Extraverted Sensor And

Table 4.3 Eight Fundamental Preferences Measured by the Myers-Briggs Type Indicator



Draws energy from the outside; involved

Draws energy from internal world of

with people; likes action and activity

thoughts and ideas



Prefers taking in information through all

Prefers information derived from a "sixth

five senses; attends to what actually exists

sense"; notices what's possible rather than


what is

Prefers logic, organization, and clean


objective structure

Prefers a person- and value-oriented way


of processing information

Prefers living a well-ordered and controlled



Prefers to live spontaneously, with room for

flexible spur-of-the-moment activities

Sources: Myers et al. (1998); Hirsh & Kummerow (1990).

Sources: Myers et al. (1998); Hirsh & Kummerow (1990).

These eight fundamental preferences reduce to four scores—you are either extraverted OR introverted; sensing OR intuitive; thinking OR feeling; judging OR perceiving. These four scores are then combined to yield types. Indeed, each person is placed into one of the 16 types yielded by their four scores. For example, you could be an ESTP type: Extraverted, Sensing, Thinking, and Perceiving. This type, according to the MBTI authors, has a distinctive leadership style in business settings. She likes to take char ge when a crisis occurs; she' s good at persuading others to adopt her point of view; she is assertive and leads the group to the most direct route to the goal; and she wants to see immediate results.

Contrast this with another type, an INFJ: Introverted, Intuitive, Feeling, and Judging. This type, according to the authors of the instrument, has a fundamentally different leadership style. Rather than take char ge and assert, INFJs are more likely to develop a vision for the or ganization; get others to cooperate rather than demand cooperation; work to inspire others rather than command others; and work solidly and with integrity and consistency to achieve business goals. One can readily imagine that different types of business leaders would be better in dif ferent organizational settings. In a time of crisis, for example, an ESTP might be better at or ganizing others to deal with immediate threats. On a plateau in business, an INFJ might be better at pausing to reflect on a long-term vision for the o ganization.

It is estimated that over 3 million people a year take the MBTI (Gardner & Martinko, 1996). Although it was developed for applications in education, counseling, career guidance, and workplace teambuilding, it is also widely used in personnel selection settings (Pittenger, 2005). Its wide use most likely comes from its intuitive appeal; people can readily understand the relevance of the personality traits supposedly measured by this test.

There are, however, several problems with the MBTI. The first problem is tha the theory on which it is based—Jung's theory of psychological types—is not widely endorsed by academic or research-oriented psychologists. For one thing, people don' t come in "types," such as extraverted types and introverted types. Instead, most personality traits are normally distributed. Figure 4.2 illustrates the dif ference between

A. Hypothetical data on the trait of introversion-extraversion if it followed a truly type-like distribution in the population. There would be a large number of introverts, a large number of extraverts, and few people in between.

Introversion Extraversion Percentage

B. Typical data on the trait of introversion-extraversion, which follows a bell-curve or normal distribution in the population. There are a large number of people in between the relatively rare extreme introverts and extreme extraverts.

Figure 4.2

Examples illustrating what the trait of introversion-extraversion would look like in terms of distributions in the population if it followed a type model (Panel A) or a normal distribution model (Panel B). Real data support the normal distribution model, not the type model.

data that would support a type model of introversion-extraversion (called a bimodal distribution) and the real data on introversion-extraversion, which is normally distributed according to a bell-shaped curve. Very few characteristics of persons follow a typological or bimodal distribution. Biological sex is one characteristic that does conform to a bimodal distribution; there are many female-type people, as well as many male-type people, and very few people in between. The distribution of extraversion-introversion is not like this at all; it has only one peak, right in the middle, suggesting that the majority of people are neither purely introverted nor purely extraverted, but are somewhere in between. Virtually all personality traits follow this normal distribution, so the concept of personality "types" is simply not justified

One consequence of forcing a typology onto a trait that is normally distributed concerns the importance of cutof f scores for classifying people into one category or the other, e.g., as introverted or extraverted. Most users of the MBTI use the median score (the score at which 50 percent fall above and 50 percent fall below) from some standardization sample as the cutof f. The problem lies in the fact that a lar ge percentage of people in any sample will be clustered right around the median score. If that median score moves a point or two in either direction, because of dif ferences in sample characteristics used to determine the cutof f score, a very lar ge number of people will be reclassified into their opposite categor . In fact, a person with an introversion-extraversion score of 20 might be classified as an introvert in one sampl (if it had a median of 21), or classified as an extravert in another sample (if it had median of 19). So, the same individual score (a 20) will be interpreted very dif fer-ently depending on the median used to perform the cutof f for classification. Despit this problem with cutof f scores and typologies, the majority of users of the MBTI continue to follow the scoring system that classifies persons into letter categor groups, a practice that has been soundly criticized in the professional consulting literature (e.g., Pittenger, 2005).

