§ Mr. CohenTo ask the Secretary of State for Health what plans he has to amend procedures relating to the disclosure of information from the next census to take into account access to (i) the electoral register, (ii) the extract of the poll tax register and (iii) anonymous poll tax data; whether access to any of these could undermine the confidentiality of the census; and whether he will make a statement.
§ Mr. FreemanAs in past censuses, no information on identified individual people or households will be passed to anyone other than employees of the registrars general. Thus current procedures already prevent such information being passed to electoral registration officers or community charge registration officers. Furthermore, stringent measures already exist to guard against possible indirect disclosure of information on individual people or households through the published statistics and no incidents of such disclosure have been reported from the 1981 census. It is planned to strengthen these measures further for the 1991 census by doubling the minimum thresholds, below which no statistics, other than a simple total of people and households will be released for any area, to 16 households and 50 people.
§ Mr. CohenTo ask the Secretary of State for Health what plans he has to reassure individuals who complete the census that they will not be identified from the information subsequently disclosed by his Department; and whether he will make a statement.
§ Mr. FreemanReassurance on confidentiality will be given on the front page of the census forms in the message addressed to heads of households and others responsible for completing the forms. It will also be reinforced in publicity material issued before the census.
§ Mr. CohenTo ask the Secretary of State for Health whether he will describe the uses of census data that specifically require the collection of the date of birth rather than the age of a person; and whether he will make a statement.
§ Mr. FreemanThe question on date of birth is asked primarily because it has been found to give more accurate statistics than can be obtained by a question simply on age. As the census is taken in April the information on date of birth is used to produce the mid-year population estimates by age as at 30 June. Also, four dates of birth have been used to select the OPCS longitudinal study sample from the 1971 and 1981 censuses. OPCS plans to include in this study the appropriate sample of records from the 1991 census.