP and INTJ represent a wildly disproportionate percentage of people on this board. Each of them only occur in 2 - 3% of the population. OTOH, the Meyers-Briggs tests aren’t nearly as statistically rigorous as the professionally administered IQ test. Those have some very strong math, especially statistics, to back them up although it is sometimes unclear how that translates to real-world results.
What type of evidence backing do you think it – or any personality typing system – should have? Although they all of flaws of one type or another, they are a lot more than “party games.” A lot of people take the little online tests and get inconsistent results and then claim it’s bogus. Or, they read a brief description of their type, it doesn’t quite seem to fit, and they make the same claim. A real test, administered and interpreted by a trained professional on an individual basis, can be very enlightening. One of the things the shorthand tests and descriptions gloss over, for example, is that the 4 scales are weighted differently in the different types and in combination with each other. I’m expressing this badly because it’s been a long time since I read up on it, so maybe someone with fresher knowledge will come along and mop up. But, just hearing “INTP” and reading a 2-paragraph description of that type isn’t really the whole story.
I ask here from complete ignorance: What’s the point? What benefit do test-takers get from this test? From my perspective it seems like an artificial constraint, learning what kind of pigeonhole my personality should fit into.
On the other hand, I am entirely ignorant of the science behind personality, or whether there even is any science behind it. I’d be hard pressed to tell you what a personality even is. So maybe I’m missing something obvious.
As mentioned, the reliability is rather high. The external validity, the ability to make sweeping generalizations about other aspects of someone’s life based on the results, is questionable. I have no cite, but I remember learning in a psychometrics class that Myers and Briggs had little science behind their original analysis, and their calculations were mostly arbitrary. It has since been shown to be stronger through research, and they may have changed the questions around to provide a stronger scientific basis.
To begin with, I want to know that the descriptions of the various types aren’t merely Barnum statements. (A Barnum statement is one that everyone can feel plausibly describes them, like “You enjoy your friends and treat them well.” or “There has been much trouble in your life.”)
Related to the above, it would be interesting to know whether there have been any studies regarding the correlation between the types and other aspects of a person’s life. (This is probably an open invitation to an endless anecdotal thread. We already have one of those in IMHO.)
Meyers-Brigg correlates pretty well with parts of the five factor model, which is the gold standard in the science of personality. That doesn’t lend any validity to some of what’s done with it.
Well, this part at least is certainly satisfied. For example, I know for sure that I am not one of the Extroverted types, and indeed every single time I take the MB test I come out strongly Introverted.
My wife thinks much the same way. I use it to make excuses for the way I act. I take them to find out what my preferences are and then take action to go outside that comfort zone. These tests also tell you how you will react in a given situation. As an I, if I have people visiting the house for a long stay, I’ll give myself room during that time to recharge myself. I might get up earlier to have some me time, for example. Not knowing why I react a certain way doesn’t allow me to take steps to work around my limitations, or to overcome them.
So it correlates well with another test of the same kind that was developed independently. That is kind of impressive.
However, that doesn’t answer the question of whether it correlates to anything outside the world of personality testing. IQ tests, as flawed as they are, do seem to correlate with things outside the realm of intelligence tests, such as job performance and (negatively) criminality.
What, in the real world, do the tests we’re discussing correlate with, and how well do they correlate?
The result you get from a Myers-Briggs ‘analysis’ tells you the result you get from a Myers-Briggs ‘analysis’. That’s it. There is no good reason or good evidence to believe anything else.
Many people want to believe there’s more to it. Just to take one example, recruitment is expensive and exhausting, and the stakes are high (hiring people who turn out to be usless is very costly for a company, and for the person who did the hiring). So anything that seems scientific and helpful may be seized upon as a useful tool, at the very least. Some companies use astrology or graphology for the same reason. And of course there’s always someone willing to cater for the demand, and make an easy buck in the process.
I knew someone who has all the relevant academic qualifications (in psychology and occupational psychology). He devises psychometric and personality tests, teaches people how to use and apply them and acts as a consultant to companies that use them. He may fairly be said to be one of the country’s leading experts in the field, and he gets well paid for this expertise.
He has cheerfully admitted to me that there is not one shred of reputable empirical evidence to suggest that the test tell you anything except the result of the test. I have also challenged him, in a friendly way, to come up with any evidence whatsoever that administering these tests is anything other than a complete waste of time (e.g. statistical data confirming that the companies that use such tools hire better people, achieve greater success or have better staff retention) than those that do not. I extended this friendly challenge two years ago, and he accepted it equally cheerfully. I’m still waiting.
What are you basing this on?
I agree, btw, that repeatability is an important factor in determining whether you are measuring something real; however, isn’t the Meyers-Briggs test always the same battery of questions?
And another important indicator of a valid measure - are the tests predictive of anything?
But then, what is the point of the test? I can extract the same information by asking you if you are intro- or extroverted; the point of interest, to me, seems more whether or not the test actually does something beyond asking those questions (in somewhat obfuscated form).