Another related consequence of using a typology scheme for scoring the MBTI is that the scores will be unreliable. Reliability is often estimated by testing a group of people twice, separated by a period of time. With the MBTI, because cutof f scores are used to categorize people into groups, and because many people are very close to the cutoff scores, slight changes in peoples' raw scores on retesting can result in a large percentage being reclassified into di ferent personality types. Indeed, a study of the test-retest reliability of the MBTI (McCarley & Clarskadon, 1983) showed that, across a five-week test-retest interval, 50 percent of the participants received a dif ferent classification on one or more of the type categories. These results are not surprising, and this is one reason why most scientific personality psychologists do no recommend using typological scoring systems for any personality measure.

Another problem with typological scoring systems is that it assumes lar ge between-category dif ferences, and no within-category dif ferences, between people. For example, all extraverted types are assumed to be alike, and introverted types are assumed to be very dif ferent from extraverted types. This, however, is not necessarily the case. Imagine two people who score as "extraverted types," yet one of these is just one point above the median and the other is 31 points above the median. These two "extraverted types" are likely to be very dif ferent from each other (they dif fer by 30 points on the scale yet are given the same type category). Now imagine an "introverted type" who scored one point below the median, and an "extraverted type" who scored one point above the median. This "introvert" and this "extravert" are likely to be indistinguishable from each other (they dif fer by only 2 points on the scale yet are given different type categories). This is another reason why psychologists who know about measurement issues avoid using type scoring systems for any personality test.

Dozens of validity studies of the MBTI have been published, mostly relating type categories to occupational preferences. These studies have been criticized, however, because most fail to report statistical details necessary to determine if the differences are significant. For example, Gardner and Martinko (1996) review 13 studie that examined the distribution of MBTI types in managerial professions. All of these studies reported the frequencies of types in different categories yet none reported scale score means that would have allowed strong statistical tests of mean personality differences between the dif ferent managerial categories. Moreover, other recent reviewers (e.g., Hunsley , Lee, & Wood, 2003) point out that no adequate tests have been done on the predictive validity of the MBTI (e.g., that the MBTI can predict future career choices or job satisfaction). Also, virtually no studies have been done examining the incremental validity of the MBTI (e.g., whether the MBTI can add meaningfully to the prediction of career choice or job satisfaction above and beyond that obtained with more traditional personality measures). T able 4.2 lists the major elements of validity for personality tests, and we see from this discussion that the research base on the MBTI does not cover many aspects of validity . The conclusion is that the evidence for the validity and utility of the MBTI is weak at best.

Every few years psychologists take a fresh look at the evidence for the MBTI and summarize what they find. In 1991 Bjork and Druckman reviewed the evidenc and concluded: "At this time, there is not suf ficient, well-designed research to justif the use of the MBTI in career counseling programs" (p. 99). A few years later, Boyle (1995) also reviewed the literature and found no strong scientific evidence support ing the utility of the MBTI. In 2003 Hunsley , Lee, and Wood reviewed the latest evidence and summarized their findings: "One can only conclude that the MBTI i insufficient as a contemporary measure of personality" (pp. 63-64). And in an even more recent review paper, Pittenger (2005) evaluated all of the scientific literature o the MBTI and concluded that, "Using the MBTI to select employees, to assign employees to work groups or assignments, or for other forms of employment evaluation are not justified for the simple reason that there are no available data to rec ommend such decisions" (p. 219).

Given the highly negative reviews on the scientific merit of the MBTI, wh does it continue to be a hugely popular tool in consulting and career counseling? There are probably several reasons. First, the popularity of the MBTI may reflect the suc cess of the publisher' s marketing campaign. In addition, the test comes with rather simple scoring and interpretation instructions, making it usable and understandable by people without advanced training in personality psychology . Moreover, the interpretations the test of fers are readily translated into seemingly sensible predictions about work and interpersonal relations. Like the popularity of horoscopes, people like hearing about themselves and their futures, even if little or no scientific evidence exist for those descriptions and predictions.

Is there any legitimate use for the MBTI? While it should definitely not be use as the single piece of evidence on which to base employment selections or career decisions, it may have a role in such areas as team-building, career exploration, or relationship counseling. The test can get people thinking about dif ferences between people. People with vastly dif ferent personalities see the world dif ferently, and if the test fosters an appreciation for this diversity , then it may be useful. The test might also be useful if it gets people thinking about the relationship between personality and behavior. If we understand that how we act toward others, and they toward us, is influenced in part by our personalities, then this increases our ability to understan and relate well to others. For example, if a teacher takes the MBTI as part of a "teacher development workshop" he or she may think about their own teaching style, or may gain an awareness that not all pupils are alike in how they relate to teachers. The test may even act as a catalyst for group exercises or team building that foster esprit de corps among group members. For example, at a "corporate retreat" a group of managers may take the test and then explore ways that they can work better as a team given the dif ferences in their personalities. So the test may indeed have some utility for getting people to think about personality , even though the test does not appear adequate as an instrument for selection.

Was this article helpful?

0 0
Anxiety and Depression 101

Anxiety and Depression 101

Everything you ever wanted to know about. We have been discussing depression and anxiety and how different information that is out on the market only seems to target one particular cure for these two common conditions that seem to walk hand in hand.

Get My Free Ebook

Post a comment