Jan 17 2020

Personality Test Pseudoscience – Swedish Edition

Our skeptical colleagues over in Sweden have been tackling one of the most popular pseudosciences there (thanks to Swedish Skeptics president Pontus Böckman for the heads up) – their “Fraudster of the year in 2018 was author Thomas Erikson for his book Surrounded by Idiots. The book popularized a dubious personality test known as DiSC, which uses four colors to characterize personality types, red, green blue, and yellow. The book is a best seller in Sweden and is now being exported to many languages. It is also incredibly popular in corporate culture.

Dan Katz, licensed psychologist and psychotherapist, does a great job of breaking down why Erikson’s claims are not based on science. I’m not going to repeat his article, just read the original. The short version is that Erikson has no qualifications, and the DiSC model is not based on any scientific evidence. The entire thing is a giant scam. What I do want to do is extend the discussion about personality tests in general.

The core scientific question behind any personality test is this – is there even such a thing as a personality type? The current best short scientific answer is – mostly no. But let’s get a bit deeper and more nuanced. First we need to distinguish personality traits from personality types. A personality trait is simply a behaviorial tendency that someone has that transcends any particular context and is consistent over long periods of time. Introversion/extroversion is probably the best established personality trait. The big five personality trait spectra that are the best established and are generally accepted by psychologists are OCEAN – openness, conscientiousness, extroversion, agreeableness, and neuroticism.

But we have to acknowledge that these are just cultural and scientific constructs. They are our best attempt at understanding some of the building blocks of personality, but we don’t know if they are bedrock, meaning that they are real neurological phenomena. My sense is that introversion/extroversion probably is a fundamental neurological phenomenon – one reason for this is that there are known genetic disorders that have extremely consistent extroversion personality traits.

To further explain what I mean by all this, what we really want to know is what are the fundamental neurological functions in the brain that ultimately result in what is expressed outwardly and we interpret as these various personality traits? To answer that we need a combination of psychology and neuroscience – understanding how human behavior works (psychology) and then correlating that with specific circuits in the brain (neuroscience). Research is just getting to the point where we can start to answer these questions, but we have a long way to go.

So some identifiable personality traits do seem to exist (otherwise everyone would be the same) but let’s put this into context. Personality (internal factors) is not the only thing that affects human behavior. There are also external factors. So the real question is, what is the relative contribution of internal and external factors to our ultimate behavior? The answer seems to be – it depends. Some personality traits in some individuals are dominant, while other are more muted. Further, some situations have a stronger influence on our behavior than others. An extreme enough situation will force almost anyone, regardless of your personality, to react a certain way.

There is, of course, an entire science that studies how people react in specific situations, social psychology. If I had to paint this entire field with a broad brushstroke I would say that in most situations people behave rather predictably, at least statistically. In short, we are mostly more similar than we might naively imagine. People are predictable, to the advantage of magicians and con-artists. In other words, there are common human personality traits, and we are all a variation on this theme. What this means is that generic human factors and external factors seem to have a greater influence on our behavior than individual internal factors. So while it is important to understand personality, and for each of us to be in tune with our own traits and tendencies, personality is not destiny. Except in extreme cases, it is not the dominant factor in determining our behavior, the situation is.

A great example of this is the marshmallow test – this is a classic psychological study in which children are offered a marshmallow (or some reward) but told that if they don’t eat it right away the tester will be back in some time period with three marshmallows. This is designed as a test of self control and delayed gratification. It was originally highly correlated with later life outcomes, and was therefore thought to be a manifestation of some internal quality of the child. However, more recent replications have cast significant doubt on this conclusion. They found that performance on the marshmallow test was more influenced by socioeconomic status and parenting, leading to the alternate hypothesis that the ability to delay gratification was likely more due to the trust children have in the adults in their life, not some internal trait.

What about personality types? Traits are individual tendencies. A personality type is a suite of traits that tend to cluster together. The pop personality tests, like DiSC, or the popular Myers Briggs, are based on the assumption that types exist, and so they cluster people into a limited set of archetypes. But the evidence shows this basic assumption is probably false, dooming any personality type system.

One way to test the notion of personality types is to see if any attempt at identifying types has internal or external validity. Internal validity means that if two different people administer the test to the same person they get roughly the same results. Also, if the same person takes the same test twice they will get the same results, which will be mostly stable over time. External validity means that the test results predict something in the real world, some tangible outcome. Does it predict vocation, the chance of going to prison, or life expectancy? None of the popular personality type tests have well established internal or external validity, which is a main reason that psychologists remain skeptical about the idea.

A recent study took a robust data crunching approach to large data sets of personality traits. They found that there was a statistical clustering of personality traits into four types, but the effect overall was weak. Further, the most common “type” was undifferentiated – not falling into any of the other four. Most people were just a mish-mash of personality traits, but there was a statistical tendency for some traits to occur more commonly with others falling into “at least” four clusters. Ultimately this was pretty weak sauce, and certainly did not support any of the existing personality type testing systems.

Ultimately, the notion that we can sort humanity into four types, or any similarly limited number of discrete personality types, is a flawed approach. It has a certain appeal, but it seems to be little more than astrology. I think people like the false sense of self-awareness, the illusion of self-knowledge. It also ironically makes them feel a little special. But mostly I think the appeal is that sorting people into personality types seems like a hack, a quick cheat, that cuts through all the complexity of dealing with people or complex situations. It can also be used as an excuse – sure, I did this stupid thing, but that’s because I am a blue, so what did you expect?

In the end it’s all highly counterproductive. As Katz points out, companies relying on this testing can easily blame dysfunction or bad outcomes on the mix of personality types at work. This whole approach is doomed to failure. Meanwhile there is a robust actual science of social psychology with tons of information about how best to optimize function in working groups and troubleshoot problems. Again – people should focus on the external factors, the working environment and the systems that are in place. People respond more to these external systems than to their programmed personality type.

No responses yet