This paper is concerned with a study commissioned by a local authority in London to check reactions of residents to a scheme by which traffic re-routing eased congestion on most roads at the inevitable expense of increased congestion on some. Pressure groups, mainly representing the residents adversely affected, had presented evidence of opposition at a higher degree of intensity than that shown by a representative sample. The paper continues with a discussion of the dangers of "amateur" research which does not used those aware of research procedures and problems. But it stresses, too, the fundamental dilemma of a survey concerned with an issue on which (unlike most of those in market research) everybody is involved whether or not they are aware of it. To inform the ignorant in the sample makes them no longer representative of the relevant universe. An experimental survey is described in which, before answering certain questions, respondents were given the opportunity to study in great detail the reasons behind the scheme under study. Results would appear to indicate that such a technique creates more problems than it solves. In conclusion, a plea is made for more continuous research in local government. A problem can then be studied at all stages rather than simply at a single point in time.