Personality type test A precise portrait: 5 traits and 30 facets

It is based on the five-factor model, the Big Five: five stable lines along which people differ across every culture studied. This is the most reproducible finding in personality psychology: the traits are measured consistently, persist over time and are visible from the outside — ratings by people who know someone well agree with that person’s self-report.

Within each trait the test distinguishes six narrower facets. Facets are what make a profile alive: two people with the same level of extraversion may differ in that one needs company while the other needs pace and activity.

• Neuroticism — sensitivity to stress: anxiety, anger, depression, self-consciousness, immoderation, vulnerability.
• Extraversion — pull towards people and activity: friendliness, gregariousness, assertiveness, activity level, excitement-seeking, cheerfulness.
• Openness to experience — interest in the new: imagination, artistic interests, emotionality, adventurousness, intellect, liberalism.
• Agreeableness — how you deal with people: trust, morality, altruism, cooperation, modesty, sympathy.
• Conscientiousness — how you deal with tasks: self-efficacy, orderliness, dutifulness, achievement-striving, self-discipline, cautiousness.

The heading says «personality type» because that is how people ask. The honest answer is that types do not exist — continuous traits do. When popular tests cut a scale in half and hand out one of sixteen boxes, the cut falls exactly where most people are. Two nearly identical people end up in different «types», and the difference between them is inflated into an opposition.

Hence the well-known problem: on retaking typological tests, between 39% and 76% of people get a different type, sometimes after only five weeks. The person has not changed — what changed is the side a borderline score fell on. A trait profile does not do this: it shows a position on a scale, not a side of a boundary.

Confidence intervals are something no comparable service we know of shows — neither free ones nor paid ones. Yet testing standards require reporting the margin of error for every scale, precisely so that people do not mistake random noise for a trait of their character.

• Five traits positioned against a normative sample — how far above or below you sit on each line.
• A confidence interval for every score — a range your true result lies in, not a single number.
• 30 facets as relief within the traits: what exactly stands out inside a trait and what does not.
• A read-out tied to your numbers rather than boilerplate under a label — stating plainly which differences are too small to interpret.

The test is built on IPIP-NEO-120, an open instrument from the International Personality Item Pool; the current edition is described in John Johnson’s 2014 peer-reviewed paper in the Journal of Research in Personality. IPIP items are in the public domain: they may be copied, translated, modified and used commercially without permission or fees.

Internal consistency of the five main scales in a large sample runs from 0.81 to 0.90, which is high. The traits themselves agree with the classic research inventory at about 0.82 on average. The 30 facets are less consistent — 0.75 on average, dropping to 0.63 for the weakest — which is why we present facets as direction, not as precise characteristics.

Russian wording of the items is our own work, with back-translation and a pilot, because existing Russian versions are either amateur and unvalidated or protected by their authors’ copyright.

• It does not measure intelligence: character traits and abilities are different things.
• It does not diagnose and does not replace a psychiatric or psychological assessment.
• It does not pick a profession and must not drive hiring decisions — the instrument’s author explicitly warns against using facets for important decisions about individuals.
• It does not predict fate, compatibility or success: traits describe tendencies, not a life script.
• It does not sort people into good and bad — every pole of every trait has its strength and its cost.

Back in 1949 the psychologist Bertram Forer handed students one and the same text assembled from an astrology booklet and got an accuracy rating of 4.26 out of 5 — everyone recognised themselves in it. The more flattering the wording, the more «accurate» a description feels, regardless of whether it is true.

A practical rule follows, and we try to keep to it: the text must be tied to your numbers and speak just as calmly about strengths as about inconvenient sides. A feeling of recognition is a poor test of accuracy. A good one is reproducibility: take the test again in a few months and compare.