Marsh C, Dale A, Skinner C. Safe data versus safe settings: access to microdata from the British Census. International Statistical Review 1994; 62 (1): 35-53.

Census offices are squeezed between users demanding ever more data and concerns about data confidentiality and privacy. They respond by releasing data either in a tightly controlled setting or in a securely anonymised form to prevent identification. These two routes have been chosen for two separate products from the 1991 British Census of Population: the Samples of Anonymised Records and the linked census Longitudinal Study. In this paper the implications of the two policies are compared. In broad terms, ease of access in the former route is traded for fullness of information in the second. In the final sections of the paper, the risks of identification disclosure in both datasets are assessed.