Psychology, as a profession, is wary of the MBTI. Its origin–the fact that Isabel Briggs et al. did not have the background and training of academic psychology is no doubt a part of the professional skepticism about the MBTI. However, as I noted in last week’s blog, there is remarkable agreement between many parts of this measure and the most accepted scale in the academic work–the Big Five or Five Factor Model. By itself, that should have improved acceptance over time.
Unfortunately, agreement ends, however, when you move past the basic measurements and on to the interpretation of the MBTI. Here we move from the workman-like construction of the individual items and the preference scales, to the interpretation of these. And here we move from science to theory, and more specifically from theory in general to the theories of just one man–the psychoanalyst, Carl Jung. I have often thought that the one thing that a graduate degree in psychology or a related field would have added to Isabel Myers quite excellent intelligence, would have been a broader sense of the many different ideas that were and are alive in the psychology world.
As it was, Myers and Briggs both bought into the Jungian model of personality and temperament without apparent dissent. Jung believed (and fairly accurately) that the preferences and thus temperaments and types were innate. He also believed, however, that inheritance would be all or none, yes or no. Therefore, when the MBTI was developed, any score over the middle between preferences put you into that preference category. Although the 1998 manual lists a number of reasons why an individual may report their preferences inaccurately (expectations of parents, siblings, friends, authorities that you behave in certain ways, or misunderstanding the terms in some of the questions, for example) it is nevertheless assumed that even one more answer toward one preference or the other is sufficient. It is true that if the individual does not feel that a reported preference or a reported whole type is accurate, there are procedures for what is called “verifying type”. In different ways these consist of having the person read preference and/or type descriptions, trying to see which is the better fit. No doubt this is helpful, but it totally ignores the possibility that greater numbers of scores in one preference direction have meaning in themselves.
Today, most of us are used to the idea that very close scores on anything are unreliable. Polls, for example, routinely report a margin of error, reflecting all the sorts of random errors that do occur, from carelessness to misunderstanding of terms. If candidate X has a rating of 47% versus a 49% rating for candidate Y, we may be told that there is no statistically significant difference and that the difference that exists is within the margin of error. An either/or scale like Introversion or Extraversion should at least reflect this. The MBTI does have what is called a “preference clarity index” setting criteria for Slight, Moderate, Clear and Very Clear preferences. However, the manual (page 121) is emphatic that “Quantitative interpretation of MBTI results as an indication that a respondent has “more” or less of a preference is incorrect. Such a practice is the most pervasive source of misunderstanding and misinterpretation of the MBTI.” Pretty strong words! They go on to say that the preference clarity index is only useful to measure how “sure” the person is about his/her preference.
In my mind, the development of the Step II MBTI contradicts this whole position. I often wonder if there are internal differences in this whole area within the organization, or if no one really sees a contradiction here. In the Step II version each main preference area is divided into facets. For example, Sensing/Intuition contains (in that order) Concrete/Abstract, Realistic/Imaginative, Practical/Conceptual, Experiential/Theoretical, and Traditional/Original facets. It is possible to be out of preference on some facets and in preference on others. Perhaps this is looking at it too simplistically, but it would seem that a very high score on your overall Step I preference would predict that you are in preference on most or all facets, and a low score would suggest that there might be some out of preference facets. In an anecdotal way, this supports a major psychology argument–that not considering the quantity of the responses, throws away a great deal of information. (Incidentally, it is of some interest that in the Step II facets they have established a mid-zone of +/- 1 as an area of no clear preference.
Lastly, either/or doesn’t make much sense given what we know of all other measures of human abilities and characteristics, or for that matter of what we know of the brain. Intelligence is not an either/or, nor is anxiety, or depression or calmness or hyperactivity. Brain chemistry will reveal differing levels of neural transmitters or brain-active hormones, but these too come in degrees. Just looking at everyday experience there are people who are a little outgoing, a lot outgoing, or can’t sit still, can’t be quiet, out-of their-mind extraverted. The same holds for introversion.
Jungian influence extends much further in the development of what are called “type dynamics”. This is the MBTI organization’s pride and joy, and critical to a proper understanding of your raw preference scores. What is established in analysis of type dynamics is (among Sensing/Intuition, and Thinking/Feeling) your dominant preference, auxiliary preference, tertiary or third preference and inferior preference. The order of importance of your preferences is determined jointly by your E or I score and your J or P score. Let us say you are an ENFJ. You have a judging preference so you will extravert whatever you chose in the Thinking/Feeling (T/F category). What does that mean? Well it means that you will use that preference more in dealing with the outside world. Since your E/I preference was E that means you will be active mostly with the outside world so your dominant preference will be Feeling. If it had been if the E/I preference had been I, the dominant preference in this case would have been Intuition. Determining auxiliary, tertiary and inferior functions takes us a few steps farther, but the general idea is that among S, N, F, T you will set up the dominant, auxiliary, tertiary and inferior preferences. You will make maximum use of your dominant preference, on down to minimum use of your inferior preference. It sounds pretty elegant and complex. Other fascinating complications are the actions of a shadow self–the inferior function that may erupt out of control, bringing strange and disturbing thoughts and feelings, in a prolonged stressful situation.
But, type dynamics has this problem. It was incorporated into the MBTI, I would guess, out of great respect and admiration for Jung. Unlike Isabel Myers’ development of the individual items and of the preference scales, which was admirable and professional, this scheme is the product of faith. No experimental research was done to substantiate it. As early a 1965, a reviewer for the Sixth Mental Measurements Yearbook raised doubts about the validity of the type dynamics aspect of the MBTI. In 1989, McCrea and Costa, the same researchers who reported positively on the relationship between MBTI scales and the Big Five, did a research study looking specifically at whether the reported stronger preference on the MBTI would be positively related to the dominant function as calculated by type dynamics. What they found was no relationship between the strongest preference and the hypothesized dominant preference. Isabel Myers had, herself, presumed that the strongest preference would turn out to be the dominant preference most of the time, but had found this to be true in only 5o% of cases at that time, suggesting that this was essentially random.
Much more recently, Reynierse and Harker reported on a relatively exhaustive series of six experimental studies looking for support or non-support for type dynamics. They concluded that “the results of this research and other research of ours….suggest that the dominant preference is simply the independently high-value preference, particularly when that preference stands out and is markedly higher than any other contributing preference.” As the expression goes, this was music to my ears, as I have long believed that would be true. In creating our own adult and child scales (www.parentingbytemperament.com) we designed the feedback to include the information on whether the preference was slight, clear, etc., and we counsel participants to focus attention on the stronger preferences.
I suspect Jung was simply wrong about some of his most passionate ideas, and would have something different to say if he were alive today and reading today’s literature. Unfortunately a very large and powerful organization bought the farm on this long ago. Change will come slowly and that is too bad, as the MBTI has a great deal to contribute.
McCrae, Robert R & Costa, Paul T. Jr. (1989) Reinterpreting the Myers-Briggs Type Indicator from the perspective of the Five-Factor Model of personality. Journal of Personality 57 1 17-40.
Myers, Isabel Briggs, McCaulley, Mary H, et al. (1998) MBTI Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator. Palo Alto: Consulting Psychologists Press
Reynierse, James H, & Harker, John B (2008) Preference multidimensionality and the fallacy of Type Dynamics: Part 2 (Studies 4-6) Journal of Psychological Type 11 113-138.