Are the criticisms of MBTI justified?
Abstract
There are alternative ways to conceive of and represent ideas in personality. One that has caused some debate is between the Type versus Trait approaches. This is sometimes quite heated and Trait theorists often criticise the Type approach – especially as it is embodied in the most popular questionnaire – the MBTI. This paper examines nine of the most common criticisms of MBTI and of the Type approach, in an attempt to correct some misunderstandings and to recognise the value of both approaches.
Introduction
The MBTI is one of the most popular personality questionnaires in use today. Since market leaders often attract criticism, it is important that such criticism is both accurate and fair. This is not always the case, and some criticisms are based on underlying assumptions that are not made clear and some are based on some quite basic misunderstandings or misinterpretations. This article identifies nine of the most common criticisms and presents reasons for reconsidering them.
As background for considering the 9 criticisms it is useful to keep in mind that each will have either an implicit or sometimes explicit position on the fundamental question of what we mean by personality. Some are tempted to believe that this has been largely resolved with the emergence of the Big Five model which now dominates discussion – at least amongst those taking a psychometric perspective. However, all methodologies contain their own biases and assumptions and it is useful to consider what these are. Even devotees of the scientific method need to recognise the method’s assumption that there is no such thing as a unique event. The need for replication, which is at the heart of evidence-based scientific practice, has the consequence that any unique event would be (and is) dismissed as chance, coincidence, or error. Perhaps science is right in making that assumption, but we should at least acknowledge the assumptions and the possibility of what we could be blind to.
One of the assumptions underlying the Big Five model of personality is the value of averaging. This is done when people complete questionnaires since they are asked to provide a typical response. It is also done in presenting the results as a point on a continuous scale to represent what could be quite a range of behaviours and preferences. Another assumption comes from using the correlational method which is at the heart of the psychometric approach. It leads to combining items that correlate to form a scale. The justification for including a range of items about behaviour, attitudes, motivations and preferences (see a review of items in the NEO and the 16PF by the author) is founded on statistical rather than psychological criteria. This makes the definition of personality less clear.
It is worth considering an alternative way of thinking about personality which can be illustrated using the concept of optimism. Ignoring the issue of how we choose items to represent this characteristic, the psychometric approach invites people to say how optimistic they are. The responses are then summed and placed on a continuous scale with optimism at one end and pessimism at the other. Questions arise about the meaning of a score that is somewhere in-between. Does it mean that the person is neither? Perhaps it means that they only express mild symptoms of optimism/pessimism. Possibly it means that they oscillate between the two extremes. This last idea is an alternative conception of personality whereby people are both optimistic and pessimistic. We then want to understand how they manifest both. What, when and where do these different characteristics emerge? It encourages us to examine context and how people learn to manage both sides. Of course, people will differ in terms of which emerges more often. However, we know that an average of 5 could be obtained from a 4 and a 6 or equally from a 1 and a 9. Perhaps viewing personality as a mosaic and identifying all the elements is more useful in helping the more optimistic to recognise how much (or how little) pessimism they have to call on in managing the danger of over-optimistic naivety. Vice versa, the more pessimistic may find it useful to recognise their potential for optimism which can help them recognise potential positive outcomes. The same concept of balancing opposing characteristics can be applied to the Big Five. An extravert is also an introvert. They may be loud and sociable and compelled to perform but they also need space and time away from the limelight at times.
With these thoughts in mind, we can examine criticisms made of MBTI and other Type instruments and consider alternative ways to view them. Below are the 9 most common criticisms with a brief discussion and pointers for consideration:
1: A Type suggests being labelled or boxed
This is one of the most common criticisms of the Type approach. Clearly the concept of ‘a type’ assumes that we can classify people. This means that they fall into one category which logically excludes them from an alternative category. In other words, people are ‘either one or the other’. To some this seems highly restricting (especially to Trait theorists) who point out how people clearly vary in the degree to which they demonstrate personality characteristics. However, Jung’s Theory of Psychological Type (henceforth called Type Theory) does not deny that people differ in degree. The difference is that, rather than thinking about high, average and low scores on a trait, Type Theory proposes a more dynamic view of personality. Using the Introversion-Extraversion dimension, it suggests that people can be both – but not at the same time. In other words, you cannot be introverting when you are extraverting – and vice versa. What differentiates people is the degree to which they prefer one over the other. A person who prefers introverting will still extravert on occasions but the balance would be different than for someone who prefers extraverting. Thus, Type Theory sees personality as a dynamic process of balancing opposing preferences. This rings true for many people who recognise that they can ‘visit’ both ends of such a trait. This recasts the Type classification not as ‘either/or’ but of ‘both’. Differences in degree then becomes differences in how we balance these preferences. It also changes the concept of average from being a less pronounced version of both ends to being a more evenly balanced expression of preferences. The label or Type classification is no longer a box but a place that represents the person’s ‘centre of gravity’ or ‘place of preference’. This is a little like finding the room in your house where you feel most comfortable or in which you spend most time – but this does not mean that you do not visit other rooms in your house.
2: Type scores are not bi-modal.
