The Myers-Briggs Type Indicator® (MBTI®) is designed to make the theory of psychological types accessible and relevant to peoples’ lives. This framework identifies basic preferences along four different dichotomies (Extraversion-Introversion, Sensing-Intuition, Thinking-Feeling, and Judging-Perceiving), creating a total of 16 different personality types. Today, the MBTI instrument is familiar in daily life and has helped thousands of people understand and improve their professional performance, personal relationships, and so much more.
Like other tools for psychological analysis, the MBTI® has been carefully developed and refined over many years. In order to make the best use of MBTI® results, it is important to have a detailed understanding of the instrument itself. In the following discussion, we overview the theoretical foundations of the MBTI® framework, the design and development of the MBTI® instrument, administrative considerations, concerns regarding the instrument’s reliability and validity, and strategies for interpreting and applying MBTI® results in a variety of different contexts.
MBTI® Design and Development
Over the years, various aspects of the MBTI® instrument have been revised and improved in a number of ways, including but not limited to:
- Reducing the total number of items in order to increase the efficiency of the instrument
- Refining items’ wording and format to remove outdated references, archaic verbiage, and ambiguous phrasing
- Revising item weight or scoring method to increase reliability and validity
- Renewing interpretation strategies, especially for breaking ties between personality types
- Reviewing guidelines for evaluating gender differences, especially given that some items have been found to be valid for one gender but not the other
Collectively, these revisions are meant to keep the instrument current, reflect changing language use and social norms, improve item-to-scale correlations, eliminate gendered scoring, and otherwise increase the indicator’s reliability and validity. Additional efforts have been made to explore how the MBTI® can be applied in a number of different contexts, including counseling and therapy, education, professional organizations, and many more. Ultimately, the goal of these efforts is to improve the MBTI® inventory, bring it into the 21st century, and apply it to help even more people more effectively.
(MBTI® Manual. Myers et al. 1998, CPP Inc. pp. 3-33)
Collectively, the body of literature that examines the validity of the MBTI® instruments has found high validity scores, particularly in the most recent versions. There are two broad categories of evidence that demonstrate the validity of the MBTI®. The first category confirms the validity of each of the four preference scales that compose the MBTI®, and the second category of evidence concerns the validity of an individual’s MBTI® type as a whole. Each of these has several sub-categories. Let’s examine each one in turn:
Validity of the Four Preference Scales: These analyses are carefully designed to examine the various dichotomies (e.g., Feeling-Judgment, Sensing-Thinking) and how different they are from one another. That said, according to Jung’s theory of personality types, the functions and attitudes of individuals are complex and dynamic, which means that there may be characteristics and behaviors that do not emerge at a preference-level scale of analysis. (MBTI® Manual. Myers et al. 1998, CPP Inc. p.172 Validity)
Analyses that fall into this category include the following:
Factor analyses provide evidence for the construct validity of the MBTI assessment tool. These analyses first select a sub-set of test items that directly address the research question (e.g., the validity of the four preference scales). Then, researchers decide whether to perform an exploratory factor analysis or confirmatory factor analysis. Exploratory factor analyses of the MBTI® instrument have shown that the MBTI result structure—the four scales of the Myers-Briggs model—are strongly correlated with the test items. Furthermore, several confirmatory factor analyses of multiple MBTI® forms have supported the hypothesized factor structure and rejected possible competing models. (MBTI® Manual. Myers et al. 1998, CPP Inc. pp.172-173).
Correlation of MBTI scores with other instruments may be a further indication of validity of the MBTI scale. While other instruments may not use exactly the same terminology, many do measure introversion, extraversion, tendency to be stressed by environmental demands, career interests, abstract thinking, logical analysis, and other factors examined by the MBTI. While this strategy has been problematized, it does generally support MBTI construct validity.
Studies of behavioral differences between types have been conducted with a broad range of foci, from proclivity for creativity to selection of medical sub-field. These studies have largely suggested significant behavioral differences between types. For example, creativity has been shown to be significantly correlated with Intuition and Perceiving types, while numeracy and technical aptitude are correlated with Thinking and Judgment types. These findings from this considerable body of research demonstrate beyond a reasonable doubt that the scale dichotomies are meaningful and distinct from one another.
Validity of Whole Types and Type Dynamics: An individual’s type as a whole is more complex than just the aggregate of their place along the four dichotomies. In other words, knowing someone’s preferences is not sufficient to understand their overall type dynamics. The goal of Whole Type studies is to demonstrate that each type is meaningful and distinct from each other type. Analyses that fall into this category include the following:
Comparison of MBTI types with self-estimates of type is useful because the chance of randomly choosing one’s assigned type is just 6.25%. In a number of studies, the percentage of adults who accurately chose their assigned type was 58-85%, suggesting that the MBTI types do align with respondents’ self-perceptions of their own preferences and personalities.
