A study published in the Journal of Personality and Social Psychology, co-authored by Michael Wilmot, a postdoctoral fellow in the Department of Management at UTSC, suggests that the Type A personality may not be a personality type after all.
The researchers replicated a 1989 study that produced the most influential empirical support for Type A and Type B personality theory. Despite conducting direct and conceptual replications to test the same idea, Wilmot and his colleagues were unable to find significant evidence for the existence of Type A behaviour as a type.
Type A and Type B personality theory is the idea that people fall under two disparate categories: either they are competitive, organized, and impatient — a Type A personality — or relaxed, non-competitive, and patient — a Type B personality.
This theory was the result of an observation by two cardiologists, Meyer Friedman and Ray Rosenman, who had found that patients who couldn’t sit still in the waiting room were more likely to develop coronary heart disease. They attributed this pattern of behaviour to a Type A personality.
Further research demonstrated a similar relationship between personality and other cardiac health concerns: Type A personality individuals had higher blood pressure and a greater risk of heart disease than their relatively laid-back Type B counterparts.
With Wilmot’s findings, there is now no normal personality construct that has uncontested evidence of a typological structure.
The existence of personality types is controversial. We typically conceptualize personality as something that goes along a continuum, rather than as a presence or absence of certain traits.
“I think people like the idea of personality types because categories are important conceptual tools we use to make sense of the world,” said Wilmot in an email to The Varsity. “However, just because we frequently like to categorize things, doesn’t mean that those things are truly categorical in some fundamental and real way.”
Most psychologists agree that personality variables are dimensional. A 2012 meta-analysis led by Australian and Belgian researchers found that few investigations into normal personality variables yielded typological results, yielding “little persuasive evidence” of types, and thus “many influential taxonic findings of early taxometric research are likely to be spurious.”
“From a scientific standpoint, I think the default should be to assume that all personality variables are dimensional unless proven otherwise,” wrote Wilmot. “As a result, the burden-of-proof rests on the person trying to show evidence for a typology. What is more, that typological evidence should be replicable across samples, researchers, and relevant data analytic techniques.”
Since it was first conceptualized in the 1970s, Type A and Type B personality theory has remained the only typological personality construct with largely unchallenged evidence. According to Wilmot, one reason for its long-standing, undisputed status is that most psychologists had shifted away from this model, focusing instead on the Big Five personality model, which proposes five traits along a continuum. But there is another reason that is more pertinent: most high-brow scientific journals seldom publish replication studies, meaning that most researchers would rather test new theories in their field than scrutinize the evidence of older theories.
Having examined the 1989 study, Wilmot and his colleagues plan to continue working toward their original goal of conducting a meta-analysis on hard-driving and competitive personality traits with respect to work and academic outcomes.
They will also aim to better understand where these traits “fit into the general taxonomy of personality traits,” such as the Big Five model.
“Even though personality types are not real, typological models of personality can still have some usefulness in the real world. If it can help to partially explain or predict other people’s behavior, then a typological model can have value,” wrote Wilmot. “Put differently, a flawed model of personality is better than no model at all. However, such a model should be taken as a useful starting place, not as a final destination.”