The Science of Values: What Research Shows

Milton Rokeach, working in the 1970s, made a distinction that still holds up. He split values into two groups. Terminal values are the destinations you want your life to reach: a sense of accomplishment, lasting friendship, inner peace. Instrumental values are the ways you prefer to travel: being honest, being capable, being disciplined. One describes ends, the other describes means.

The split matters because people conflate the two and then wonder why their goals and their habits pull in opposite directions. You might hold freedom as a terminal value while behaving in tightly controlled, cautious ways day to day. Naming which layer a value lives on often explains that friction. Rokeach also argued that values are relatively few and organized by priority, not held in equal measure, which is the idea underneath any ranked result.

Shalom Schwartz took the next big step. Across surveys in dozens of countries, he found that basic human values organize into a recognizable structure that repeats across cultures. He grouped them into broad types, such as achievement, benevolence, security, and self-direction, and arranged them in a circle.

The circle's key insight is adjacency and opposition. Values near each other tend to be compatible and easy to pursue together. Values across the circle tend to conflict, so gaining ground on one usually costs you the other. Openness to change sits opposite conservation; concern for self sits opposite concern for others.

This is why a values result should not read like a list of things you happen to like. If everything scores high, the instrument is measuring your desire to look good, not your actual priorities. A useful map shows the trade-offs, the places where valuing one thing genuinely means giving up another.

Here is the measurement problem. Ask people to rate each value on a scale from one to five, and almost everyone rates almost everything near the top. Honesty, family, growth, integrity: who checks a low number next to those? This is social-desirability inflation, and it flattens the very differences a values test exists to find. Researchers call scale-based scoring normative, because each item is judged against an outside standard.

The alternative is ipsative measurement, where you compare items against each other rather than against a scale. Forced choice is the cleanest version: shown four statements, you pick the one most like you and the one least like you. You cannot rate them all highly, because the format will not let you. To choose one, you have to set another aside.

That constraint is the point. It surfaces genuine priority instead of general approval, and it blunts the pull toward answers that simply sound good. This test uses that format: 20 sets of four statements, one most and one least in each, which takes about five minutes. The result is a ranked top five drawn from ten values, shown on screen. Where the gap at the top is decisive, a single core value is named; where it is close, you see an honest cluster or a near-tie rather than a manufactured winner.

Forced choice fixes one problem. It does not fix all of them, and no reputable page should pretend otherwise.

Every values test is self-report, which means it captures how you see yourself on the day you take it, filtered through mood, recent events, and how you read the wording. People also differ in self-insight; some describe their values accurately, others describe an aspiration or a memory. And a snapshot cannot see context. The value you lead with at work may not be the one you lead with at home, and a single result cannot separate those settings.

This test is built on an original framework of ten values, with original items written for it. It is informed by the thinking above, but it does not use Rokeach's or Schwartz's questionnaires, and it does not claim to measure the categories they defined. Their work is described here for context, not borrowed as an instrument.

Just as important: this framework has not yet been empirically normed. There are no published reliability or validity figures, because the studies that would produce them have not been done. That is a deliberate statement, not an oversight. Treat the result as a well-designed self-report snapshot, a strong prompt for reflection and conversation, not a clinical or diagnostic measure. The right use is to read your ranking, notice what surprises you, and test it against how you actually spent last week.