Openness to Experience · Facet O6
Liberalism in the Big Five has little to do with parties, ballots or policy. It measures how readily you reopen a question most people treat as settled: whether authority, custom and inherited rules keep having to justify themselves, or whether long service has already earned them the benefit of the doubt.
Free, no sign-up, results right away
The ten statements circle around one relationship: yours with rules you did not write. They ask about tradition, about authority, about whether established ways deserve another look. What they never ask is which answer you reached. Two people who agree on nothing concrete can land at the same point: the scale registers a stance, not a conclusion.
The facet sits inside Openness to Experience, the domain that collects appetite for the unfamiliar. Imagination aims it at the inner world, Artistic Interests at beauty, Adventurousness at unfamiliar places. Liberalism aims it at norms, and holds that slice almost alone: it travels with its own domain more weakly than any of its five sisters.
The name deserves a word. Liberalism here is a term from personality research, coined well before current party systems. It does the same job as Anxiety or Modesty elsewhere in the test: it points at a measured tendency, not at a group you belong to.
In the research literature this territory usually carries the label Values, and older writing calls it unconventionality. The wording used here follows the public domain items published by Johnson rather than any commercial questionnaire.
This is the facet whose name misleads people most, so these three lines matter more than usual.
The test never asks about taxes, borders, parties or elections, so no result here points at a ballot. The scale registers how you hold rules in general, and people at the top of it argue for change in every direction, including the return of arrangements dropped long ago.
A low score gets read as devoutness and a high one as its absence, and both readings overshoot. The statements ask how readily settled rules are reopened, never what a person believes. A believer who argues with their own tradition scores high, and someone with no faith scores low if inherited structure looks load bearing to them.
Questioning conventions and giving people a hard time are separate things, and the data is clear. This facet correlates 0.01 with Cooperation, minus 0.02 with Anger and 0.04 with Trust, which on a scale where 1.00 means identical is no relationship at all. Asking why a rule exists says nothing about how you treat people.
Neither end of this scale is the enlightened one. Each buys something a group needs and charges for it in a currency the other end rarely notices.
You treat a rule as an argument that can be reopened, so you notice when a practice has outlived the reason it was built for. Groups that include someone like you do not carry dead procedure for decades. The bill comes as friction: every reopened question costs the room time, and several at once remove the ground people stood on.
Your first question to any rule is why. Conventions have to earn their keep with you, and authority gets your respect only after its arguments do.
You read inherited arrangements as compressed experience: somebody had a reason, and it is often invisible from here. That makes you steady to work with and a careful custodian of what breaks quietly. The cost is inertia. A rule that has run out of purpose keeps its place, because nothing in you pushes to reopen the file.
You respect what has stood the test of time. Rules and traditions earn your defense, and you change positions slowly, for good reasons.
About four people in ten sit in the middle band, and that middle is a working stance rather than the absence of one. A few rules get reopened, most are left alone, and which is which depends on the case. Repair without demolition.
Low, average and high follow the 30/40/30 percentile convention used by Johnson, not any political scale. The score describes how you hold rules, not whether your conclusions are sound.
The raw score is the sum of ten answers, so it runs from 10 to 50. The curve below comes from 135,947 completed profiles, and it sits well below the curves of the neighbouring facets.
| Percentile | Raw score |
|---|---|
| 10 | 21 |
| 25 | 25 |
| 50 | 30 |
| 75 | 35 |
| 90 | 40 |
Half of all profiles fall between 25 and 35 points, the median is 30, and a raw 30 lands on the 52nd percentile. The other five facets of Openness average between 35.3 and 40.2 points, so this one sits below every one of them, by five points at the nearest and ten at the furthest. Percentiles compare you with people of your sex and age.
Two group differences get assumed on a scale with this name, and neither one turns out to be large.
Women average 29.97 points and men 29.66, a difference of 0.31 on a 40 point range. Only two of the thirty facets split the sexes by less than that. Age moves more: both curves lose one to two points between the under 21 and over 60 groups, while the conscientiousness facets climb. Those oldest cells hold about 250 people each, thin enough to read lightly.
