MADIP - FAQs
 

ABOUT MADIP

HOW TO ACCESS INFORMATION AND DATA

PRIVACY AND SECURITY

TECHNICAL



ABOUT MADIP

What are the benefits of MADIP?

The Multi-Agency Data Integration Project (MADIP) shows how combining existing public data can be used to:
    • Inform policy development and better targeting of services, such as health and early childhood services, to people and communities who need them
    • Provide insight into the effectiveness of government programs to ensure they are delivering value to the Australian public

A series of MADIP Case Studies have been developed to highlight the kind of insights that MADIP data can provide. The MADIP Research Projects page provides information about the approved research projects that use MADIP data.

What is the governance for the project?

There are currently three key levels of governance for MADIP:
    • The Secretaries Data Group and Deputy Secretaries Data Group have oversight of data integration involving Commonwealth data for statistical and research purposes.
    • MADIP forms part of the Public Sector Data Management agenda and is a core component of the Data Integration Partnership for Australia (DIPA) which is driven by the DIPA Board, reporting to the Secretaries Data Group.
    • The MADIP Board is composed of senior executive staff from the MADIP partner agencies and guides the strategic direction of MADIP and has oversight of data sharing, linking, and use in MADIP, and related issues.
    • The ABS, as the accredited Integrating Authority for the project, is responsible for linking the data, providing access to authorised users via highly secure ABS systems, and managing data in accordance with the Commonwealth Arrangements for Data Integration and terms agreed by the ABS and data custodians.

How is the project developing?

From June 2015–2018, the MADIP partners evaluated the technical feasibility of linking multiple datasets longitudinally, as well as the usefulness of the integrated data as a resource for policy analysis, research, and statistical purposes.

MADIP became fully operational on 1 July 2018 as a core asset of the Data Integration Partnership of Australia (DIPA), and will continue to expand to include new collections of data (from a range of sectors and sources, including surveys) as well as longitudinal information to enable insights into trends over time.

Is MADIP related to My Health Record?

No. My Health Record is a separate project undertaken by the Australian Digital Health Agency. Data from My Health Record is not linked with MADIP data.



HOW TO ACCESS INFORMATION AND DATA

How can I apply for access to MADIP data?

MADIP data products (for research) are listed in the 1700.0 catalogue series.

Access to MADIP data is only provided to authorised users for approved policy analysis, research, and statistical purposes. Government officers and non-government researchers authorised by the ABS and relevant data custodians have access to data for approved projects. All users are legally obliged to use data responsibly for approved purposes, comply with the conditions of access, and maintain the confidentiality of data.

All researchers and projects must be approved for access under the Five Safes Framework - see the MADIP Research Projects page for more information about the approval process.

To apply for access to MADIP data or to submit a query, email dipa@abs.gov.au.

Can I access or correct my personal information?

Under the Privacy Act 1988 agencies that collect your personal information may permit you to access or correct it, where it is reasonable and practicable for them to do so. However, data collected under the Census and Statistics Act 1905 is subject to legal exemptions. It is also important to note that personal information, such as name and address, is removed when combined with other datasets as part of the data integration process in line with the Separation Principle. This makes it unlikely that the ABS would be able to locate your information in MADIP to update or correct it.

Any enquiries about accessing or correcting personal information in MADIP should be directed to the ABS Privacy Officer at privacy@abs.gov.au or by phone on 02 6252 7203. Mail can be directed to:

Privacy Officer
Policy and Legislation Section
Australian Bureau of Statistics
Locked Bag 10
Belconnen ACT 2617

Any enquiries about accessing or correcting personal information outside of MADIP should be directed to the privacy officer of the relevant agency.

Who can I contact for more information about the project?

For more information about MADIP, email dipa@abs.gov.au.



PRIVACY AND SECURITY

How is my privacy protected?

The privacy and security of personal information in MADIP are maintained through strong legislative protections and best practice data management. Key measures include:
    • MADIP data can only be accessed by authorised users via highly secure systems. Storage and access are strictly controlled using the Separation Principle and Five Safes Framework. This ensures that personal information is separated from other information during storage and linkage (and cannot be accessed at the same time).
    • Only data that are reasonably necessary for an approved purpose are shared for linkage or made available for research or statistical uses. This is in line with the principle of data minimisation.
    • Personal information (e.g. name and address) are not available in files used for analysis, which are designed and only made available in a way that ensures an individual is not likely to be identified.
    • MADIP complies with the Privacy Act 1988 and other relevant legislation. Authorised researchers that breach their obligations under the Census and Statistics Act 1905 can face fines of up to $25,200, and/or up to two years in prison.
    • The privacy impacts of the project were identified and addressed as part of the MADIP independent Privacy Impact Assessment (PIA) consultation processes.

