This paper describes the standard version of the MRC Social Performance Schedule, which covers 21 behaviour areas such as hygiene, initiating conversations, etc. and was designed to be used with long-stay populations, within a hospital or the community. Data on the schedule's consistency over raters, informants, time, and situation showed that the reliability was reasonable over all four areas. The schedule is recommended as a standard instrument to describe the behavioural difficulties of patients likely to be dependent on psychiatric services for some time. Recent data are provided on the levels of problem behaviours found within both long-stay hospitals and community settings, together with references for more detailed behaviour profiles of such groups.