The Big Five personality traits: five spectrums, not five types

The Big Five are five broad dimensions of personality: openness, conscientiousness, extraversion, agreeableness and neuroticism, which spell OCEAN. Each is a continuum. Everyone has a position on all five, and none of them is a box that a person is either in or out of.

The model was not built from a theory of the mind. It comes from the words people use to describe each other, which researchers collected, rated and analysed until the descriptions fell into five broad groups. The name "Big Five" says that each group is broad. It does not say the traits are the greatest ones.

No pole of a trait is simply good or bad. Conscientiousness has the most consistent link to performance at school and at work, but that is a finding about group averages, and how much a pole helps depends on the setting.

• OCEAN stands for openness, conscientiousness, extraversion, agreeableness and neuroticism. Each is scored as a position on a scale.
• The traits come from language. Allport and Odbert listed 17,953 English personality terms in 1936, and a 1961 US Air Force report found "five relatively strong and recurrent factors" in trait ratings.
• A 1996 twin study put broad genetic influence on the five traits between 41 and 61 percent. That describes differences within a group, not how much of one person's personality is inherited.
• The Big Five describe personality and do not explain it.
• Trait-outcome findings mostly hold up: in a replication project, 87 percent of 78 published trait-outcome links held up, at about 77 percent of their original strength.
• The Myers-Briggs Type Indicator sorts people into 16 types by four either/or preferences. The Big Five give five scores on continuous scales.

The five traits at a glance

Each trait is a broad domain that contains narrower facets. The table gives the definitions from a standard handbook chapter on the model and the six facets that the NEO PI-R, the best-known questionnaire, assigns to each domain (John, Naumann and Soto, 2008).

The Big Five literature writes extraversion with an "a", as this article does. The APA Dictionary of Psychology lists extroversion as the variant spelling of the same word.

Spectrums, not types

A type puts a person in a category: one kind or the other. A dimension asks how much. On the Big Five, introversion is not a kind of person set against extraverts. It is the low end of one scale, and the APA Dictionary describes extraversion as a trait that, like introversion, "exists on a continuum of attitudes and behaviors".

The best-known type system is the Myers-Briggs Type Indicator. The Myers and Briggs Foundation describes four preference pairs (E-I, S-N, T-F and J-P) that combine into 16 types, and says the assessment "sorts for preferences; it does not measure traits, ability, skill, or character". That is the structural difference. The Big Five ask how much of a trait a person shows, and the MBTI records which side of a pair a person prefers.

In 1989, McCrae and Costa compared the MBTI with the Big Five in 468 adults aged 19 to 93. They found "no support for the view that the MBTI measures truly dichotomous preferences or qualitatively distinct types". The four MBTI indices instead measured four relatively independent dimensions, and these matched aspects of four of the five Big Five dimensions (McCrae and Costa, 1989). The Foundation's own overview describes the instrument as scientifically reliable and valid, so the 1989 study tests a narrower claim: that people fall into distinct types.

Where the five came from

The model starts from the lexical hypothesis: the differences between people that matter most become words in the language. In 1884, Francis Galton tried a first count. He sampled the index of Roget's Thesaurus and estimated that it held "fully one thousand words expressive of character" (Galton, 1884).

In 1936, Gordon Allport and Henry Odbert did the job properly. They went through Webster's New International Dictionary and listed 17,953 words that distinguish one person's behaviour from another's. The first of their four columns held 4,504 terms, 25 percent of the list: names for "consistent and stable modes of an individual's adjustment to his environment", such as aggressive, introverted and sociable. A second column of 4,541 terms held temporary states, moods and activities (Allport and Odbert, 1936).

Raymond Cattell took roughly 4,500 trait terms from that list and cut them to 35 variables in the 1940s. In a 1961 US Air Force technical report, Ernest Tupes and Raymond Christal reanalysed correlation matrices from eight samples of trait ratings and found "five relatively strong and recurrent factors and nothing more of any consequence", as John and colleagues quote them. Warren Norman replicated the structure in 1963, and the factors were first labelled Extraversion or Surgency, Agreeableness, Conscientiousness, Emotional Stability and Culture.