Privacy protections are reviewed on an ongoing basis and will evolve as new technology becomes available and new data is added.

For detailed information on how privacy is protected, see the MADIP Privacy Policy or the Privacy, Secrecy, and Information Security page.

How is my data kept secure?

The ABS is committed to data security. Visit the Privacy, Secrecy, and Information Security page to learn more about the safeguards the ABS has in place to your keep your information secure.

Can I opt out of MADIP?

No, it is not possible to opt out of MADIP. Names and addresses are removed during the data integration process, and without this information the ABS is unable to identify an individual’s personal information to remove their records.

It may be possible to opt out of collections which feed into MADIP, however, some collections feeding into MADIP are compulsory (such as the Census) while others are needed to deliver particular services or programs to you (such as welfare payments) or government functions (such as taxation).

Why was I not asked to consent to my personal information being used in MADIP?

Under the Privacy Act 1988, personal information may be used for the purpose(s) it was collected for and related purposes where the person consents or as authorised by law. MADIP uses data collected for research and statistical purposes, and is authorised by legislation (for more information on the legal basis of MADIP, see the MADIP data and legislation page).

Direct consent is not required or sought, however the ABS does seek to inform you about uses of your personal information by providing information about the project online, in the MADIP Privacy Policy, and in informative materials for particular collections which feed into MADIP. Public interest is also taken into account during the data sharing and research project approval processes.

How do I make a privacy complaint?

If you think the ABS may have breached your privacy rights or its privacy responsibilities in relation to MADIP, a complaint should first be made to the ABS Privacy Officer using the information provided above.

If you are not satisfied with how the ABS Privacy Officer handles your complaint, or the outcome reached, you may refer your complaint to the Office of the Australian Information Commissioner.



TECHNICAL

How is MADIP data linked?

Data linkage is conducted by a dedicated team within a secure ABS environment.

Underpinning MADIP is a ‘spine’, created through a three-way linkage between the Medicare Enrolments Database (MEDB), Social Security and Related Information (SSRI), and Personal Income Tax (PIT). Together, these datasets have very high coverage of the Australian population. The high coverage of the spine enables high quality linkage of other datasets to the spine.

Linkage is performed using identifying information from the datasets. For most MADIP linkages this is anonymised name, date of birth, geocoded address, and sex or gender. Names are anonymised using a character replacement method to prevent the name information from being recognised. Addresses are geocoded to Address Register Identifiers (ARIDs), which are unique identifiers representing individual physical Australian addresses. Broader statistical geographies for addresses (e.g. Mesh Block) are also used in linkage.

There are some variations to the linkage variables used for particular datasets. In particular, note that MADIP does not use original names collected in the Census of Population and Housing, as this information has been destroyed (Census 2011) or irreversibly encoded (Census 2016) prior to linkage in MADIP. For more information on the management of Census data, see Census Privacy, Confidentiality and Security or the Census Privacy Policy.

At all points in the data linkage process, identifiable personal information is stored separately from other (analytical) information, and cannot be seen at the same time – in accordance with the Separation Principle. Additionally, analytical variables from each dataset are stored separately until they are brought together in a linked analytical dataset. These analytical datasets are created for specific purposes and include only the subset of MADIP analytical data required for that purpose. Note also that the MADIP “spine” is simply a ‘map’ or concordance of identifiers from one dataset to another, and does not contain any linkage or analytical data.

For details about the datasets involved, see the MADIP data and legislation page.

How is my personal information (e.g. name and address) used?

Experience shows that having name and address information is crucial to achieving high quality linkage between administrative datasets and other collections, and thus creating high quality integrated datasets.

Personal information used in linkage (either in original form, or changed into an unrecognisable form to protect privacy) includes name, address, date of birth, and government identifiers. In particular, names are anonymised or encoded prior to linkage in MADIP. Other demographic information which does not directly identify a person (such as country of birth) may also be used to link datasets together where necessary to ensure high quality linked data.

The linked data available for analysis does not contain names and addresses.

Note relating to Census name information

MADIP does not use original names collected in the Census of Population and Housing, as this information has been destroyed (Census 2011) or irreversibly encoded (Census 2016) prior to linkage in MADIP . For more information on the management of Census data, see Census Privacy, Confidentiality and Security or the Census Privacy Policy.

For more information on how we use your personal information in MADIP, see the MADIP Privacy Policy.

How long will MADIP data be retained?

Data supplied for and integrated as part of MADIP is retained by the ABS while there is a business need to do so, for instance to maintain and expand the integrated data. The need for retention is reviewed annually, or more frequently if appropriate. This is consistent with the Privacy Act 1988.