What the Big Five Test Actually Measures

The Big Five test measures five broad domains of personality: openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism. Each domain is a continuous spectrum, and the result doesn’t put you in a box; it places you at a point on each of the five scales.

That’s the short answer. The long one is more interesting, because “what it measures” hides three questions that few pages answer: at what level of detail it measures, from what source of information, and what it leaves out. Where the five factors came from is covered in the article on the lexical hypothesis; here we focus on the instrument itself.

Domains and facets: two levels of detail

The Big Five are very broad categories. Within each one sit more specific traits, called facets, which can point in different directions in the same person.

The reference inventory, Costa and McCrae’s NEO-PI-R (1992), has 240 items and splits each domain into six facets, 30 in total:

Domain	NEO-PI-R facets
Neuroticism	anxiety, angry hostility, depression, self-consciousness, impulsiveness, vulnerability
Extraversion	warmth, gregariousness, assertiveness, activity, excitement-seeking, positive emotions
Openness	fantasy, aesthetics, feelings, actions, ideas, values
Agreeableness	trust, straightforwardness, altruism, compliance, modesty, tender-mindedness
Conscientiousness	competence, order, dutifulness, achievement striving, self-discipline, deliberation

The facet level matters more than it seems. Two people with the same overall extraversion score can be very different: one highly assertive and not very gregarious (she runs meetings with ease but would rather have dinner with two friends than go to a party); the other highly gregarious and not very assertive. The domain makes them equal; the facets tell them apart.

Other instruments use fewer facets. Soto and John’s BFI-2 (2017) has 60 items and three facets per domain, 15 in total. Ten-item versions can only give a rough estimate of each domain.

How an answer becomes a score

The most widely used Big Five tests are self-reports: statements about your own behavior that you rate on an agreement scale, usually from 1 to 5.

Many of the statements circulating online (“I am the life of the party,” “I have a vivid imagination,” “I get stressed out easily,” “I pay attention to details”) come from the International Personality Item Pool (IPIP), the public-domain item bank that Lewis Goldberg (1999) launched so that anyone could do research without paying for licenses.

The calculation is simple:

1. Each item belongs to a domain and, in the longer versions, to a facet.
2. Some items are reverse-keyed (“I prefer to go unnoticed” counts against extraversion), and their value is flipped before adding up. This reduces the effect of people who tend to say “yes” to everything.
3. The items on each scale are added up or averaged.
4. In professional instruments, that raw score is compared with norms from a reference population (percentiles or T-scores). How to read that comparison is explained in the OCEAN scale.

The test on this site follows the same logic. It comes in three versions: a quick one with 10 statements (two per trait), a standard one with 50 (ten per trait), and a full one with 120 statements, 24 per trait, grouped into six facets of four items each. In the full version, half of the items are reverse-keyed. It doesn’t use population norms: it shows your relative score on each domain and, in the full version, the breakdown by facet. How those statements are built is covered in what the Big Five test questions look like.

What you see in yourself and what others see

A self-report measures how you describe yourself, which isn’t exactly how you are or how others perceive you. That’s why research also uses observer ratings: partners, friends, coworkers.

The two sources agree moderately, and each one sees different things more clearly. Simine Vazire (2010) showed that other people are more accurate on highly visible or highly valued traits, such as intellect, while you are the better reporter of what goes on inside, such as anxiety. A meta-analysis by Connelly and Ones (2010) also found that ratings by others predict job performance better than self-reports do.

The practical takeaway is modest but useful: your result is your version of you. If someone who knows you well answered on your behalf, the profile would look similar, though not identical.

How reliable it is

In psychometrics, reliability has two sides: whether the items on a scale measure the same thing (internal consistency) and whether the result holds when the test is repeated (stability).

• Internal consistency. The domain scales of published inventories, such as the NEO-PI-R and the BFI-2, usually reach alpha coefficients around 0.8 or higher. Facets, with fewer items, come in somewhat lower.
• Stability. Traits are fairly stable, but not fixed. In Roberts and DelVecchio’s meta-analysis (2000), the correlation between two measurements several years apart rose with age: around 0.54 in the college years, 0.64 around age 30, and 0.74 between 50 and 70.

With fewer items, reliability drops. That’s why a 10-question version is good for getting your bearings, while a 120-question one gives a more precise profile.

What the test doesn’t measure

The Big Five describes tendencies in behavior, thought, and emotion, not the complete portrait of a person. Psychologist Dan McAdams (1995) proposed three levels for getting to know someone, and traits are only the first:

1. Traits: what the Big Five measures.
2. Characteristic adaptations: goals, values, motives, and interests; what a person wants and how they try to get it.
3. The life story: the narrative each person uses to make sense of their own biography.

Nor does it measure abilities. Scoring high on openness doesn’t mean being more intelligent, even though the two correlate moderately. And it isn’t a clinical tool: high neuroticism describes greater-than-average emotional reactivity, not a disorder.

The five domains aren’t fully independent, either. Colin DeYoung (2006) found that they cluster into two broader factors, stability (agreeableness, conscientiousness, and low neuroticism) and plasticity (extraversion and openness). The test doesn’t score them, but they help explain why certain traits tend to go together; they’re explained in Big Five personality theory. And the five-factor structure doesn’t replicate in the same way across every culture, one of the best-documented criticisms of the model.

If you’d like to see your profile across the five domains, and across their facets if you choose the full version, you can take our Big Five test.

Sources

• Connelly, B. S. and Ones, D. S. (2010). An other perspective on personality: Meta-analytic integration of observers’ accuracy and predictive validity. Psychological Bulletin, 136(6), 1092-1122.
• Costa, P. T. and McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Psychological Assessment Resources.
• DeYoung, C. G. (2006). Higher-order factors of the Big Five in a multi-informant sample. Journal of Personality and Social Psychology, 91(6), 1138-1151.
• Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. In I. Mervielde, I. Deary, F. De Fruyt and F. Ostendorf (Eds.), Personality Psychology in Europe (Vol. 7, pp. 7-28). Tilburg University Press.
• McAdams, D. P. (1995). What do we know when we know a person? Journal of Personality, 63(3), 365-396.
• Roberts, B. W. and DelVecchio, W. F. (2000). The rank-order consistency of personality traits from childhood to old age: A quantitative review of longitudinal studies. Psychological Bulletin, 126(1), 3-25.
• Soto, C. J. and John, O. P. (2017). The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power. Journal of Personality and Social Psychology, 113(1), 117-143.
• Vazire, S. (2010). Who knows what about a person? The self-other knowledge asymmetry (SOKA) model. Journal of Personality and Social Psychology, 98(2), 281-300.