I am a big fan of the Myers-Briggs Type Indicator (MBTI), and its offspring, the Kiersey Temperament Sorter (KTS). Most of my knowledge of these comes from the latter, since I have a copy of David Kiersey & Marilyn Bates “Please Understand Me”, in which the KTS test is found, and an explanation of the theories behind it. I actually used “Please Understand Me” to create KTS results for my three central characters in my novel (and found myself struggling when a 4th character turned up for whom I hadn’t done the test).
However, a lot of psychologists who like to imagine they are “proper” scientists, are not. Every so often, I will encounter a piece stating their case against these personality tests, and dismissing them. Last night, one such dismissal crossed my twitter feed, and I thought I should finally set out why they are flat-out wrong.
Myers-Briggs is based on the ideas of Carl Jung, whose thinking has always run somewhat against the grain: first a follower of Freud, then a rebel, and ploughing furrows far afield of what others were doing at the time, he doesn’t seem to fit properly in academia, and yet he brought an academic approach. It’s no wonder people who think of themselves as scientists don’t like him! He did not limit his beliefs or his ideas and research to the physical, tangible, natural, world that scientists like: in the linked piece he is dismissed as an, “Early 20th century thinker who believed in things like ESP and the collective unconscious.”
The criticisms levelled at MBTI are:
- It has no predictive value.
- It creates false binary designations
- not a bimodal distribution
- inconsistent over time
- The Forer Effect
I intend to show that these criticisms come not from any flaw with MBTI and KTS, but rather, from a mode of thinking that is intrinsically linked to capitalist exploitation and Patriarchy. They are, above all, a way of psychologists making themselves feel clever, and superior to others.
This is the oddest criticism, because it’s based pretty much on what Jung (he’s actually quoted on this point, several times, in the linked article above), and Kiersey/Bates (and therefore presumably Myers-Briggs, in between), say about the traits. The so-called criticism is that most people are somewhere in the middle, and only a few people at either end.
What we’re seeing with this criticism is a classic case of misunderstanding the purpose. I don’t believe anyone in the chain of ideas, ever said, “We must divide people up into binary categories”: they all said we can understand people as being at some point on a spectrum between two extremes.
This effectively nullifies the point that it’s not a bimodal distribution: we’re not trying to find one in the first place!
It also explains why it seems inconsistent over time. For myself, I know I am strongly Introverted and strongly iNtuitive, but somewhere in the middle on the other two scales (Thinking/Feeling and Judging/Perceiving). It only takes a small change to slip me from J into P or vice versa, for example. If you interpret the test as a binary rather than a spectrum, yes, that looks inconsistent. But if you sample over time, you’ll find that I am “in the middle” a lot.
To claim that it’s inconsistent is also a false claim purely on an understanding of statistical sampling. The most relevant example I can think of that’s widely understood, is an IQ test. An IQ test is a statistical sampling of a person’s “intelligence” (although it’s arguable that IQ tests don’t really test intelligence at all). They generally have a ±3% error margin. If you scored 101, then it’s possible you’re really “below average” intelligence (98 IQ) when you look above average. (You could also be even higher above average!) And oh look, it’s another false binary of “above” or “below” average. Furthermore, if you took another IQ test on a comparable scale, you would expect to come out at a different score within that ±3% error range.
Because the KTS and MBTI have fewer questions per category, the error margin is correspondingly broader; for someone in the middle, only one or two answers on the other side changes the apparent outcome.
The criticisms are not relevant to MBTI, but rather, to the ways in which, according to the linked articles, corporations use them. Which leads on to the next point, where I expose the capitalist, Patriarchal thinking behind the criticisms:
No predictive value
“Another indicator that the Myers-Briggs is inaccurate is that several different analyses have shown it’s not particularly effective at predicting people’s success at different jobs.”
And there you have it. To a modern psychologist (imagining himself to be a “proper” scientist), what matters is how useful something is at predicting something else. And specifically, how effective it is at slotting people into employment.
(To be fair, the linked article above makes it clear that MBTI is marketed for the same purpose by a corporation in the US. This just goes to show that capitalism rewards the unscrupulous.)
The criticism comes from a mode of thinking that believes (a) that the value of a human being is in their usefulness to business, and (b) people claiming expertise have a right to determine others’ existence and reality. It also epitomises the White male establishment notion of “rationality”.
The fundamental error of the criticism is to focus on what people do, when the focus of Jung’s work, and of MBTI, and KTS, is on why people do things, and on understanding the inner world. To people imagining themselves to be “proper” scientists, to focus on the “why” is unthinkable. But to the rest of us, particularly those who are curious about ourselves, the “why” is the whole point.
Of course knowing why someone behaves the way they do doesn’t give good predictive results of the sort they want: people can do the same thing for very different reasons, or do different things for very similar reasons (as anyone who spends any time talking to kinky people would know, if they aren’t trying to find the One True Cause Of Kink). What it does do is give that person power over their own choices, and help them navigate the world in a more effective and self-fulfilling way. It’s about their outcomes, rather than business outcomes or “science”.
