ABSTRACT
There is some ambiguity about the usage of 'census'. Kendall and Buckland's {1960) A Dictionary of Statistical Terms says {p. 37) of 'census' that 'The word is used to denote a complete enumeration as distinct from the partial enumeration associated with a sample'. The alternative distinction between a census and a survey is taken to lie in the nature of the information collected. The former term is then generally confined to enquiries that are more or less straightforward counts, like censuses of population, distribution and production, the term 'survey' being applied to enquiries which go beyond simple enumerations. With this terminology, it is perfectly sensible to speak of sample censuses: for example the sample enquiry into fertility conducted in 1946 by the Royal Commission on Population was called the Family Census (Glass and Grebenik, 1954) and the 1966 Population Census of Great Britain was taken on a ten per cent sample basis. We shall broadly follow this second distinction here. It is by no means a hard-and-fast one, but, fortunately, it need not detain us, since this book is mainly concerned with the kinds of investigations which one instinctively calls surveys rat~er than censuses.
