The hypotheses that self-reported symptoms are a valid proxy for physicians' ratings, and that people with different social characteristics mayvary in their recognition and evaluation of symptoms, were tested on volunteers attending the MRC Common Cold Unit. Eleven-hundred volunteers assessed the presence and severity of colds at the end of their stay at the unit, using the same two measures as a trained clinical observer (all ratings were double blind). On both measures there was an extremely high correlation between self- and observer-assessments, a finding encouraging for those using self-reports in community surveys. Although this high level of agreement was found within all the age, sex, occupational class and marital status groups, men were significantly more likely than women to over-rate their symptoms in comparison with the clinical observer. No such differences in over-rating were found by age, class or marital status. This finding of a sex difference in the tendency to over-rate physical symptoms is of major significance to theories of illness behaviour, and is particularly convincing because it is based on a double-blind comparison between self-reports and assessments made by a doctor.