Values are the abstract priorities that sit above your opinions: what you treat as worth pursuing in a life. Psychologists have mapped them into a surprisingly tidy, near-universal pattern, a circle in which some values sit happily side by side and others pull against each other.
Values are the most abstract layer of what a person carries: not an opinion about a particular thing (that is an attitude, L3-02) but a standing principle about what is worth pursuing in general, freedom, security, success, tradition, kindness. The first thing to understand is counterintuitive: what matters is not which values you hold but how you rank them. Ask people whether they value honesty, or security, or adventure, and almost everyone says yes to almost everything. The information is in the trade-offs, what you will give up for what, which is why values are measured by priority rather than by yes or no.
An early influential split came from Milton Rokeach, who separated terminal values, the end-states you want from life (equality, a comfortable life, salvation), from instrumental values, the ways of behaving you prize on the way there (being honest, ambitious, courageous) (Rokeach, 1973). His survey, which had people rank lists of each, was the standard tool for two decades.
The deeper structure came from Shalom Schwartz, and it is the heart of this article. Across data from dozens of countries he found that values are not a loose list but a circle (Schwartz, 1992; refined to nineteen finer values in Schwartz et al., 2012). Two axes organise it. One runs from openness to change (self-direction, stimulation) to conservation (security, conformity, tradition). The other runs from self-enhancement (power, achievement) to self-transcendence (universalism, benevolence). The arrangement is not decorative; it is motivational. Values that sit next to each other are compatible, pursuing one tends to serve the other, while values on opposite sides of the circle conflict, so chasing one works against the other. You can prize both achievement and power easily, because they are neighbours, but prizing both power and universal compassion at the top of your list is a genuine strain, because they sit across the circle from each other.
What makes this more than an elegant diagram is how well it holds up. The same circle, with the same neighbours and the same oppositions, has been found across more than thirty countries in large representative samples, and across repeated waves over time (Bilsky, Janik and Schwartz, 2011; reviewed in Sagiv and Schwartz, 2022). In a field with plenty of findings that crumble on replication, the value circumplex is one of the sturdier structures we have.
The honest weak spot is the link to behaviour. If values are guiding principles, you would expect them to guide action, and they do, but loosely and unevenly. When researchers lined up each value against the behaviours that should express it, some matched up reasonably well (people who prize stimulation do seek novelty; people who prize tradition do observe it), while others, security, conformity, benevolence, achievement, barely tracked their matching behaviours at all (Bardi and Schwartz, 2003). A big reason is norms: people often do the done thing regardless of what they personally prize, which muffles the signal from values to action. So values motivate behaviour, but they are one quiet voice among several, and they predict broad tendencies (volunteering, voting, the general shape of a life) far better than any specific act.
There is also a deeper worry about direction. Because stated values are abstract and flattering, they are wide open to social desirability and self-presentation (the measurement problem from L2-08), and people are very good at reaching for a value after a decision to dress it up as principle. A purchase made for status gets retold as a choice about quality; a vote cast from group loyalty gets retold as a stand for freedom. Values are superb at explaining the story we tell about our choices, and less reliable as the thing that actually drove them.
Finally, values are neither fixed nor freely changeable. Tracking the World Values Surveys across 65 societies, economic development reliably pushes a culture's values toward the more rational, tolerant, and secular, yet the imprint of a society's religious and cultural heritage persists for generations underneath that drift (Inglehart and Baker, 2000). Values move, but along channels their history has already carved.
Almost all of this rests on self-report, with all its exposure to how people like to see themselves (L2-08), and the constructs are abstract enough that the gap between a stated priority and a real one can be wide. The circumplex is a map of how priorities relate, not a set of personality boxes to sort people into, and reading it as "there are four types of people" misses the point (the dimensional caution from L2-01). And the modest behaviour link means values are better for understanding and framing than for predicting what someone will do next.
Do values genuinely drive behaviour, or mostly justify it after the fact? How much do a person's priorities really shift across a lifetime, as against simply being re-described? And can we measure what someone actually prizes without relying on what they are willing to say they prize?
The usable core: read people by the priority order of their values, not by which they endorse, respect the circle's conflicts (no one can sit on opposite sides at once), and use values to shape the story and the framing rather than to predict the specific act.
Values are powerful for positioning and useless for forecasting a basket. They work when you align a brand with a coherent region of the circle: self-transcendence for purpose and sustainability, openness for novelty and early adoption, conservation for safety, heritage, and trust. The single most useful rule is the conflict structure: a brand cannot credibly stand for opposite values at once. "Daring, disruptive, and never changes" or "exclusive status symbol and humble everyman" reads as incoherent because those values sit across the circle from each other. Pick an arc and own it. And do not expect a customer's stated values to predict their actual purchases, that link is weak; values predict the narrative they will accept, not the transaction.
The Schwartz axes line up closely with real political divides: conservation and security on one side, openness and universalism on the other, which is why some disagreements feel less like arguments about facts and more like clashes of priorities. Persuasion lands far better when it is framed in the audience's own values rather than the speaker's, asking a security-prioritising voter to support a policy "because it protects what we have built" rather than "because it is progress". The same conflict structure warns that a coalition stretched across opposite values is inherently unstable.
Many of the hardest policy trade-offs are value conflicts in disguise, liberty versus security, growth versus environment, individual reward versus collective welfare, which map onto the opposite poles of the circle and are felt as impossible precisely because the values genuinely oppose. Naming the trade-off honestly, and framing a measure in terms of the values it serves rather than the ones it costs, does more than piling on facts. And behaviour-change campaigns travel further when the desired action is framed as an expression of what the audience already prizes.
Three habits. First, read priorities, not presence: everyone values most things, so the question is always what they will trade for what. Second, respect the conflicts, you cannot appeal to opposite values at the same time, so choose your arc of the circle and stay coherent. Third, use values to understand and frame, not to predict, since the link from stated value to specific behaviour is real but modest.
.png)