Preference interactions analyses of national sample data found evidence of interactions among all possible combinations of separate preferences, and each of these interactions is significant in predicting the study’s independent variables. Fifty-nine variables were predicted by one or more main effects, while 34 were predicted by one or more interactions. The percentage of variance accounted for ranged from 0.5% to 10.30% for main effects and from 0.5% to 2.90% for the interactions.
Studies of whole types can examine the types themselves or how people who belong to different types interact with one another. Evidence in this category may include observations by independent raters regarding the behavioral tendencies of each of the 16 MBTI personality types in a variety of different contexts, as well as psychological analyses that involve the relationships between different types, such as relationship satisfaction in couples, conflicts and conflict resolution between co-workers, leadership style and proclivities, and more. Overviewing each individual study here is not possible, but broad literature reviews that aggregate results from multiple studies have yielded a number of interesting and useful findings, including: FJ and FP types have the highest family happiness, while ENTP is the lowest; of all types, ENTPs thrive most in competitive, challenging environments, while INFJ and ISFP are lowest; TP types value free expression more than other types; NTJ types are strongest in financial analysis, while STP are weakest. The fact that types behave significantly different from one another in a variety of contexts demonstrates the validity of the MBTI instrument and of the types themselves.
The reliability of the MBTI® instrument has been repeatedly tested, verified, and improved over the several decades. Reliable instruments must be internally consistent and yield replicable results over time. Let’s consider each of these criteria in turn:
Internal Consistency: Results are considered internally consistent if respondents give similar answers for similar prompts. One method for determining internal consistency is called Split-Half Reliability. In this method, the item pool is split in half, either using the consecutive items procedure which assigns the first half of the items on the scale to one set and the second half to a second set, or using a matched pairs method, that considers item format, sub-scale coverage, item-to-total correlations, maximum amount of item information, and other variables to ensure the two sets are similar to one another. Then, the internal consistency of each half is calculated and corrected to account for the length of the instrument itself—longer instruments are generally more reliable and internally consistent than shorter scales. Finally, the results from each half are statistically compared to confirm that respondents’ answers are not significantly different from each other. A second method for determining internal constancy involves using the statistical coefficient alpha, which is the average of all of the item correlations.
All four MBTI® scales have relatively high internal consistency regardless of the method used to split the item pool. Furthermore, studies have demonstrated that the coefficients determined by the split-half and coefficient alpha methods are almost identical.
It is important to note that participant characteristics, such as gender, ethnicity, and age, could affect the consistency of their responses and therefore the instrument’s internal consistency. Research that has investigated this question suggests that there is no significant difference in reliability between male and female respondents, respondents of different ethnicities, or respondents over the age of 18. Internal consistency may be lower for younger respondents, perhaps because their sense of self is not as concrete, or because their reading level was not high enough to fully understand some items. Furthermore, reliabilities tend to be higher in higher-achieving groups and those with higher intelligence scores.
Test-Retest Reliability: This measure estimates how stable a characteristic is over time. In other words, if the MBTI® is reliable, then the same person should get similar results if they retake the MBTI® instrument a few months or even a few years later. There are several sub-types of test-retest reliability. Correlations of continuous scores compares the correlation coefficients of each item over a given time period. In one study that compared items from Form M over a 2.5-year period, the reliability coefficients ranged from .83 to .95, while earlier forms might have coefficients as low as .59 over just 9 months. This and other similar studies suggest that more recent forms of the MBTI® instrument are more reliable than earlier forms.
A second, less sophisticated measure of test-retest reliability is the percentage of people who exhibit the same preferences on retest. The chance probability of choosing all four preferences a second time (that is, coming out exactly the same type) is 6.25%. In a sample of 424 respondents, 65% chose all four preferences on retest and 93% chose at least three of four on retest – both of these numbers are significantly above chance and demonstrate the test-retest reliability of the MBTI® instrument.
How and to whom the inventory is administered is just as important as the design itself. The MBTI® has been shown to be most accurate for individuals over the age of 14, in part because younger children may not be as confident in their self-perception and may be less able to self-report their attitudes and beliefs. Other contributing factors are reading level—a recent analysis using the Fry readability formula estimated that the inventory had about a seventh-grade reading level—and language ability. There are currently 14 commercial translations that have been rigorously tested to ensure that they are valid and reliable, as well as 15 additional translations that are in development.
