Each stage of the Census is subject to stringent quality assurance procedures which result in data of high quality. However, in a Census there are recognised sources of error and some of these may survive in the data produced. Potential sources of error in the Census are: undercounting, respondent error, processing error and introduced random error. Introduced random error is used to protect the confidentiality of individuals. The effect of such errors on overall Census results is generally insignificant and does not impair the usefulness of Census data.
A series of working papers will be produced to assess and report on various aspects of 2006 Census data quality. Census topics will be examined in detail and the impact of form design, collection procedures and data processing on data quality will be evaluated. Some topics to be evaluated include labour force status, Census undercount, and housing.
In addition, the ABS publishes Census Fact Sheets which assist users to understand and interpret Census data. They will be published in response to issues that arise during and after the publication of Census data. Some issues which may be covered include Income Imputation, Confidentiality, and Changes to variables between Censuses.
For a more detailed explanation on how the ABS ensures data quality in the Census, see the chapter titled
Managing Census Quality in this dictionary.
See also Derivations and imputations, Data processing, Introduced random error, Undercounting and/or underenumeration.