The fifth label did not last. The items that loaded on it were "open" and "intellectual" characteristics such as artistic, curious, original and wide interests, and reading the factor as Culture was a historical accident. Lewis Goldberg chose the name "Big Five" in 1981 to stress that each factor is extremely broad, not that the factors are intrinsically great, so the name does not mean that personality reduces to five traits.

The questionnaire route reached the same five. Paul Costa and Robert McCrae designed the NEO Personality Inventory, published in 1985, to measure neuroticism, extraversion and openness. In 1983 they saw that their system matched three of the Big Five factors, and they added preliminary scales for agreeableness and conscientiousness. The 240-item NEO PI-R of 1992 then gave all five domains six facets each. That route is usually called the five-factor model, and the agreement between the two routes led to a dramatic change in how widely the five factors were accepted (John, Naumann and Soto, 2008).

The five traits, one at a time

Each trait below comes with one finding. The findings are group averages from meta-analyses, so they say what is more or less likely across many people, not what a given person will do.

Openness

Openness is one of the debated labels: John and colleagues call it broad enough but somewhat vague. In Barrick and Mount's 1991 meta-analysis of job performance, openness and extraversion were both valid predictors of training proficiency across occupations (Barrick and Mount, 1991). In a 2022 meta-analysis of academic performance, openness had weak overall importance but mattered in the early school years (Mammadov, 2022).

Conscientiousness

Conscientiousness has the most consistent record. Barrick and Mount studied five occupational groups and three performance criteria, and conscientiousness was the one dimension that related to all of the criteria in all of the groups. For school, a 2009 meta-analysis with cumulative samples above 70,000 found that conscientiousness correlated with academic performance largely independently of intelligence (Poropat, 2009).

The 2022 meta-analysis pooled 267 samples with 413,074 participants. Cognitive ability and the Big Five together explained 27.8 percent of the variance in academic performance. Of that explained share, cognitive ability accounted for 64 percent and conscientiousness for 28 percent. Conscientiousness therefore matters for grades, but most of the variation in grades is explained by neither.

Extraversion

Extraversion is about energy toward the social world, with sociability, activity, assertiveness and positive emotion among its facets. It is not the same as being loud, and its low end is introversion. Barrick and Mount found it to be a valid predictor of performance for managers and in sales, the two occupations they describe as involving social interaction. That fits the point that how much a pole helps depends on the setting.

Agreeableness

Agreeableness contrasts a prosocial orientation with antagonism. John and colleagues point out that the label is a trap, because it suggests a tendency to agree with others and so wrongly implies submissiveness, which is closer to the introverted end of extraversion. In Poropat's 2009 meta-analysis, academic performance correlated significantly with agreeableness, conscientiousness and openness.

Neuroticism

Neuroticism contrasts emotional stability and even-temperedness with negative emotionality. The APA Dictionary describes it as "a chronic level of emotional instability and proneness to psychological distress". Many studies score the trait from the other end and call it emotional stability, as Barrick and Mount did.

Neuroticism is not a diagnosis. A 2010 meta-analysis of 175 studies found that groups with anxiety, depressive and substance use diagnoses scored high on neuroticism on average (mean Cohen's d = 1.65), and that it was the strongest correlate across the board (Kotov and colleagues, 2010). That is a result about groups of patients: a high score is not a diagnosis, and it does not mean a person has such a condition. If worry or low mood lasts or disrupts daily life, a doctor or a licensed mental health professional can assess it and a Big Five questionnaire cannot. Our article on the psychologist, psychotherapist and psychiatrist explains who does what.

How the Big Five are measured

Most Big Five measures are self-report questionnaires in which people rate how well statements describe them. Four are worth knowing.

The International Personality Item Pool holds more than 3,000 items and 250 scales that anyone may use without paying a fee, among them scales that measure constructs similar to the five NEO domains. The NEO details come from the publisher and from John and colleagues.

