The Say/Do Gap: Why Stated Preference Misleads

Ask people what they prefer and they answer with their slow, deliberate, justifying mind. They decide with a faster one. Respecting that gap is the difference between a tidy chart and a true one.

The practice

Treat stated preference as one input, not the answer. Where a choice is made fast and emotionally, measure the doing — not only the saying.

The gap between the survey and the shelf

A respondent in a survey has all the time in the world, an interface that invites consideration, and a quiet wish to look like a sensible person. A shopper at a shelf has none of those things. The first explains; the second reacts — and the choices they make can diverge sharply. That divergence is the say/do gap, and the gap between what people say and what they do is long-documented.¹

It is not that people lie. It is that the question and the moment of choice engage different machinery. Asked to explain a preference, we construct a reasonable-sounding story after the fact; in the moment, we mostly act before the story is written.

Why it happens

Three forces pull stated answers away from real behaviour. Rationalisation: we report the reasons we can articulate, not the ones that actually moved us. Social desirability: we shade answers toward what sounds good. And simple lack of access: much of what drives a choice is non-conscious, and you cannot accurately report what you cannot introspect.²

What to do about it

You do not close the say/do gap by writing a cleverer question — you close it by adding a different kind of measurement. Pair the explicit answer with a behavioural one: a timed choice, an implicit association, a reaction captured before deliberation. Treat the stated response as a hypothesis and let the behaviour test it. When the two agree, you can be confident; when they diverge, you have just found the most interesting thing in the study.

¹The divergence between stated attitudes and actual behaviour is long-documented: LaPiere (1934), “Attitudes vs. actions,” Social Forces 13(2), 230–237; and on the intention–behaviour gap, Sheeran (2002), “Intention–behavior relations: A conceptual and empirical review,” European Review of Social Psychology 12(1), 1–36.²On the limits of introspective access to the causes of our own behaviour, see Nisbett & Wilson (1977), “Telling more than we can know: Verbal reports on mental processes,” Psychological Review 84(3), 231–259.