A quote in the article from Adam Grant says, “The characteristics measured by the test have almost no predictive power on how happy you’ll be in a situation, how you’ll perform at your job, or how happy you’ll be in your marriage.”
This is the problem. Appropriate use of MBTI or KTS is not to predict those things, but rather, to enhance them. That’s why the Kiersey/Bates book is called “Please Understand Me” and not “Please Predict My Outcomes”. I don’t want to know how happy I’ll be: I want to know how I can be happy, and why I’ll be happy. In a job, knowing what my motivations and tendencies are, I can adjust my environment to maximise my effectiveness and my happiness in the role. Knowing why I do things and how I relate emotionally to others, I can be more aware of my actions and make my partner more aware of what is important to me, and thus negotiate a happier marriage.
The supposed weakness of MBTI for a psychologist is that it gives power and agency to the person taking the test, to affect their own outcomes and make sense of their life, rather than placing power in the hands of the person interpreting the results.
(Again, noting that the corporation that owns the rights to MBTI sells it as a way of giving power to employers over their workers. I would consider this to be an inappropriate and unscrupulous use of the model.)
The Forer Effect
Any time you get a personality test you don’t like, you compare it to astrology:
If the test gives people such inaccurate results, why do so many still put stock in it? One reason is that the flattering, vague descriptions for many of the types have huge amounts of overlap — so many people could fit into several of them.
This is called the Forer effect, and is a technique long used by purveyors of astrology, fortune-telling, and other sorts of pseudoscience to persuade people they have accurate information about them.
This is daft. Of course there’s overlap. On a four-scale classification like MBTI, then any classification will have four other classifications with which it shares 75% of the traits, and another six with which it shares 50% of the traits. Once again, people using these tests (and I’m working here from the KTS, because that’s what I have the book for) are advised to read around the type they’ve been identified with to see which of several is most like them, because it’s not a binary.
Maybe I’m unusual (no, no maybe about it – but it’s not proved for the following point) but for me the interesting thing about any result is the ways in which I feel I differ from the answer given. Strictly speaking, the Forer Effect is that any outcome would sound plausible to the listener; that isn’t the case with MBTI or KTS in my experience. And I suspect most people feel a particular type or small subset of types is the best description for them, and that others are increasingly inaccurate as you move away from them.
It is perfectly possible to give a vague, flattering description, and have people disagree with it. If someone says to me, “You’re a bold, outgoing, happy-go-lucky person who has lots of friends,” one might think that’s going to apply to a lot of people, it’s flattering, and it’s vague. But it is most certainly not me! In fact, “You’re a timid, reserved, cautious person who doesn’t have many friends,” is a description I would be much more likely to agree with (and it sounds much less flattering). In fact, the truth is that I’m somewhere in between the two descriptions: Timid and reserved, and a few close friends, but also relatively happy-go-lucky and sometimes quite outgoing if I put my mind to it, also.
Which is why Keirsey & Bates invite their readers to read around and find the description that suits them best.
Once again, the criticism of MBTI on the grounds of the Forer Effect is actually a way of psychologists claiming power over others’ minds and lives, rather than letting them determine their own experience of the world. The Forer Effect works because you don’t know what your other options are. If you don’t know what other descriptions are available, you can’t measure how accurate the one you’ve been given is. (If you find all the other descriptions are the same, then you can legitimately cry bullshit! Or, indeed, if they all use synonyms.)
The Strength of MBTI vs “Psychologists”
As I’ve mentioned, I used the KTS rather than MBTI. My test I came out strongly Introverted and iNtuitive. (and consistently so) My results for the other categories are somewhere in the middle.
Understanding myself in this way gave me power to name what was in me, and to make sense of how the world affected me.
When I took the Five-Factor Model test (the supposedly scientific, empirical, test preferred by “proper” scientist psychologists), the corresponding factor of Extraversion, I scored low (as one might imagine an introvert would) but that gave me no clues about how to understand my world. Indeed, the Five Factor Model’s definitions often challenge or confuse matters because they don’t quite match people’s understandings of the labels. What’s more, the FFM (or OCEAN, from the initials of the five factors) is “empirical” only in that it describes broad statistical tendencies, rather than individual experience or outcomes. That’s what they mean by “predictive value”: “Is it more likely that a person with this trait will have that outcome?” It is, in fact, a “big data” approach, and I have written before about how flawed big data is when it comes to individuals. It’s built around averages, and therefore is worse than useless for those who deviate from average.
This is why the approach of Jung, and Myers-Briggs, and Kiersey/Bates, is stronger: it’s about understanding, rather than determining, what people do. Because it doesn’t look for bimodal distribution, it allows for the unusual to be as significant as the usual (if anything, the “false binary” criticism comes out of putting more focus on the unusual). Thus, it allows the individual to be an individual in hir own right.
The supposedly meaningless MBTI has a lot of meaning, it’s just that the meaning it has is contrary to the received wisdom of academia, and outside of their purview. It challenges the capitalist, Patriarchal, idea of people as being parts of a mechanism, and instead describes them as having an inner life, and as being agents in their own lives.
And that’s why I continue to love MBTI.