Note: This post is part of the “classic” RIP series, which is described in this initial post for context. (Originally posted April 1, 2024)
We’re long overdue for another RIP, so… let’s rip!
The history of research on personality and individual differences has a long and fascinating history of really, really smart people trying to understand the “best” way to describe the components and structure of various psychological phenomena. Early work on intelligence (or today, often referred to as “mental ability”) was marked by numerous people working to understand what types of mental abilities people had — what were all of the ways in which the brain could be tested for how well it could do different types of things? What were all of those different types of things that the brain could do? And so on. The story of intelligence testing is interesting in and of itself — it’s a truly fascinating tale that ranges from public education to the military to eugenics, and involved more than one of the people who would later be considered some of the most influential personality psychologists in history — but perhaps that’s for another RIP. Ultimately, the study of intelligence came down to a discussion of the g factor (i.e., “general intelligence”) — a superordinate, latent variable to which all other sub-factors of intelligence were linked in various ways. Put simply, a growing body of research found that all of the “mental abilities” that were being studied seemed to be loosely federated in such a way as to where if you scored higher on one, you’d be likely to score higher on several others as well. For a long time, the question of “how many dimensions of intelligence are there?” seemed to be largely resolved.
In the realm of personality theory, several of my favorite debates raged over what the “best” or “most correct” number of personality dimensions existed. Raymond Cattell (an early adopter/pioneer of factor analysis, who was also massively influential in the study of intelligence) came up with an understanding that personality could be organized into multiple, hierarchical levels, and that 16 factors was a good place for most people to start. Some people still use his PF-16 questionnaire to this day. Hans Eysenck had relatively well-articulated theory of personality (strongly) suggesting that a 3-factor solution was best — Extraversion, Neuroticism, and Psychoticism (often referred to as the PEN model of personality). Eysenck also battled extensively with the emergence of the Big Five folks in the late 90’s, blasting them repeatedly with many a question that have yet to be answered satisfactorily (Eysenck has been posthumously relegated to the doghouse for a number of years now, as it appears that he may have played fast and exceptionally loose with some of his data). And, as we all know, the Big Five people have largely won the debate (with some caveats around the lack of universality of the Five Factor Model, etc.) — but you might occasionally run into someone who I would consider to be a “personality psych hipster” who prefers the 6-factor HEXACO model. The HEXACO people are the personality psych equivalent of the person who swears that listening to any mastering of Dark Side Of The Moon that isn’t the Black Triangle release gives them headaches, and they refuse to wear anything but Acronym jackets because — in spite of being excruciatingly esoteric — they’re “just better.”
But, I’m going to let you in on a dirty little secret. There is no “best” solution. Personality Theory isn’t an optimization problem because, in practice, it’s not especially clear that there even is an optimal solution that translates to all domains or research questions across varying levels of psychology. And, one thing that we like to commonly ignore is this: even in the most dominant models of personality (I’m looking at you, Big Five), they actually aren’t all cleanly separated as we all pretend they are. The standard line in Intro to Psychology classes is that we have these five beautiful factors — OCEAN — and we love this model because they’re all orthogonal and they seem to represent the “space” of personality incredibly well. But when we actually go out and measure them… they tend to all be inter-correlated to varying degrees. Put another way: they’re not as orthogonal as we all pretend they are (or want them to be or even – sometimes – need them to be.
Is this a huge problem with the Five Factor Model? Does this ruin everything that we do when we give people a Big Five questionnaire? Have these folks been lying to us? Why, those dirty tricksters… Well, not really, no. But it does bring us back around to a recurring trend in the study of personality and individual differences (and really, many, many various areas of psychology), which is this: we want to break things down into increasingly small bits in order to understand them better, which is a standard method in scientific inquiry. However, when we get mired in the atomic, we need to occasionally step back and look at the “big picture” of how it all fits together.
Enter: the K-factor (also called the “Super K-Factor” as well as the General Factor of Personality [GFP] or the “Big One” of personality). I’ll warn you in advance that the K-factor isn’t some widely-discussed thing in personality psych. It’s something that a lot of personality psych folk don’t talk about, and most of them certainly don’t like to talk about it for various reasons (some more legitimate than others). But, it is true that, in many samples, people notice over and over that most (and sometimes all) of the Five Factors are inter-correlated, often modestly, sometimes more strongly than they’d like. Well… if they all tend to reliably correlate with each other, what does that mean?
One of the standard interpretations of this that will come up in casual conversation is that the K-factor (and thus, really, the Big Five) seems to reflect in very, very general terms something like “socialization.” That is: in a zoomed-out sense, a person who has been well-socialized is more likely to be Extraverted, Conscientious, Open to Experiences, Agreeable, and Emotionally Stable. Sounds pretty reasonable, right? So far as I know, nobody who has this understanding of the K-factor would claim anything as sweeping as “well, then, the Big Five only reflects socialization” because, really, that’s a gross over-reach and unfair interpretation of variance accounted for here. However, it does raise some interesting questions about how this interpretation “trickles down” to our understanding of lower-level personality factors, traits, states, and so on.
A more complex interpretation of the K-factor is… well, complicated. However, if you’re interested in digging further, one important thing to note is that this factor seems to extend far beyond the Big Five. That is: it’s not simply an artifact of the Five Factor Model, or really, any N-factor models that you may like. Research has looked at this phenomenon by measuring things like ranging from relationships with others to beliefs in the meaningfulness of society to financial status, physical health, religiosity, subjective well-being… and on and on. There are a few proponents of this model and many, many detractors, and I’ll admit that I’m not especially well-read on the debates that go back and forth every few years on the subject. But, what I do know is this: the K-factor is an interesting, dirty little secret that a lot of personality psychologists kind of wish they didn’t have to know (or hear) about at all.
With that, I give you today’s RIP, which is another two-fer. Please do enjoy:
- Figueredo, A. J., Vásquez, G., Brumbach, B. H., & Schneider, S. M. R. (2007). The K-factor, covitality, and personality: A psychometric test of Life History Theory. Human Nature, 18(1), 47–73. https://doi.org/10.1007/BF02820846
- Rushton, P., & Irwing, P. (2011). The general factor of personality: Normal and abnormal. In The Wiley-Blackwell handbook of individual differences (pp. 132–161). Wiley Blackwell. https://doi.org/10.1002/9781444343120.ch5