Identifying data for release

What data does my agency collect and manage?

The public sector creates, collects and manages a range of data in the course of its everyday operations, including through administration, delivering and monitoring programs and services, and research. Agencies should review their data assets, considering:

  • Existing data (current and historical data): Some of this data may already be made available to the public, for example on the agency’s website, in annual reports and publications, and in databases.

  • Future data collection and creation (new data): This may include upcoming or future programs, projects, services, reports and other initiatives. Agencies should consider and apply the Policy to the development of systems that collect or create data, and any modernisation projects that update existing data systems.

  • Data collected by the agency’s contractors, consultants and grant recipients: Agencies should consider including provisions surrounding the collection of data, consistent with the principles of the Policy, in their contracts and agreements with contractors, consultants and grant recipients.

What data should I consider for release?

Agencies are encouraged to adopt a position of “open data by default”. That is, considering all data for release to the public. However, there are instances where data is not appropriate or suitable for release and, thus, access should be restricted or precluded (outlined in Step 1.3). From an implementation perspective, the focus is high-value data. That is, where there is some value, outcome or benefit for the State from opening access to that data. This includes:

  • Potential value and outcomes for government – including reduced silos and improved collaboration, better decision-making, new and improved solutions and services, and efficiencies and savings (including reduced duplication and tools and service delivery more quickly and at lower costs); and

  • Potential value and outcomes for industry and the community – including supporting the development of new industries and business (and jobs), increased entrepreneurship and productivity, improved research outcomes, and better business and community decision making.

This assessment will help agencies evaluate against the cost and potential implications of making that data open, decide how best to manage the data (including whether to release it and under what conditions), and determine priorities for release.

In the first instance, agencies should improve the discoverability and usability of existing datasets, prioritising those that:

  • are already publically available, for example on the agency website or in public reports;

  • are considered high-value (as outlined above); and/or

  • are in high demand by the public – front-line helpdesk requests, Freedom of Information (FOI) logs, and feedback and engagement with the public are useful ways of gauging this.

Other resources:

Australian Government – Open Data Toolkit – Assessing what datasets agencies should publish to improve productivity

Global Open Data Index – Dataset overview

New Zealand Government – Guidance and Resources – Open Data Case Studies

Is the data appropriate and suitable for release?

A large amount of data collected by the public sector will be suitable for public release. However, agencies should consider the potential implications and risks of releasing data, weighing them up against the potential value or benefits of openness. There are instances where agencies need to modify the data prior to release, or restrict or preclude access, including:

  • Privacy – where personal or sensitive information is involved that can be identified for an individual, or may be involved as an unintended result of data linking. Data that is released must not be (or be able to be) associated with any individual. Before releasing data, agencies should carefully consider and address any privacy concerns. Agencies may be able to mitigate privacy risks by de-identifying the data, which means removing anything that can identify a person. However, caution needs to be exercised where disparate datasets, individually de-identified could be linked to reidentify an individual. The Government of Queensland’s guidance on Dataset publication and privacy and Dataset publication and de-identification techniques is useful.

  • Security – because of the nature of the data or information. For example, data that is associated with national security or State interests vital to stability and integrity.

  • Confidentiality – arising because of the nature of the data or information itself or because a contractual arrangement has been made in relation to the data or information. For example, data that reveals Western Australian Cabinet information, is subject to medical practitionerpatient privilege, or is commercial-in-confidence.

  • Legal privilege – attached to certain legal advice.

  • Commercial – such as commercial-in-confidence, patent pending or intellectual property considerations.

  • Public interest – if there are public interest considerations against disclosure and, on balance, those considerations outweigh the public interest considerations in favour of disclosure.

Agencies should also consider their own specific legislative provisions relating to the release of data and information.

Refer to Appendix A of the Western Australian Whole of Government Open Data Policy for the definitions of terms used throughout (indicated in italics).