In full disclosure, I’ve been a long-time hater of the MBTI. As a matter of fact, one of my favorite pastime activities is responding with this 2-min Adam Ruins Everything episode about the MBTI whenever someone asks for advice about an MBTI facilitator.
But this initial hate turned over time to curiosity around questions like: why do people find this assessment so compelling? why do workplaces keep using it? and are there better alternatives out there?
As I noodled on these questions I realized that they extend to a broader category of assessments that can best be grouped under the psychometrics label: “the science of measuring mental capacities and processes.” It’s a bit of a generous label for some of these assessments, as many don’t meet the “science” qualifier (MBTI, for example) but their intent aligns with the second part of that sentence.
The appeal of “clustering tests”
Most people don’t like to take tests. And not everyone views feedback as a gift. Yet a subset of psychometrics tests seems to be highly popular and some people even enjoy taking them. Why?
This subset of tests is part of a category I’d call “clustering tests” — their aim is to group human behavior into clusters and then figure out which cluster you belong to, often referred to as your “style” or “type”. Wilson and Walton’s theory provides one of the best explanations for the appeal of “clustering tests”: we are all meaning-making machines, and interpret a given situation in a way that serves three underlying needs:
- The need to understand — make sense of things around us in a way that allows us to predict behavior and guide our own action effectively.
- The need for self-integrity — view ourselves positively and believe that we are adequate, moral, competent and coherent.
- The need for belonging — feel connected to others, accepted, included and valued.
We find “clustering tests” appealing because they check all three boxes: they make it seem easier to understand ourselves and others, reducing our and their complex set of behaviors into a simpler, more predictable “type”. They strengthen self-integrity by highlighting all the positives in our “type”. And they foster our need for belonging by connecting us to a tribe: the INTJs, the ENTPs — a group of people who are “like us”.
But what do we find when we dig below simply satisfying our meaning-making cravings?
The bright side
The core benefit of psychometrics comes from acknowledging that other meaning-making strategies, such as self-reflection, or the feedback and perspectives of others are not without their shortcomings and limitations either. They too are often inaccurate, subjective and incomplete. Therefore, when used properly, psychometrics can be a valuable complement for other meaning-making strategies.
Furthermore, grappling with complexity is hard. And sometimes, but definitely not always, grappling with a simpler challenge helps us better understand the more complex challenge. The simpler can sometimes be used as a starting point, gradually layering on complexity in later stages. Therefore the simplicity that psychometrics offer, when held loosely, can be helpful.
The dark side
Picking up from where the bright side left off, holding on to simplicity too tightly creates problems. Especially when used for prediction, a simplistic model of complex phenomena will often yield false predictions.
Furthermore, the way “clustering tests” create a sense of belonging can also lead to othering: there’s us, the INTJs, and there’s them, the ENTPs…
Finally, coming full circle to the beginning of this post, many of the assessments do not sit on solid scientific foundations. While I’m not sure that this is intrinsically problematic, it is certainly a serious issue where psychometrics are meant to support fair and objective decision-making. More on this shortly.
How to best use (and not use) psychometrics
The inherent attributes of both the bright and dark sides of psychometrics: meaning-making power, reductionism, scientific credibility, belonging, etc. are often at odds with one another making this an optimization problem that is very situational. In different circumstances, some of these attributes matter more, or less, than others.
This pithy advice from a Forbes piece captures this sentiment fairly well:
First distinguish between real tests and masquerading fortune cookies. Select an assessment that is psychometrically sound, situationally relevant and provides actionable insights.
Let’s unpack this a little further, given the different situations in which psychometrics are often used.
Perhaps the least contentious use-case of the three. The most dominant factor here is the assessment’s meaning-making ability as the value comes not just from highlighting the opportunity for growth but also from building the motivation to do so.
Since scientific credibility has a positive impact on my personal motivation, I tend to prefer assessments such as Hogan and the Leadership Circle Profile, but I don’t think it’s a hard requirement. If you find the Enneagrams narrative more meaningful to you — use that.
Under this use-case, psychometrics are used as data to support organizational decision-making in the hiring or advancement/promotion processes.
The key factor that outweighs all others in this situation is the scientific validity of the assessment. I’ll also repeat a somewhat controversial statement that I’ve made before: if the scientific basis is sound, evidence of systemic bias should not automatically lead to exclusion of the assessment. It is almost certainly the case that other data points used in the decision-making process (interview team scores, manager’s evaluation, etc.) are also biased, it’s just that the bias there is more erratic and difficult to quantify. The “upside” so to speak, of systemic bias, is that it can be systemically corrected for.
Building on the research outlined here, the assessments that are worth considering center around evaluating general mental ability (O*NET Ability Profiler, Slosson Intelligence Test. Wonderlic Cognitive Ability Test) and integrity (Stanton Survey, Reid Report, PSI).
Team dynamics and collaboration
Here, psychometrics are often used to validate/normalize the different work and communication styles of the different team members and serve as the basis for discussion on strategies that’ll allow the team to collaborate effectively despite those differences.
However, I would argue that the dark side of psychometrics here is perhaps the most dominant. Both reducing people to their “type” and accomplishing a sense of belonging through tribalism have the most negative effects in this context.
It’s the self-reflection and team dialogue that the assessment triggers that are most valuable. So instead of using an assessment, I’d advocate for using other self/team exercises that accomplish the same goal. The NY Times Kickoff Kit outlines three such exercises (“muppet analysis”, “how we work best”, ”hopes, dreams, and non-negotiables”) offering a much better alternative.