1 BRINGING TOGETHER 2011 CENSUS DATA WITH A SMALL NUMBER OF PREDETERMINED DATASETS DURING CENSUS PROCESSING USING NAME AND ADDRESS, FOR QUALITY STUDIES
The ABS is intending to undertake a small number of predetermined data integration projects associated with the 2011 Census of Population and Housing which will involve linking the Census to ABS and non-ABS datasets using name and address, for quality studies.
Each of these studies has been vetted to ensure the studies demonstrate significant potential community benefit; are statistically appropriate; cannot be done without name and/or address to bring the data together; and do not require identifiable information to be accessible outside the ABS. After careful consideration, the Australian Statistician has determined that each of these studies should proceed.
The key features of these studies are:
What has changed since the 2006 Census
- once the purpose of each study has been met, all linked datasets will be destroyed;
- the studies can only be undertaken during the Census processing period when name and address are available;
- the linked datasets created through these projects will not leave the ABS and will only be accessible by those ABS officers directly involved in the study; and
- the ABS will not retain Census names and addresses once Census processing is completed. The only exception is if a person explicitly agrees by answering the relevant question on the Census form to have their name-identified responses retained by the National Archives of Australia for release in 99 years time (see Glossary for further detail);
The 2006 CDE project made use of name and address information from the 2006 Census during the Census processing period to undertake a number of identified quality studies (see Appendix 1
). These quality studies have demonstrated the potential for significant statistical improvements to be made by bringing together data from the Census with other datasets. As noted in Appendix 1
, this included immigration and mortality data. For the 2011 Census, the datasets proposed to be temporarily brought together with the Census for quality studies are listed below.
There will be no change to the methodologies used in 2006 to bring together datasets in 2011. Names and addresses collected on the 2011 Census will be destroyed at the end of Census processing.
The purpose of each quality study to be undertaken during the 2011 Census processing period are to:
- assess the quality of and provide a benchmark for, data integration projects that do not make use of name and address; and
- assess the data integration methodology to help in improving processes for future data integration projects.
Four quality studies will be undertaken for the 2011 Census. These are:
Benefits of the Quality Studies
- Census 2011 Dress Rehearsal to 2011 Census data to provide a benchmark standard for the 5% SLCD;
- 2011 Census to Department of Immigration and Citizenship's Settlements Database to compare linkage outcomes and quality with earlier work linking the 2006 Census to the Settlements Database, enabling evaluation of subsequent improvements implemented by DIAC to the Settlements database and to evaluate opportunities for further development. The study will also provide a benchmark for projects which bring together the Settlements Database with both the 5% SLCD and the 2011 Census data without using name and address;
- studies related to an overall strategy to develop an Australian Longitudinal Learning Database: a) 2011 Census to school and early childhood education student enrolment data; b) 2011 Census to a sample of student data from the National Assessment Program; and c) 2011 Census to the Australian Early Development Index; and
- 2011 Census to a Western Australian Enhanced Mortality dataset to provide input into development of national best practice guidelines for data linkage related to Indigenous people.
Undertaking these studies provides increased quality assurance for proposed as well as future data integration projects in terms of both the methodologies applied and the quality of the output from the projects.
Data involved in the Quality Studies
These quality studies will temporarily bring together the 2011 Census data with a range of ABS and non-ABS datasets using name and address during Census processing. Names and addresses collected on the Census will be destroyed at the end of Census processing and the linked datasets created for the quality studies will be deleted once the purpose of the project has been fulfilled.