Why everyone’s favorite personality test is BS

Myers-Briggs tests have a persistent hold on many of us. Employers administer them to new recruits. Singles put their results in dating profiles, next to their astrological signs. And, to my dismay, the test was even featured recently in a Fast Company article about remote work and personality types.

Why, you might ask, was I dismayed by this?

For that, we have to dig into the field of personality psychology a bit more.

The Myers-Briggs Type Indicator, or MBTI, is an assessment that was developed in the early 1940s by Katharine Cook Briggs and Isabel Briggs Myers, based on the writings of Swiss psychologist Carl Jung and his discussion of personality archetypes. The assessment is based around four dimensions, and individuals taking the assessment are classified along each of those dimensions. The dimensions are:
• Extraversion (E) vs. Introversion (I)
• Sensing (S) vs. Intuition (N)
• Thinking (T) vs. Feeling (F)
• Judging (J) vs. Perceiving (P)

You have probably seen people highlight their “type” using a code like ENFP.

This all sounds good. A simple classification system for understanding people’s personality ought to be a great tool for people to understand themselves and for managers to get to know their reports. Anything that helps people recognize the ways in which their motivations differ from those of the people around them should be useful in resolving conflicts and influencing behavior.

Unfortunately, the MBTI is a bad tool in ways that are likely to be misleading.

The discipline of personality psychology is a real success story in the field of psychology. Constructs like the Big Five personality characteristics (Openness to experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism) have been measured, replicated, and studied extensively for decades. Starting with observations about individual differences, the field has also made strides toward understanding the motivational basis of these constructs.

There are a few key lessons from personality psychology that undermine the MBTI. First, a good personality inventory needs to have test-retest reliability. That is, when you give an inventory to the same person over time, you’d like to get similar results. The Big Five (for example) is very stable across the lifespan, even though it drifts a bit over time. In contrast, the various forms used to measure the MBTI vary in their reliability, but for many of them, the reliability is quite low.

An even more serious limitation involves the classification system used by the MBTI. Extensive research on the Big Five personality characteristics shows that when you look at the distribution of scores across people, the distribution is well-behaved and centered around the middle of the scale. Well-behaved means that most observations are in the middle with fewer away from the mean. That is, most people’s scores on the Big Five dimensions fall toward the center of the dimension. That is likely to be true for any broad measure of basic personality traits. Yet, the MBTI creates a forced classification of people (which is why people can give a four-letter personality type for themselves).

The reason this matters is that most people have a mix of the motivations that underlie various personality traits. For example, agreeableness reflects people’s motivation to get along with others in interpersonal situations. Most people have some settings in which they are strongly motivated in this way (and others in which they are not). However, a personality system like the MBTI that classifies people will force them into a category, even if they are really a mix of more than one. That will give people a false sense of how extreme they are along that dimension, which can cause people to miss subtle deviations in people’s behavior from the expectations of members of that category.

As wonderful as personality psychology is—and as valuable as strong tools about personality can be—personality traits are not destiny. Situations also exert a strong influence on people’s behavior. Even a highly conscientious person may be a total slacker when on vacation. Even the most extraverted person will sit quietly and listen when the head of their company is giving a talk. Even the best measurements of personality characteristics predict only about 20% to 30% of the difference in behaviors among people. It is helpful to understand the influence of personality on the behavior of people in the workplace, but also important to pay attention to how the people around you react in different contexts.

In the face of these problems, why has the MBTI persisted? Part of the reason is its simplicity. There are only four dimensions and sixteen combinations of the dimension values. That makes it easy for people to feel like they have a handle on how to treat others. The descriptions of the dimensions are vague. As a result, it is easy to see the categorization you get from taking the MBTI as reflecting aspects of yourself. The ways in which you don’t fit the mold will not be as obvious at first, and so the categorization you get from the test is likely to resonate with you when you hear it.

Another part of the problem is that many consultants have been trained on how to administer and talk about the MBTI. But academic psychologists have not been as likely to go into corporate consulting. As a result, many people in businesses are exposed to the MBTI without knowing its limitations.

Given all of the limitations of the MBTI, I hope companies stop using it, and publications stop promoting it. Companies do not use bad tools to do finance and accounting. They do not use bad tools to make market forecasts. The behavior of people in organizations is the single most important factor that will drive future success. We should not use bad tools to understand ourselves and the people we work with, either.