The MBTI® instrument should also be carefully introduced, especially to students or others who may feel vulnerable or nervous about a psychological assessment. For example, examinees may need to be assured that the MBTI® does not assess mental or emotional health, and that their results are confidential and belong to them.
Key Strategies for Interpreting and Applying MBTI® Results
Accurately interpreting the results of the MBTI® survey is essential for respondents to apply their results to improve their lives. Interpreters and respondents alike should also consider their stage in life and any relevant roles and responsibilities. For instance, a student will probably not apply their results in the same way as a seasoned professional, who will not apply theirs in the same way as an organizational leader. Interpreters should use their familiarity and knowledge of their client while guiding them in how to apply insights from their MBTI® to yield the maximum benefit for them. In this section, we overview a number of key strategies for interpreters and respondents in different contexts.
Strategy 1: Interpreting MBTI® results should be a collaborative process between the professional interpreter and the respondent. If you are an interpreter, presenting the results in an open-ended way may help your respondents feel more comfortable and amenable to the analysis, and might even provide insights that the indicator itself could not unearth. For example, you might say “This report shows that you tend to prefer A, B, and C. Do you think that describes you?” This approach allows for conversation and discovery, while a more top-down presentation, “Your personality type is INTJ” may be intimidating or cause the respondent to shut down.
Strategy 2: As you converse with respondents, it is important to remember that the MBTI® is an indicator not an exam. There are no right or wrong answers and there are no good or bad personality types. The purpose of the MBTI® is for respondents to learn more about themselves and to use those insights to improve their professional performance and personal relationships. Along the same lines, try to avoid “black-and-white thinking”—everyone uses both sides of each dichotomy, in the same way that we all use both of our hands. Right-handed people still use their left hand for many things and may even do some things better with their left hand. In the same way, even a highly sensitive, feeling person will still use critical judgment at times. Understanding this flexibility can help respondents better understand the scope and applicability of their MBTI®.
Strategy 3: If respondents are high school or college students, they may be able to use their personality type results to better understand and even improve their academic performance. For example, Thinking personality types tend to earn high scores in persuasive writing and justify their position using objective quotes or data, while Feeling personality types tend to make self-references or use persuasive tactics. Similarly, Intuitive and Perceiving personality types tend to perform well on foreign language assessments, while these skills come less naturally to Thinking and Judgment personality types.
All of this said, students should keep in mind that their personality type does not determine their academic, vocational, or personal outcomes. Rather, personality type can be a tool for making sense of why they have certain preferences or skills, as well as how they can improve their performance in certain contexts given those preferences.
Strategy 4: If respondents are seeking a job, they may be able to use their MBTI® results to help focus their job search. For example, a highly technical, competitive career may not be optimal for someone with Sensing and Intuitive tendencies. Career counselors should also consider peripheral factors such as work environment when helping their clients explore careers. For example, ESTPs often value highly structured hierarchies and environments, while most Intuitive types instead value having a wide variety of tasks to complete. In addition to field and work environment, the degree of schedule flexibility, professional independence, and supervisory oversight might also be relevant for jobseekers choosing a career in which they will be both professionally successful and personally fulfilled.
Strategy 5: Organizational leaders should apply their team’s MBTI® results to improve how their team members function, both as individuals and as a group. For example, MBTI® workshops can help team members understand how they and their colleagues communicate, so they can anticipate misunderstandings and resolve conflicts quickly, amicably, and before they escalate. The MBTI® can further help leaders motivate their team members as well as manage organizational change effectively. For instance, leaders may need to give Introverted personality types adequate time to think and process the changes individually, while extraverts might benefit more from an open forum or discussion. Similarly, in order to become invested in the change, Thinking personality types may want to know the logic and reasons behind the changes, while Feeling personality types may be more concerned with how the changes will impact employees and how their needs will be met during the transition.
Strategy 6: While there may be some differences in interpretation when the MBTI® is used in multicultural or international contexts, interpreters should be aware that the psychological personality type framework and MBTI® inventory have been used effectively in dozens of different cultures around the world. A number of studies have confirmed that the instrument is both valid and reliable, and that it can improve workplace function and personal relationships in much the same way that it does in the United States. That said, interpreters should also be familiar with and willing to adapt their interpretation to their specific context and respondents as is needed.
These strategies for applying the MBTI® insights serve as a starting point, but there is always more to learn. We encourage you to continue to explore our website to learn more about the MBTI® instrument and how the 16 personality types affect daily life.
MBTI Manual (Isabel Briggs Myers et al., 1998, CPP Inc.)