The Big Five Personality Traits

In brief

The Big Five is the dominant model of personality structure in modern psychology. It claims individual differences in personality organize under five broad factors, remembered by the acronym OCEAN: openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism. Each is a spectrum, not a type: everyone has some standing on all five, and the traits are statistically close to independent of one another. The model traces to Gordon Allport and Henry Odbert's 1936 catalog of English trait words, was narrowed by Raymond Cattell's factor analysis in the 1940s, rediscovered as a five-factor solution by several independent teams from 1949 through 1963, named the "Big Five" by Lewis Goldberg in 1981, and turned into a standard instrument by Paul Costa and Robert McCrae's NEO Personality Inventory in the 1980s and 1990s.

The full treatment

The problem it answers

Early twentieth-century personality psychology had a proliferation problem. Freudians, Jungians, humanists, and dozens of test builders each proposed their own traits and scales, with no shared vocabulary and no way to compare one system to another. Allport and Odbert's response, in 1936, was the lexical hypothesis: if a personality difference matters enough for people to notice, it will leave a trace in ordinary language. They combed an unabridged English dictionary and pulled out nearly eighteen thousand words describing individual differences in behavior, several thousand of which named comparatively stable traits. That list was too large to use. The Big Five is, at root, the end product of a decades-long effort to compress it.

How it works

Cattell reduced the list by clustering synonyms, then ran the result through factor analysis, a statistical technique that groups variables which rise and fall together into a smaller number of underlying dimensions. He arrived at sixteen factors, published as the 16 Personality Factor Questionnaire in 1949. Other researchers reanalyzing similar data kept finding a simpler structure underneath: Donald Fiske in 1949, then Tupes and Christal in 1961, then Warren Norman in 1963, each recovering close to the same five factors from independent samples using different methods. Goldberg's 1981 paper gave the pattern its name. Costa and McCrae then built and repeatedly revised a self-report questionnaire, the NEO Personality Inventory, culminating in the 1992 NEO PI-R, which measures each factor along six narrower facets and remains the model's most used instrument.

What it claims

The Five-Factor Model is explicitly descriptive rather than explanatory. It does not say why these five dimensions exist, only that they are the smallest set that reliably captures how people differ once redundant trait words are collapsed. Each names a spectrum: openness (curiosity and appetite for novelty versus a preference for convention), conscientiousness (organization and self-discipline versus spontaneity), extraversion (sociability and assertiveness versus reserve), agreeableness (cooperation and warmth versus skepticism), and neuroticism (a tendency toward anxiety and instability versus calm resilience). Scores on the five are treated as fairly stable across adulthood and moderately heritable, though not fixed.

The key study or demonstration

Tupes and Christal's 1961 report is the model's quiet founding document. Working for the United States Air Force, they factor-analyzed peer ratings of personality across eight separate samples, using trait scales developed by Cattell, and found the same five factors emerging almost every time despite the different raters and settings. The finding sat in an obscure technical report and was largely ignored for two decades until Norman's 1963 replication and Goldberg's advocacy pulled it into mainstream view. Its later strength is cross-cultural: McCrae and Costa reported in 1997 that the five-factor structure emerged from translated personality inventories across dozens of languages, and further work extended the replication to roughly fifty cultures, evidence the structure is not an artifact of English vocabulary alone.

Related distinctions

Hans Eysenck built a rival, longer-running model from the 1940s onward, arguing personality reduces to just three biologically grounded dimensions: extraversion, neuroticism, and (added later) psychoticism, together the PEN model. Eysenck tied extraversion to differences in baseline cortical arousal and treated his smaller set as more parsimonious and more clearly linked to physiology than the Big Five. More recently, Kibeom Lee and Michael Ashton's lexical studies across many languages found a six-factor solution fit the data better than five, adding Honesty-Humility (sincerity and fairness versus manipulativeness) and renaming neuroticism's counterpart Emotionality. Their HEXACO model is now a serious competitor, especially in research on unethical and antisocial behavior, which the extra factor predicts better than the Big Five alone.

Lineage

The line runs from Allport and Odbert's 1936 catalog, through Cattell's factor-analytic reduction in the 1940s, to the independent five-factor rediscoveries of Fiske (1949), Tupes and Christal (1961), and Norman (1963), to Goldberg's 1981 naming, to Costa and McCrae's NEO inventories (1985, revised 1992). Eysenck's three-factor model developed in parallel from the 1940s onward, sharing the factor-analytic method but not the five-factor conclusion. HEXACO followed in the early 2000s as a lexical-tradition rival built on the same method applied more broadly across languages.

The strongest case for it

The model's strength is convergence: independent researchers, using different data (dictionary trait words, self-ratings, peer ratings) and different samples across decades, kept arriving at nearly the same five factors, unusual in a field where theories often just reflect their theorist. Its predictive record is substantial. Brent Roberts and colleagues, in a 2007 meta-analysis in Perspectives on Psychological Science, found Big Five traits predict important life outcomes, mortality, divorce, and occupational attainment among them, about as well as socioeconomic status or IQ do. The traits also show meaningful heritability in twin studies and replicate across dozens of cultures, evidence the structure is not a quirk of English or of any single test.

The strongest case against it

Walter Mischel's Personality and Assessment (1968) opened what became known as the person-situation debate, arguing broad trait scores correlate only weakly, often around 0.3, with any single instance of actual behavior, since situations shape behavior more than stable dispositions do. The debate was never resolved cleanly in favor of either side; the field's working compromise is that traits predict behavior aggregated across many occasions far better than any one occasion, and that traits and situations interact rather than compete.

Jack Block's 1995 paper pressed a deeper worry: factor analysis is sensitive to which variables and rotation methods a researcher chooses, so the "discovery" of exactly five factors partly reflects those choices, and other solutions fit comparably well under different assumptions. Dan McAdams, in a 1992 critical appraisal, argued the Big Five describes what he called the "psychology of the stranger": traits capture how a person strikes you on brief acquaintance but say little about their goals or the life story that gives their behavior meaning. Ashton and Lee's HEXACO research is itself a standing objection from within the same tradition, arguing Honesty-Humility captures real variance the Big Five misses.

Where it stands now

The Five-Factor Model remains the working standard in personality and industrial-organizational psychology, embedded in instruments like the NEO PI-R, the shorter Big Five Inventory, and the public-domain International Personality Item Pool. It has not silenced its critics: the person-situation debate reshaped how psychologists talk about prediction rather than ending trait theory, and HEXACO has carved out a niche in research on dishonesty and workplace misconduct, where its sixth factor earns its keep. Most researchers treat five versus six factors as an open question, while using the Big Five as the default language for describing how people differ.

Test yourself

Think of the person you know best. Would you place them high or low on each of the five, and could you defend each placement with a specific memory rather than a vague impression? If one trait is easy to score and another isn't, that gap is worth noticing: it may mean the trait fits them less well, or that you know less about that side of them than you assumed.