What is the problem?

In a nutshell, the essential problem driving claims that the MBTI instrument does not have acceptable levels of reliability and validity are driven by the basic opposition between type theories of personality and trait theories of personality. These two positions are each represented by a range of theories and assessments but by far the most representative of each camp are the MBTI instrument for type theory and the Five Factor Model (FFM) for trait theory. The FFM is the most used personality model and assessment in academia and the MBTI assessment is the most used personality model outside of academia, in professional practice, personal development, and in popular culture.

Importantly, trait theory and type theory are measured differently, and this is at the heart of the matter. Type theory through the MBTI instrument measures personality preferences using dichotomous scales and sorts results into preference categories, e.g., results are in either the category of extraversion or introversion; type theory does not measure how much of that preference a person possesses. Trait theory through the FFM measures personality traits on a continuous scale and reports where results fall on the spectrum from a lot to a little of that trait; e.g., 72% extraversion, 35% agreeableness, 55% openness, etc.

Critics of the MBTI instrument make claims that the dichotomous scoring of the MBTI instrument does not provide reliable results over time, particularly in test-retest studies. Reliability means that results are consistent over time; if results were not consistent over time, the results of any assessment would not be meaningful.

Trait theorists have conducted studies and make claims that the MBTI instrument does not have acceptable levels of reliability and validity and should not be used to provide reliable, meaningful insights into people and how they grow and change. In response, MBTI theorists have conducted studies and make claims that the MBTI instrument has more than acceptable reliability and validity and should be used to provide meaningful insights into human personality. Both ends of the argument have published scientific studies to back up their positions. So how do you wade through these positions and come to your own conclusions?

This page provides guidance on how to understand the claims made against and for the MBTI assessments, important responses to the major claims, and where to find quality reference material for yourself so that you can review the research and determine the merits of opinions from all sides.

How to Evaluate Claims

Most importantly, an article or opinion that makes absolute claims—for or against the MBTI model or instruments—should be suspect. As with almost all areas of life, absolute claims do not represent reality but extreme opinions.

Consider also the fact that the MBTI instrument has been through over 70 years of application, study, analyses, criticism, and theory-building. But the global online content is overwhelming and primarily unvetted for accuracy, which means that anyone offering an evaluation of the MBTI instrument's psychometric properties will need to rely on selective resources. When evaluating claims and opinions about the MBTI instrument, it is also a good idea to check references and sources and to look up and consume as many primary sources as you can to come to your own informed conclusions.

It is our intention to present a balanced, measured, fair, and objectively informed position, with instructions on how you can find quality reference material for yourself. There is a large volume of academic publications and studies on the MBTI instruments, much of it positive though not all of it. Because the average reader does not have access to academic journals (something typically only academics read), they do not know where to access them.

At the end of this page, there is a link to all studies and articles cited on this page, and additional articles for those people who want to dig deeper. We aim to be a trusted source for education on the MBTI framework, including providing access to research articles that are publicly available, but many people do not know where to find them.

Claims and Facts

Claim: The MBTI instrument was created by non-academics who did not have PhDs.

This is true. Isabel Briggs Myers and her mother, Katharine Cook Briggs, did not have PhDs, nor did they work in academia. This claim implies that people without PhDs or academic standing cannot be smart enough to create a quality assessment. Of course, we know that is not true at all.

This claim also misses the fact that the MBTI instrument has gone through decades of refinement, testing, analyses, and growth by people with PhDs, many of them academics (as well as some very bright people who were not PhD academics).

Claim: The MBTI instrument is based on pseudoscience and does not have reliability or validity.

This is not true. The MBTI instrument has over 70 years of scientific testing, research, and practical application. An instrument's reliability emerges from the samples and conditions in studies conducted and is not a property of the instrument itself. While there are some studies that demonstrate poor reliability, there are many studies of the MBTI instrument that demonstrate strong reliability and validity. For example, Capraro (2002) conducted a meta-analysis of studies of the MBTI instrument and found that reliability across studies was good (i.e., .80 to .87) with some variability, particularly in the Thinking–Feeling scale (0.64 to 0.87). Scientific and academic standards for acceptable reliability are .70 or higher.

Variability in performance across samples and varying demographics in studies is not unusual for self-report assessments. A review by Gnambs (2016) of reliability studies for a Big Five Trait Model instrument found both good reliability and poor reliability demonstrated for the same instrument in different samples. These are examples of the relativity and nuance of the reality of psychometric instrument performance and why we should be wary of absolute claims.

As for validity, we refer to a 2018 article by Moyle and Hackston published in the Journal of Personality Assessment. The authors provide a particularly clear and comprehensive response to criticisms of reliability and validity and lay out a case for experiential validity.

Claim: Academics do not use or like or support the MBTI instruments.

This is not true. There are several hundred studies using the MBTI instrument published in academic journals. Each year our research team collects, vets, selects, and summarizes research from across the world that uses the MBTI instrument and is published in academic journals. The in-house Journal of Psychological Type^®, published from 1975–2015, offers a wide range of studies in methodology (empirical, correlational, qualitative, theoretical) and topic (relationships, career, education, and more). A separate in-house publication, the Journal of Psychological Type Research Digest has published summaries or reviews of the collected academic research with the MBTI instrument annually since 2016. Over the years academics have published opinion pieces on the quality and reliability of the MBTI instrument, such as Johson's (2016) and Gordon's (2020) thoughtful pieces in Psychology Today, where they present a balanced, clear evaluation of the MBTI assessment's strengths and weaknesses.

Fact: The Myers-Briggs Company's research has been under-utilized in evaluations by academics, critics, and the public.

The Myers-Briggs Company (formerly Consulting Psychologist Press or CPP) is the exclusive worldwide publisher of the MBTI assessments and has produced high-quality, rigorous studies for many years that we rarely see referenced in critiques. As an example, through our library we provide an excerpt from the 2018 MBTI Manual that details the results from global research on the latest versions of the MBTI Global Step I^™ and Step II^™ instruments. In a global sample of 1,721 adults who took the MBTI Step I twice between 6 and 15 weeks, the test-retest reliability coefficients were 0.81 to 0.86 across all four scales.

The Myers-Briggs Company has an excellent MBTI Facts page with detailed explanations of the reliability and validity of the MBTI instrument and references to published studies.

Where to Access Quality Reference Material on the MBTI Instrument

We have created a collection of research articles, including the ones cited on this page, on the topic of reliability and validity of the MBTI instrument. The collected articles can be accessed and downloaded for free through our library, Mary and Isabel's Library Online MILO^®. If you want to download the articles, you will be prompted to first create an online library account which is free to do. With a MILO account, you can search for more material and work with the librarian on specific searches of material—just reach out to library@myersbriggs.org.

Through MILO, the Journal of Psychological Type Research Digest, published since 2016, can be accessed online, as well as 50 years of publications in the Journal of Psychological Type.