A typical argument against the Type approach is that scores on Type questionnaires show a range from high to low. It is presented as evidence of a continuum as assumed by trait theorists. It is argued that, as Type proposes a dichotomy (i.e. the two opposing preferences) then the distribution should be bi-modal which would look like this:
Whilst this is argued by many academic psychologists, it is based on a fundamental misunderstanding. This is easily illustrated with a simple exercise – the Preference Test which is available in section 7.2 of the online course.
When people complete the Preference Test (which explores left and right handedness) they could potentially score between 15 and 45. When completed in a group you can plot the distribution. From doing this many times a typical distribution looks like this:
Invariably, scores range from low to high with very few at the extremes. This is often used as an argument by those who criticise the Type approach but the exercise shows how, even with a very clear dichotomy like left and right handedness, a range of questions can still produce a distribution that looks continuous. This demonstrates how arguing that the shape of a distribution proves an underlying continuity is wrong. It is surprising how many people in psychometrics miss this point. The example also helps to reframe the meaning of a score in the middle because a middle score on the Preference test does not mean that people change gear with a part of the body somewhere between your left and right hand – perhaps your chest! Instead, it shows that you have a more balanced use of the two alternatives or opposites.
3: Type questionnaires are unreliable.
This is a criticism often aimed at the MBTI. It is based on the premise that personality does not change and that a self-report questionnaire should identify it accurately. However, we need to recognise that self-report questionnaires do not measure personality directly. Instead, they are mediated by a person’s self-perception which, in turn, is related to a person’s self-awareness. Hence, even if our personality does not change a person’s increasing (or decreasing!) self-awareness may well be changing. And we need to recognise that the goal of much coaching, training and facilitation is precisely to increase self-awareness. Instead of thinking of self-report questionnaires as measuring instruments (the predominant model) we could think of them as instruments that help people to articulate their current auto-biographical narrative. This can then be used to confirm, communicate, discover, and challenge current self-perceptions which can be prone to distortions.
If this process is successful, then people may well change their self-perception, which demonstrates how the concept of reliability is predicated on the assumption of stability which may not be entirely appropriate – and which we do not need to adopt or accept. Therefore, even if personality is relatively permanent (which can be further debated), a person’s narrative about their personality clearly can change. In fact, it could be argued that if people do change after they have experienced such a questionnaire, the process has been successful!
In summary, the argument that people change their type or their profile is not sufficient to dismiss an instrument and could even be presented as an argument for its sensitivity to changes in awareness.
4: A good questionnaire should predict behaviour and performance.
This assumption is prevalent in the psychometric model which is not only predicated both on a reliability model based on an assumption of ‘no change’ (as described in the previous section) but also on a validity model which assumes that the aim is to predict future behaviour and/or performance. This suffers from several misconceptions. The first is that it does not address the difference between Being and Doing (see next section) – the fact that who we are on the inside (our preferences) does not determine what we do in different situation (our roles). This is linked to the belief that we are locked into our underlying personality – hence it is deemed to restrict our ability to change our behaviour whereas success is often based on people being able to adapt, to choose context-appropriate behaviour. Hence linking a ‘score’ on a personality questionnaire at one point in time to the prediction of successful/appropriate behaviour at a future point in time involves many moving parts.
Hence something missing from the current validation model is a more dynamic concept of questionnaires helping people to increase their behavioural flexibility, rather than linking two static pieces of evidence – a predictor score at time x and a criterion score at time y.
5: Type questionnaires focus on preferences rather than behaviour.
As mentioned in the previous section, there is a significant difference between Being and Doing. This is because ‘what I do’ is based on a dynamic tension between ‘who I am’ and how my circumstances constrain ‘what I do’. For those who view behaviour as primary (which is a strong tradition within psychology), asking people what they prefer is inviting a ‘wish’ rather than reflecting their ‘real’ personality. However, most people accept that behaviour is modified by context which suggests that what behaviour reveals will necessarily be limited. This does not deny that behaviour can reveal personality traits, but it may not differentiate sufficiently between people’s role and their underlying personality. Furthermore, everyone has had a limited context in which to express and discover themselves. Few people really know what they would be like in very different circumstances such as a war or earthquake or refugee camp. Hence current behaviour is limited by current challenges whereas preferences help people articulate what they believe would be their natural choice of behaviour.
The concept of preference is complicated and you may like to explore this further by viewing this video:
6: The theory of opposites is unfounded.
This criticism stems from the proposition that certain characteristics are opposites. This is then mis-interpreted as meaning either one or the other. However, it was previously made clear that Type Theory uses the idea of opposites as being both. It simply proposes that they cannot be expressed at the same time – and hence people oscillate between the two. Some Trait theorists point to how people can score highly on both Thinking (i.e. rationally supported decisions) and Feeling (i.e. values supported decisions). Type Theory proposes that these are opposites and that some people tend towards allowing Thinking to dominate more of their decisions, whereas others let Feeling dominate more. But fundamentally we all do both at different times and in different circumstances. Hence the separation of Thinking and Feeling in statistical terms does not invalidate the Type Theory proposition.
