Are Scores on the MBTI Totally Meaningless?
Common criticisms of the MBTI are misguided.
Posted March 21, 2016 | Reviewed by Ekua Hagan
The Myers-Briggs Type Indicator (MBTI) and its spin-offs are among the most popular personality inventories in the world. The MBTI is widely used in organizational workshops to demonstrate how people with similar or different personalities interact with each other. Hundreds of thousands of people have enjoyed discovering their personality type by completing the MBTI and similar inventories on the web.
At the same time, the MBTI has been the target of extremely harsh criticism from the community of professional personality psychologists. A friend recently asked me what I thought about a recent article by Joseph Stromberg and Estelle Caswell that described the Myers-Briggs Type Indicator (MBTI) as "totally meaningless." I read the article and found that its authors cited the same complaints about the MBTI that I have heard for decades. This is what I told my friend.
As I see things, to say that the Myers-Briggs Type Indicator (MBTI) is "totally meaningless" is to exaggerate the shortcomings of the instrument and how it is used. The main complaints about the MBTI that have been lodged over the years (and are repeated in the Stromberg and Caswell article) are as follows:
- The MBTI was developed by Katharine Cook Briggs and her daughter, Isabell Briggs Myers, neither of whom had formal training in psychometrics or psychological assessment. Briggs earned a degree in agriculture, Myers in political science.
- The MBTI is based on psychoanalyst Carl Jung's theory of types. Jung is disrespected by many academic psychologists, who consider him to be a mystic without any ideas of scientific relevance.
- The MBTI sorts people into 16 type categories, but most personality psychologists agree that individual differences in personality are better described by continuous traits than discrete type categories. They note that distributions of scores on the MBTI scales are continuous, with most scores in the middle rather than piling up at the low and high end, as type theory might predict.
- Critics claim that there is no research indicating scores on the MBTI predict significant life outcomes such as job performance and satisfaction.
I have a response to each of these criticisms.
- Briggs and Myers may not have had formal training in psychological assessment, but they were highly intelligent, college-educated, observant, thoughtful, and passionate about understanding personality. Research by Ashton and Goldberg demonstrated that even individuals without formal psychological training can create personality scales that are just as valid as professionally developed scales. Imagine what two smart, highly motivated women might accomplish if they put their minds to it.
- Jung's scientifically dubious ideas about archetypes, alchemy, synchronicity, the collective unconscious, the paranormal, and so forth are irrelevant to his theory of psychological types. Jung's theory of types gave us the concepts of introversion and extraversion, which modern, scientific personality psychologists are perfectly happy to use today. While it is true that most modern personality psychologists would be afraid to conduct research based on Jung's theory of types or the MBTI, that has not always been the case. For example, Rae Carlson and Ravenna Helson (both highly respected, award-winning psychologists) have published empirical research based on Jung's theory of types in the top journal in the field.
- Type theory is a very complex subject that does not boil down to how scores are distributed and whether people fall into discrete categories. There are some extremely successful and well-supported type theories today, especially John Holland's theory of six personality-vocational types, currently the most widely used theory in vocational psychology. Even trait psychologists occasionally think in terms of types when they consider people who score at the high end of an extraversion scale as "extraverts." Types can (and often are) thought of as traits when we talk about the degree of resemblance to a type. I have written on the similarity of type and trait theory in practice in the reference below. Those who are interested in the intricacies of type theory should also read the monograph by Grant Dahlstrom.
- The psychological tendencies measured by the MBTI are not very different from four of the traits in the widely-accepted Five-Factor Model (FFM), as McCrae and Costa (1989) demonstrated. The MBTI does lack reference to the neuroticism dimension, which critics sometimes cite as a failure of the MBTI to assess "bad" traits. However, further research by Harvey, Murry, and Markham (1995) has indicated that normally unscored items on the MBTI can be scored to yield a measure of neuroticism if one desires. Given that there is an enormous amount of research indicating the impact of the five major personality factors on significant life outcomes, and the scales of the MBTI are similar to the factors of the FFM, it follows that scores on the MBTI can predict significant life outcomes.
A hard look at these common criticisms of the MBTI indicates that none of them hold up. I sometimes wonder if academics are not a little jealous of the commercial success of the MBTI and therefore look for ways to shoot it down.
This does not mean that the MBTI is beyond criticism. As I see it, the major problem with the MBTI is not with the inventory itself, but with the way it is normally scored and interpreted. No personality inventory is reliable enough to sort people into 16 type categories, which is why people can get different type profiles when they take the inventory on multiple occasions. So it does not make much sense to classify people with four-letter codes such as INFP and ESTJ and to regard these type codes as stable portraits of personality.
It would be more scientifically advisable to score the MBTI scales continuously to show people the degree to which they resemble the types. If the scales are scored as most personality trait scales are scored, the MBTI could still be useful in workshops designed to increase self-insight and insight into differences and similarities between people. One could still create group activities and dyadic exercises in which people have very similar or very different scores on the MBTI; one would simply talk about these similarities and differences as a matter of degree rather than a function of totally different types.
Scoring and interpreting the MBTI the way other personality inventories are scored and interpreted might be less fun than finding—like one's astrological sign—a single label for one's "type." All of the folklore about INFPs, ESTJs, etc. would have to be dismissed. But, in the end, the MBTI is sufficiently reliable and valid enough to be useful in a number of real-world contexts.
Ashton, S. G., and Goldberg, L. R. (1967). In response to Jackson's challenge: The comparative validity of personality scales constructed by the external (empirical) strategy and scales developed intuitively by experts, novices, and laymen. Journal of Research in Personality, 7, 1-20.
Carlson, R. (1980). Studies of Jungian typology II: Representations of the personal world. Journal of Personality and Social Psychology, 38, 801-810.
Helson, R. (1982). Critics and their texts: An approach to Jung's theory of cognition and personality. Journal of Personality and Social Psychology, 43, 409-418.
Dahlstrom, W. G. (1972). Personality systematics and the problem of types. Morristown, NJ: General Learning Press.
Johnson, J. A. (1997). Units of analysis for description and explanation in psychology. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of personality psychology (pp. 73-93). San Diego, CA: Academic Press.
Harvey, R. J., Murry, W. D., & Markham, S. E. (1995). A “Big Five” Scoring System for the Myers-Briggs Type Indicator. Paper presented at the Annual Conference of the Society for Industrial and Organizational Psychology, Orlando, FL.
McCrae, R. R., & Costa, P. T. (1989). Reinterpreting the MyersBriggs Type Indicator from the perspective of the five-factor model of personality. Journal of Personality, 57, 17-40.