WHAT IS THE CENSUS DATA ENHANCEMENT PROJECT ALL ABOUT?
The ABS is undertaking a new project called Census Data Enhancement. The project, involving a 5% sample of the population (approximately 1,000,000 people) is an exciting new way of looking at Census data. Census Data Enhancement will increase the amount of relevant information available about Australian society to support decision making.
The project contains three parts:
- The central feature of the project is the creation of a Statistical Longitudinal Census Dataset (SLCD). The SLCD will be based on a 5% sample of the population. Records for this sample group will be brought together from each Census by statistical techniques which do not involve the use of name and address.
- Using the same statistical techniques, the dataset will be used with other non-ABS datasets for approved statistical projects. The other datasets being considered are: birth and death register data, long-term immigration data, and national disease registers.
- During the period of Census processing, name and address information will be used to bring together Census data and other selected datasets for ABS quality studies. Once Census processing is completed, all name and address information held by the ABS will be destroyed.
Project findings are available from the following webpage: Findings from Census Data Enhancement - Quality Studies
WHY DID THE ABS CHANGE THE ORIGINAL PROPOSAL WHICH INVOLVED USING THE WHOLE POPULATION?
The ABS commissioned an independent Privacy Impact Assessment into its original Census Data Enhancement proposal. This confirmed that ABS procedures and the current legislative environment provided a safe basis for the proposal to proceed.
However, the assessment noted that there was a residual privacy risk to the proposal - that the information gained could prove so attractive to a future government that legislation might be changed to enable inappropriate access to it.
The report noted that this would run totally counter to the experience of the past 100 years. During that time governments had actually improved legislative protection of ABS data.
In spite of this reassuring history, a number of submissions indicated concerns about the comprehensive nature of the original proposal.
While the ABS is very confident about its privacy safeguards, it responded to these concerns. The central consideration for the ABS is maintaining public confidence in the Census. As a result the ABS decided to use a 5% sample. This is still a very useful sample as 5 % of the population is about 1,000,000 people.
WHY IS THE ABS PLANNING THIS NEW USE OF CENSUS DATA?
In its existing form the Census is a 'snapshot' about Australian people and households once every five years. While it provides very good information on how many people are in a particular situation at a point in time, it doesn't explain the underlying causes.
This new way of looking at the data, across Censuses, can help to tell us what led to that situation, and what might be the likely consequence of it on the rest of the person's life? What sort of intervention might have prevented adverse situations, or encouraged positive ones? What benefits for the future might be achieved by early investment? The ability to combine personal characteristics with area characteristics would also be useful in many studies.
The potential benefits are substantial. They provide information on patterns in individual experiences over time and therefore provide insight into the effectiveness of policy or the need for new policy interventions. Examples of studies that could be undertaken are: pathways undertaken by migrants in their early years of settlement, particularly in employment; links between employment outcomes and education qualifications; transitions to higher education and work for young people from low income households; and the extent of income and employment mobility.
The ABS views the Longitudinal Census Dataset as an investment in the future, with the value of the dataset growing over time as more data is added.
WHAT WILL BE THE BENEFITS OF COMBINING DATA OVER TIME?
In its existing form the Census provides a 'snapshot' about Australian people and households once every five years. The creation of the Statistical Longitudinal Census Dataset (SLCD), by combining Census data for a 5% sample of the population, would provide the means to identify patterns of change in social and economic circumstances, for individuals and households, over time.
Examples of studies that could be undertaken are: pathways undertaken by migrants in their early years of settlement, particularly in employment; links between employment outcomes and education qualifications; transitions to higher education and work for young people from low income households; and the extent of income and employment mobility.
While the real value of the SLCD would not become apparent for some years, as data from the 2006 Census are brought together with data from 2011, 2016 and so on, the ABS sees this as an important investment in our future to create a significant national statistical asset.
DOESN'T THE ABS COMBINE CENSUSES AND OTHER INFORMATION ALREADY?
No, currently the ABS does not combine information in this manner.
WHEN WILL THE BENEFITS OF THE PROJECT START TO SHOW?
It will take several years before the full benefits of the project start to show. Longitudinal Census data, combining information from successive censuses, will only become available in 2012 after the 2011 Census is processed. An analysis of the quality of the dataset created will be undertaken by the ABS around this time.