The instruments do not define every facet in the same way. The NEO PI-R puts warmth under extraversion, while lexical studies put a very similar warmth and affection facet under agreeableness. John and colleagues link this to the finding that the NEO-FFI's extraversion scale converged much less with other instruments than those instruments did with each other. Scores from two different tests are therefore not interchangeable.

Self-report also has weak points. One study of the Big Five Inventory found small but systematic acquiescence effects, a tendency to agree with items whatever they say, and Paulhus and John argued that part of the overlap among the factors may reflect self-enhancing bias, both as reported by John and colleagues. Observer forms are one check: the NEO PI-3 comes with informant forms, and the NEO PI-R scales have shown convergent validity against spouse and peer ratings.

What the model does not do

It describes people, it does not explain them

The five dimensions, in John and colleagues' words, represent no particular theoretical perspective. They were derived from the natural-language terms people use, so the model says that people differ along five broad lines and is silent on why.

Twin studies add one piece of the answer. In a 1996 study of 123 pairs of identical and 127 pairs of fraternal twins, broad genetic influence on neuroticism, extraversion, openness, agreeableness and conscientiousness was estimated at 41, 53, 61, 41 and 44 percent (Jang, Livesley and Vernon, 1996). Heritability is a statistic about differences within a sample. It does not say that 41 percent of one person's neuroticism is inherited.

Links to life outcomes shrink on replication

The Life Outcomes of Personality Replication Project ran preregistered, high-powered replications (median N = 1,504) of 78 published trait-outcome associations. Of the replication attempts, 87 percent were significant in the expected direction, and the effects were typically 77 percent as strong as the originals. The author concluded that the literature gives a reasonably accurate map of trait-outcome links that would still benefit from better replicability (Soto, 2019). The project tested links between traits and outcomes, not the five-factor structure itself.

The structure may not hold in every society

A 2007 study translated the BFI into 28 languages and gave it to 17,837 people in 56 nations. The five-dimensional structure was robust across major world regions, and people from South America and East Asia differed significantly in openness from those in other regions (Schmitt and colleagues, 2007).

A 2013 study of the Tsimane, a largely illiterate forager-horticulturalist society in Bolivia (n = 632), reached a different result. The Tsimane completed a translation of the 44-item BFI, and the authors "failed to find robust support" for the five-factor model. They argue that Tsimane personality shows two principal factors and that the structure may not be the same across all human societies (Gurven and colleagues, 2013).

Other models exist

In 2004, Ashton and colleagues analysed lexical studies in seven languages (Dutch, French, German, Hungarian, Italian, Korean and Polish) and found a similar six-factor solution. One of its factors, honesty-humility, is not among the Big Five. Its agreeableness factor includes anger and ill temper at the negative pole, and its emotionality factor leaves anger out, so neither matches the Big Five factor of the same name (Ashton and colleagues, 2004). The resulting HEXACO model has honesty-humility, emotionality, extraversion, agreeableness, conscientiousness and openness, and its inventory comes in self-report and observer forms (HEXACO website). The Big Five are the dominant model, not the only one.

Traits, states and defences

A trait, in Allport and Odbert's first column, is a stable mode of adjustment, and a mood is temporary. Defence mechanisms are a different kind of idea: the unconscious operations that Freud's tradition proposed for keeping unwanted thoughts and feelings out of awareness (our article on defence mechanisms lists twelve). A Big Five score claims no such mechanism. It describes the pattern a person shows, not the process behind it.

Learning the five for an exam

OCEAN already gives you the five names. The part that takes drilling is each trait's contrast, its facets and the finding attached to it. Our guide to memorising psychology terms explains why such sets are learned by testing yourself, not by rereading. In Psychology World you can save terms to a named list and drill that list with flashcards, multiple-choice questions once it has four or more words, or typing. A Personality list with the five traits, their contrasts and the facet names is a good first set.