Work values

Ask anyone what they want from work and the answer is a list: interesting problems, decent money, good people, some autonomy, the sense that it matters, and enough stability to sleep. Nobody says otherwise. Everybody wants all of it.

Which is why that list is worth nothing as information, and why the whole field of work-values measurement is built around a single manoeuvre: taking things away. A values instrument that lets you rate autonomy 5 out of 5 and security 5 out of 5 has learned nothing about you, because you have not yet been asked what happens when they pull in opposite directions — which is the only form the question ever takes in real life, at the moment you are choosing between two offers.

A work value is a tie-breaker. It is the thing you protect when you cannot protect everything. The measurement problem is that you cannot see a tie-breaker until there is a tie, so the scale has to manufacture one.

The technical name for the failure is that values are socially desirable across the board. Nobody rates "doing work that helps people" as unimportant. The result on an ordinary agreement scale is a ceiling: high scores on everything, small differences between items, and a profile that mostly records how generously the person uses the top of a scale rather than what they care about. That is a response style, and it swamps the signal.

Rokeach's answer in 1973 was to abandon rating entirely. His value survey handed people eighteen values on cards and made them rank-order the lot, on the argument that a value has no meaning except relative to other values. Everything since has been some negotiation with that idea: paired comparisons, forced choice between equally attractive statements, budget allocations where you have a hundred points and they run out. The trade-off is documented and real — forced-choice and Likert formats each buy something and cost something — and the cost of ranking is specific and worth stating.

Ranked scores are relative, not absolute. If autonomy comes out top, that means it beats your other values. It does not mean you want autonomy more than the next person does, and two people with identical profiles can differ enormously in how much any of it matters to them. A ranked result answers which, never how much. Nearly every misreading of a values report starts by forgetting that sentence.

The second thing worth knowing is that a tie is a finding. When two values come out level, that is not the instrument failing to resolve them. It is frequently the most useful line in the report, because a job that satisfies one and starves the other is precisely the job people take and cannot explain why they hate.

Instruments in this space descend from one of two lineages, and the difference is not cosmetic.

Super's tradition treats work values as job rewards. Donald Super's Work Values Inventory, from 1970, asked what you want a job to provide — achievement, independence, security, prestige, altruism, variety, economic return — and his broader life-span theory placed those alongside the other roles a person occupies. Dawis and Lofquist's theory of work adjustment formalised the same logic: the person has needs, the environment has reinforcers, satisfaction is the correspondence between them. Edgar Schein's career anchors belong here too, and add the useful claim that after a decade of working most people discover one anchor they will not give up, and only discover it when a job forces the choice.

In this tradition values are a catalogue. Items can be added and removed freely, and nothing in the theory says two of them must conflict.

Schwartz's tradition treats values as a structure with conflict built in. Shalom Schwartz's model, tested in twenty countries by 1992 and in far more since, arranges basic values in a circle where adjacent values are compatible and opposite values are structurally incompatible — you cannot simultaneously maximise self-direction and conformity, or achievement and universalism, because acting on one obstructs the other. Ros, Schwartz and Surkiss mapped work values onto that circle in 1999: intrinsic work values sit with openness to change, extrinsic ones with security and conformity, social ones with self-transcendence, prestige with self-enhancement.

The consequence is sharp. In Super's frame, "I want autonomy and stability" is an ordinary pair of preferences. In Schwartz's frame it is a report of a person standing on opposite sides of the circle, and the tension is not a personal failing to be organised away — it is the geometry. Most people asking why no job has ever felt right are describing this, and the Career Values test is built to name the opposing pair rather than declare a winner, which is the more honest output when the two are genuinely level.

The applied claim is that matching your values to an organisation's makes for a better working life. This one holds up better than most fit claims, with a caveat about what is being measured.

Kristof-Brown, Zimmerman and Johnson's 2005 meta-analysis separated the kinds of fit people talk about, and found person–organisation fit — the values match — related strongly to organisational commitment, satisfaction with the organisation and intent to stay, while person–job fit — the demands-and-abilities match — related more to job satisfaction and performance. Those are two different questions and the popular literature runs them together. Values fit tells you whether you will stay and identify with the place. It says much less about whether you will be good at the work or enjoy Tuesday.

The caveat: the biggest effects in that literature come from perceived fit, where the same person reports both their values and the organisation's. That design cannot separate fit from general contentment, and it inflates the correlation. Where researchers measure the two independently the relationship survives and shrinks. Still real, still smaller than the consulting decks say.

Three things make a values result usable instead of decorative.

Check it against your last three decisions, not your self-image. Values are cheap to endorse and expensive to act on, and the gap between the two is where most of the self-knowledge is. If stability ranks fourth and you have turned down two jobs in three years for being risky, believe the behaviour. The ranking is a hypothesis; your actual choices are the data.

Find the value behind the dissatisfaction. A job that is fine on paper and quietly draining is usually starving one specific value rather than failing generally, and naming which one changes what you do about it. Sometimes the answer is leaving. Often it is smaller — job crafting is the evidence-backed version of adjusting the job you have, and it works best when you know exactly which value you are buying back.

Do not confuse this with interests. Vocational interests describe what kind of activity holds your attention; values describe what you want the activity to be worth. They come apart constantly. Work you find fascinating that pays badly and work you find dull that buys you your afternoons are both perfectly coherent choices, and which of them is a mistake depends entirely on a ranking only you have.

And a caution the literature earns. Values shift with life stage — what a person will trade changes when a mortgage or a child arrives — so a profile from five years ago is a record, not a current reading. If the last one you took no longer sounds like you, the honest interpretation is usually that it was right then.