This is the loosest scale in the test. Its strongest relationship with anything is minus 0.28 with Orderliness, a facet from another domain, and nineteen of its twenty nine partners stay under 0.15. Openness is already the loosest family, and this is its loosest member.
A high score earns its keep when a process is being reviewed. You ask what a step protects against, and the honest answer is sometimes nothing. The same instinct is expensive mid delivery, when the question is settled and the team needs it to stay settled. A low score reads as stewardship: you keep the memory of why a rule exists, so old accidents stop repeating.
At home the disagreement is rarely about the answer. It is about whether the question is open at all. One partner treats the way a holiday is kept or a child raised as something to work out from scratch, the other as something received and worth passing on. Couples far apart here do fine once they see the argument is procedural.
At the high end the useful move is triage, not restraint. Almost any convention can be questioned, few are worth the cost, and the skill is telling them apart before goodwill is spent. At the low end the gain comes from putting a review on the calendar, since nothing here prompts you to reopen anything.
Twin research puts roughly half of the variation in Big Five traits on the genetic side, and most of the rest on experiences siblings do not share. That leaves real space for movement, and the movement is slow: across a decade people mostly keep their position relative to everyone else, and change arrives as drift.
Our norms show that drift running downward, while the conscientiousness facets climb steeply over the same span. One caveat weighs more here than anywhere else in the test: norm tables compare different people at different ages instead of following anyone through life, so ageing and generation look identical in them. On a facet about inherited rules that is no technicality, because people who were twenty in different decades inherited different rules.
High Liberalism next to high Orderliness behaves nothing like high Liberalism next to low Orderliness. One rebuilds the system and writes the new manual, the other only dislikes the old one. Combinations are what the paid analysis works on.
Free, no sign-up, results right away
The 300 item version measures Liberalism with ten statements, the 120 item version with four. Seven of the ten are keyed in reverse, so agreement lowers the score rather than raising it, and only one facet carries more. Clicking straight down one column therefore leaves a meaningless number. Answers run from 1 to 5, reversed items are flipped, and the norm table for your sex and age turns the sum into a percentile.
Alpha shows how closely the statements of one scale agree with each other. In research, 0.70 and above is usually treated as solid. We counted these numbers on Johnson’s open data, so they can be checked against the published values.
The statements are deliberately absent from this page. Reading the items in advance changes the way people answer them, so they appear only once the test has started.
300 questions · about 35 minutes
Take the full test120 questions · about 15 minutes
Take the short versionFree, no sign-up, results right away
Move the slider to the place you think you occupy next to other people, then take the test and see how far off the guess was.
0 means lower than almost everyone who took the test, 100 means higher than almost everyone.
This is a guess, not a measurement. The test compares your answers with norms from real people.
No, and the test could not tell. None of the 300 statements mentions a party, a policy or an election, so the data holds no political variable. The scale registers how readily you reopen a rule, and people at the top push in every direction, including back towards arrangements dropped long ago.
The label comes from personality research and predates the party sense of the word. In the literature the same scale often appears as Values, which describes the content better. We keep Liberalism because the test, your result and this page must call one scale by one name.
There is no good score. A high one buys willingness to revise and pays in friction, a low one buys continuity and pays in rules that outstay their use. Both are needed, and the useful information sits in how this score meets the rest of your profile.
Not a contradiction, and it is common. This facet ties to the rest of its own domain at 0.34, the second loosest such link in the test. The domain average hides the gap: the other five facets of Openness average between 35.3 and 40.2 points, this one 29.8.
The data says no. This facet correlates 0.01 with Cooperation, minus 0.02 with Anger and minus 0.01 with Altruism, about as close to nothing as measurement gets. Its strongest link anywhere is minus 0.28 with Orderliness, so the pattern is about wanted structure.
On Johnson's open data the ten statement version has an internal consistency of 0.79 and the four statement version 0.65, the lowest of all thirty short form facets. The short test shows the five domains well, but here four items give a hint, not a measurement.
Ten statements per facet in the long version, four in the short one, thirty facets either way. Nothing to sign, nothing to pay, and percentiles appear the moment you answer the last item.
Free, no sign-up, results right away