7: Type Questionnaires are not suitable for comparing people.
The traditional concept of norm tables does not fit with Type Theory. People derive meaning from Traits by examining how different they are in degree from other people. With Type Theory meaning is derived from how types differ in the nature of their preferences – which is defined by the construct rather than defined by differences from others. Hence type questionnaires are designed to operationalise Type Theory, by identifying which constructs a person prefers rather than how they compare to others. Of course, it is interesting to see how types are distributed in different populations and so type tables (i.e. the proportion of the various types in different populations) are more relevant than norm tables.
Another issue that arises in relation to comparing people is the claim that the MBTI is ipsative. For those who don’t remember what ipsativity is – you can view this short video:
Many academic papers and many well-known academics perpetuate the myth that MBTI is ipsative. This is a shocking misapprehension. Ipsative questionnaires force you to choose between relatively independent traits such as Anxious, Friendly, Detailed – often inviting people to say which is most like and least like them. This makes ipsative tests self-referenced and unsuitable for making inter-personal comparisons. However, trait questionnaires do force you to choose between one end and another end of a scale – such as Extravert versus Introvert. The forced choice in the MBTI is no different from that – it is a choice between two ends and NOT between two independent constructs.
In summary MBTI and other type indicators are NOT ipsative and many academics are either misunderstanding or misrepresenting type questionnaires when they claim it is.
8: The reports simply describe positive characteristics.
Some people find that the reports generated by MBTI (and other Type Indicators) make the personality sound too positive – too good. This is not really a criticism of the instrument or measurement but rather a criticism of the interpretation. It is sometimes argued that the positive descriptions are a technique for gaining greater agreement (and hence popularity) of the tool. If that were the case, it would appear to be working since MBTI is the most popular personality questionnaire in use today! However, whilst it would be possible to write negative interpretations as well it should be recognised why the positive descriptions are preferred.
Ethical – personality questionnaires claim that there is no right or wrong which makes a negative interpretation somewhat incongruous
Theoretical – each preference describes a different way of understanding and engaging with the world. An analogy would be that using one preference is like seeing the world in 2D whereas using the opposite preference gives a 3D image. In other words, each preference provides a partial but valuable view of the world and this is what the reports describe.
Practical – it is more effective to encourage learning, personal development and self-awareness by highlighting strengths before addressing potential weaknesses. The reports clarify the upside of people’s preferences (if the person manages themselves well). The potential downsides (sometimes conceived as a characteristic ‘overdone’) are then easily explored by examining the opposite type which highlights preferences that a person may not understand, value or may find more difficult to emulate.
This may be more easily understood with an example. Some people immerse themselves in detail which can mean that they lose sight of the big picture. Others focus on the big picture and fall down due to lack of attention to detail. Both preferences are useful and even necessary for managing real world activity. The theory of dynamic opposites highlights what people may find easier to do well and de facto suggests what people may find more difficult. A common misunderstanding is that the Type report describes a well-developed and balanced example of someone with those preferences, which the individual can aspire to.
9: Type Theory is not comprehensive.
Finally, MBTI and Type Theory have been criticised for not assessing anxiety. This is a bit like criticising a car for not being an aeroplane. Jung specifically argued that personality theory was dominated by how people handle their emotions. He proposed Type Theory on the grounds that people still have a personality even when they are not in the grip of their emotions. In fact, it can be argued that the strength of Type Theory is precisely that it allows us to examine personality without the complication of emotional reactivity. Of course, each type will manage and express itself differently if they are either high or low on anxiety, but we should not criticise a car for not having wings.
Dr Roy Childs
Managing Director & Chartered Occupational Psychologist
Roy has worked in organisations at senior levels for more than 20 years and focuses on developing capability and building relationships. He works both as a coach and facilitator, assessments expert and psychometrician, and also delivers some of our advanced training courses.
TEAM FOCUS LIMITED
Telephone: (01628) 637338
E-mail: teamfocus@teamfocus.co.uk
References
Furnham, A. (1996) The big five versus the big four: the relationship between the Myers-Briggs Type Indicator (MBTI) and NEO-PI five factor model of personality. Personality and Individual Differences, Volume 21, Issue 2, August 1996, Pages 303-307 https://www.sciencedirect.com/science/article/abs/pii/0191886996000335#!
Furnham, A. (2022) The Big Five Facets and the MBTI: The Relationship between the 30 NEO-PI(R) Facets and the Four Myers-Briggs Type Indicator (MBTI) Scores. Psychology, 13, 1504-1516.
doi: 10.4236/psych.2022.1310095.
Stein, R., A. B. Swan (2019) Evaluating the validity of Myers-Briggs Type Indicator theory: A teaching tool and window into intuitive psychology
Psychology Today
https://www.psychologytoday.com/us/basics/myers-briggs https://www.psychologytoday.com/us/blog/give-and-take/201309/goodbye-mbti-the-fad-won-t-die
Videos
Michael Pierce https://duckduckgo.com/q=criticisms+of+MBTI&iax=videos&iai=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DWexxkbOTg6I&ia=videos
Dr Steve Myers
Comments