There is a poll about it right now in IMHO. There have been several others throughout the years. Some people have taken the real test and some just the online tests but INTP and INTJ are always overrepresented here. Those types aren’t common in the general population.
Do they have to be predictive of anything? Seems like they can be useful without being predictive in the same way the physical sciences are (which, of course, no model of the brain except for each individuals brain itself can accurately predict the outcome from any given input, so we shouldn’t compare it to predicting the path of an electron).
It could improve social interaction in some circumstances by understanding the differences between your own and another persons general tendencies.
Like just about any substantive question in the social sciences, there is a lot of empirical evidence on both sides of the discussion and social scientists make a living arguing the validity of the evidence with each other. Here are a number of citations and abstracts (gathered from a brief perusal of Google Scholar) to illustrate that there is a substantial amount of applied research in this area.
The association between cognitive style and accounting students’ preference for cooperative learning: an empirical investigation, Alan Ramsaya, Dean Hanlonb and David Smith, Journal of Accounting Education Volume 18, Issue 3, Summer 2000, Pages 215-228
This paper investigates the use of cooperative learning in accounting education, in particular, the issue of whether an individual’s cognitive style impacts upon their preference for cooperative learning techniques. The study extends the existing literature by using the Myers-Briggs Type Indicator to examine the association between all four dimensions of cognitive style and accounting students’ preference for cooperative learning. Results show that preference for cooperative learning is significantly associated with the extroversion/introversion dimension [a result which extends the findings of Hutchinson, M., & Gul, F. (1997). The interactive effects of extroversion/introversion traits and collectivism/ individualism cultural beliefs on student group learning preferences, and with the thinking/feeling dimension, which is a new result in the literature.
Journal of Personality Assessment, Recent Assessments of the Myers-Briggs Type Indicator, John G. Carlson, August 1985
Abstract: The present paper focuses on approximately two dozen recent published studies that examined reliability and validity of the Myers-Briggs Type Indicator (MBTI) in clinical, counseling, and research settings. Several assessments of split-half and test-retest reliability of the standard Form E and shorter Form G of the Inventory have Yielded generally satisfactory correlations
far all four scales. A larger number of studies of construct validity of the MBTI have
yielded support for research hypotheses in situations ranging from correlations of the MBTI with a personality inventory, to couples problems in a counseling setting, to line judgment in groups, and others, Therefore, the applications of the MBTl have been broad, although somewhat unsystematic, and with generally favorable validity assessment. Continued attempts to validate the instrument in a variety of settings are needed.
Journal of Accounting Education, 21:2 (2003) Personality preferences of accounting students: a longitudinal case study Stacy E. Kovar, , Richard L. Ott and Dann G. Fisher.
As the skills needed to succeed in accounting broaden, accounting programs must develop recruiting strategies to attract and retain individuals with a broader range of personality characteristics. As a form of recruitment evaluation, this study uses the Myers–Briggs Type Indicator (MBTI) to examine the characteristics of accounting students recruited into and retained by one of the original Accounting Education Change Commission (AECC) grant programs. Results indicate that, contrary to expectations, the personality types attracted to and retained in the program have not become more diverse over the course of 8 years. These results may imply that certain requisite preferences are needed to succeed in accounting or that homogeneity has benefits not previously considered. They certainly suggest a need to further refine recruiting processes, to focus on broader perceptions of the profession, to evaluate faculty’s own preferences and biases and/or to develop curriculum to help students learn to utilize their less preferred personality traits.
Patrick Wheeler, The Myers-Briggs Type Indicator and Applications to Accounting Education and Research Issues in Accounting Education 16 (1), 125 (2001);
Many of the challenges facing the accounting profession involve personality characteristics of accountants. Personality theories have generated a rich research stream outside accounting and are widely applied in other disciplines and professions. Yet little research using these theories has been done in accounting. To redress this imbalance, this paper examines Jungian personality-type psychology—one of the main personality theories—and the major psychometric instrument that has arisen from it—the Myers-Briggs Type Indicator (MBTI). The MBTI has been extensively tested for reliability and validity, and used in a large number of basic and education research studies. Results from reliability and validity testing indicate that the MBTI reliably measures personality characteristics predicted by Jungian theory. A small amount of published research has been conducted in accounting using the MBTI. These 16 articles are reviewed, with suggestions for additional research.
@Half Man Half Wit - Why is it better to have a test than just to ask someone which category they think they are in? This is a fundamental point about psychometrics. Asking the subject a single self-assessed question, particularly using psychological jargon, is about the weakest type of psychometric question imaginable. It doesn’t tend to yield stable or predictive results and is much more subject to biases, such as social desirability bias.
The first study you quote may be relevant, it’s hard to say. None of the others obviously are: which part of the above suggests to you they are?
Read the article on the five factor model. It doesn’t go into a lot of detail, but it’s clear that the five factors model something that describes real world behavior.