Jobs expert Lily Garcia wrote in The Washington Post for January 31, 2010 that we should “skip the test if you don’t want your personality pegged at work.” She was responding to a reader whose employer was requiring employees to take the Myers-Briggs Type Indicator (MBTI), and who lamented that “I’d just rather not be defined by a label.
We agree whole-heartedly. No one should be defined by a label. The French poet Paul Valery said at one point (in French, sorry!) that “To see is to forget the name of the thing one sees.” Names are labels: Sarah Palin, Jimmy Carter, blonde, jock, politician, and automobile are names of things. Each of these names stirs up specific meanings for different people. It is the meanings of these words, and not the things/people themselves, that are the breeding ground of stereotypes and labels. To use the name and its associated meanings without attempting to get to know the thing or person is stereotyping, or labeling.
Many personality tests attempt to reduce one’s complex personality to a “manageable” set of labels, or types, or styles. The problem is that such reductions inevitably lead to oversimplifying. Allow me an example:
The MBTI assigns one of sixteen four-letter types to every individual. These four-letter types derive from the extreme scores on four scales. The MBTI pegs you as belonging at one end or the other of scales that measure sociability (Introvert or Extravert), perception (Sensing or Intuiting), judging (Thinking or Feeling), and deciding (Perceiving or Judging). There are sixteen possible combinations of the bolded initial letters of those eight terms, taking one from each pair for a total of four per “type”: ESTJ, ISTJ, and so forth.
But here is the problem, traits on these four scales are normally distributed, which means that about 1/3 of the people will score lower, 1/3 higher, and 1/3 in the midrange. The MBTI, in essence, omits the midrange.
How many individuals are likely to exhibit any one of the sixteen four-letter types? Let us take the type “ESTJ” as an example:
- 1 person out of 3 will score higher on the I/E scale, and could be considered an “E.”
- 1 out of 3 will score lower on the S/N scale, hence an “S.”
- 1 out of 3 will score lower on the T/F scale, hence a “T.”
- 1 of 3 will score higher on the P/J scale, hence a “J.”
Hence, the number of people scoring E, S, T, AND J all together will be 1/3 times 1/3 times 1/3 times 1/3, or 1/81. One person in 81 is equivalent to 1.2% of the population. So, 1.2% of the population would be predicted to fall in any one of the 16 types. By talking of only 16 types, we are describing only 19.2% (i.e., 16 x 1.2%) of the population, or 1 in 5. In essence then, MBTI type theory is relevant to roughly 20% of the population. This is precisely why so many people are uncomfortable being “labeled” as being one of the 16 types. The probability that a person has been inaccurately described is approximately 80%.
We recommend that such labels not be used in reporting trait scores. Rather, we like to say:
- Fran is very extraverted
- Henry is somewhat extraverted
- Celina is equally extraverted and introverted (ambiverted is the term we suggest to indicate this combination)
- Chou is somewhat introverted
- Marianne is very introverted
- Henry is more extraverted than Celina
- Chou is more extraverted than Marianne
- …and so forth
In fact, if the MBTI were to use five levels for each of the scales, they would have a total of 625 four-letter types. As you likely sense, that number of “types” is unmanageable. Good. Let’s just stop using types, and, instead, we can use a descriptive sentence to be accurate in describing ourselves. Thus, I would describe myself as: “Pierce is slightly more introverted than he is extraverted, highly intuitive, equally prone to thinking and feeling, and less obsessive about structure than most.”
Of course, this assumes that you want to use the MBTI language. We recommend that you move not only from typing people to describing them more fully, but from the MBTI and other outdated models to the Big Five, or the Five-Factor Model.