The ABS views the Longitudinal Census Dataset as an investment in the future, with the value of the dataset growing over time as more data is added.
WHO WILL BE INCLUDED IN THE 5% SAMPLE?
Anyone in Australia, counted in the Census, could potentially be included in the 5% sample (approximately 1,000,000 people) that will be used to form the Statistical Longitudinal Census Dataset (SLCD).
The SLCD sample will be chosen using random sampling techniques. It will produce a file that includes people from all walks of life across Australia.
As in past Censuses, ABS will destroy any records containing names and addresses once Census processing has been completed. Because names and addresses will not be used, it will be impossible to identify who is included in the SLCD.
WILL THE ABS BE KEEPING MY NAME AND ADDRESS FROM MY CENSUS FORM?
No. The project does not involve the ABS keeping names and addresses from the Census.
The same watertight conditions of privacy and confidentiality of information will continue to exist as they have for the past 100 years. As in past Censuses, the ABS will destroy any records containing names and addresses once Census processing is completed.
WILL ANY CENSUS FORMS BE PRESERVED?
For the 2001 Census, all people completing the Census form were given the option of having their name identified Census information archived for the research use of future generations. It was called the 'Census Time Capsule' project. Those who took this option agreed to pass their 2001 information to the custody of National Archives of Australia. This information is not accessible to anyone, including the ABS for 99 years.
This option will again be available for the 2006 Census and again the information will be kept secret for 99 years.
This option is in no way connected to the Census Data Enhancement Project.
WHAT TOPICS WILL THE CENSUS ENHANCEMENT PROJECT COVER?
All data items as collected in the Census will be available for statistical purposes, subject to confidentiality constraints.
The detail of information available to users (for example geographical or occupational detail) will be constrained to ensure no data is released which is likely to enable the identification of an individual.
The SLCD will also be available to be used for approved statistical projects, with other specific data sources, where these are available for this purpose within the ABS. The data sources considered for this use are:
- birth and death register data (including cause of death)
- long-term immigration data
- national disease registers.
WHAT DATA WILL NOT BE AVAILABLE FROM THE SLCD?
Information that is likely to enable the identification of an individual will never be released. As with any other ABS data the same watertight conditions of privacy and confidentiality of information will continue to exist as they have for the past 100 years.
PRIVACY AND CONFIDENTIALITY
Will I know if I am in the sample?
No. Anyone in Australia, counted in the Census, could potentially be included in the 5% sample (approximately 1,000,000 people) that will be used to form the Statistical Longitudinal Census Dataset (SLCD). Names and addresses will not be used, and the individuals included in the sample will not be identified.
As in past Censuses, ABS will destroy any records containing names and addresses once Census processing has been completed.
How can I be sure that the confidentiality of my information and privacy will be guaranteed?
The ABS is obligated to comply with the provisions of the Census and Statistics Act (1905) and the Privacy ACT 1988 to respect the privacy of individuals and to protect the confidentiality of their data. The provisions under both these Acts would govern the use and release of data under this proposal.
Under the Census and Statistics Act, under which the Census is collected, it is a criminal offence to reveal names or addresses of Census respondents. All ABS officers are legally bound to secrecy under the Census and Statistics Act.
Can anyone release the data?
All ABS officers are legally bound to secrecy under the Census and Statistics Act. Any ABS officer found to be releasing identifiable data will be criminally prosecuted.
Who can use the dataset?
Those people wanting to access the SLCD, or to use the SLCD with another data source, would need to apply to the ABS, specifying the intended statistical use of the project. The project would be assessed by the ABS to ensure:
- the use was statistical.
- the use was likely to be of community benefit.
- the data proposed to be used was appropriate to address the research question.
- the use would not require access to identifiable information.
No identifiable information will ever be released outside the ABS.
Is my information going to be provided to other government agencies, such as Centrelink, Tax Office, Department of Housing?
No! The ABS is legally bound to protect the privacy of individuals and cannot give identifiable information to any third party.
Organisations outside the ABS will not be given data about an individual person or individual household. The confidentiality of your data is protected by the Census and Statistics Act and the Privacy Act. Both these Acts ensure that data is not provided to anyone where that data can be used to identify an individual. All ABS staff, including temporary employees are legally bound never to release personal information outside the ABS.
By law, organisations such as the Tax Office and credit reference groups cannot have access to personal details from the Census. |
