The Big Five model explained

The model measures where a person sits on five independent dimensions, each of them continuous. Nobody is an extravert or an introvert: everyone sits somewhere between the two, and most people fall around the middle. The five dimensions were kept because they recur from one study to the next, not because they exhaust the description of a person.

• Openness to experience: the relationship to novelty, abstract ideas, art and imagination.
• Conscientiousness: how someone organises, plans, meets deadlines and persists.
• Extraversion: where energy comes from in dealings with others, and appetite for social stimulation.
• Agreeableness: how someone handles relations with others, between seeking agreement and speaking bluntly.
• Neuroticism: the intensity and duration of emotional reactions, particularly under stress. Its other end is called emotional stability.

The acronym OCEAN takes the initials of the five factors: Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism. It is a memory aid, nothing more: the order of the letters implies neither a hierarchy nor an importance.

It comes from vocabulary, not from a theory of the brain. The starting point is a simple idea: the behavioural differences that matter in social life have ended up deposited in everyday language, in the form of adjectives. Analysing statistically how those adjectives cluster when people rate one another yields a small number of dimensions.

The first clear demonstration dates from 1961, in a technical report commissioned by the United States Air Force and written by Ernest Tupes and Raymond Christal. They analysed 35 rated traits across eight very different samples and found five recurrent factors. The text stayed hard to obtain for thirty years before being republished in 1992 in the Journal of Personality.

Two further steps established the model. In 1987, Robert McCrae and Paul Costa showed that the same structure emerges from different instruments and different sources: what a person says about themselves, and what those around them say. In 1990, Lewis Goldberg published the structure under the name that stuck. The model was not invented in one go: it was found again and again.

Because five is the number of dimensions that reappears most reliably when you change sample, instrument or judge. It is not a law of nature, it is a stable level of division. McCrae, Terracciano and the 78 members of the Personality Profiles of Cultures Project recovered the same structure from data collected across 50 cultures, published in 2005.

Below those five factors you can divide more finely: the BFI-2 inventory published in 2017 by Christopher Soto and Oliver John places three facets under each of the five factors, fifteen in all. Above them you can group. Five is a compromise between detail and stability, not a truth.

That is the main difference from typologies that sort people into a fixed number of profiles. A continuous position absorbs small variations in answering; a box does not. The price is that the result is harder to remember and less easy to share.

No, and this is the most widely misunderstood point of the model. None of the five factors is a desirable factor. High conscientiousness helps you hold to a plan and gets in the way when improvisation is needed. High agreeableness makes cooperation easier and saying no harder. High neuroticism often goes with a strong sensitivity to signals, including those others do not pick up.

Nor is a score a grade. On our questionnaire, the percentage shown is the share of points reachable on the items of a factor, not a rank against other respondents. Our weights and bands were set by the editorial team and come from no calibration on a population: 70 out of 100 therefore does not mean higher than 70% of people.

Finally, a profile is not fixed. The meta-analysis by Brent Roberts, Kate Walton and Wolfgang Viechtbauer, published in 2006 and covering 92 longitudinal studies, shows average increases in conscientiousness, emotional stability and social dominance, especially between the ages of 20 and 40. What you read today is a snapshot, not a signature.

• It measures neither skills, nor intelligence, nor values. A position on an axis says nothing about what you can do.
• It does not predict compatibility between two people. No pair of profiles is announced as matched by the structure of the model.
• It does not describe a disorder. None of the five factors, at either end, corresponds to a mental health category.
• It captures only what you agree to describe. Every answer passes through the image you hold of yourself, and through your mood at that moment.
• It does not replace observation over time. A person describes themselves differently depending on whether they have work or private life in mind.

These limits do not disqualify the model: they define its use. The Big Five is useful for putting words on tendencies, comparing two moments in your own life, or understanding why a situation costs one person more than another. It is not useful for ranking, selecting or predicting.