official statisticsany data collected and published by government departments. Such data have varied reliability and utility Although many of these data can be highly valuable to the sociologist, e.g. those presented in the CENSUS, GENERAL HOUSEHOLD SURVEY and EXPENDITURE SURVEY, major debates exist concerning the limitations of official statistics (e.g. CRIME STATISTICS). The way in which statistics are collected, e.g. as a by-product of the work of administrative agencies, sometimes by a large number of untrained recorders, can lead to major inconsistencies, unreliability and uncertainties about the meaning and worth of data. Also, statistics may possess positive or negative implications for the agencies and individuals that collect them, which can affect what is recorded. Finally statistics are always collected for some purpose, which will usually be different from that of the sociological researcher; above all, they are the result of a process of categorization and the attachment of numbers which involves inherent difficulties of the kind identified particularly by ethnomethodologists (see CICOUREL, 1964). A classic example of the issues that can arise is the debate concerning DURKHEIM's use of Suicide statistics (see Douglas, 1967). see also MEASUREMENT BY FIAT.
Collins Dictionary of Sociology, 3rd ed. © HarperCollins Publishers 2000