1001.0 - Australian Bureau of Statistics -- Annual Report, 2016-17  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 19/10/2017   
   Page tools: Print Print Page

PERFORMANCE REPORT - ANNUAL PERFORMANCE STATEMENT

INTRODUCTORY STATEMENT

As the Accountable Authority of the Australian Bureau of Statistics (ABS), I present the 2016-17 annual performance statement of the ABS, as required under paragraph 39(1)(a) of the Public Governance, Performance and Accountability Act 2013 (PGPA Act). In my opinion, based on advice from ABS management and the Audit Committee, this performance statement accurately reflects the performance of the ABS against the performance criteria included in its Portfolio Budget Statement and Corporate Plan, and complies with subsection 39(2) of the PGPA Act (endnote 1).

David W. Kalisch
Australian Statistician


ABS PURPOSE

The ABS Purpose is to provide trusted official statistics on a wide range of economic, social, population and environmental matters of importance to Australia (endnote 2).


RESULTS

Results against the criteria from the ABS Corporate Plan 2016-17 to 2019-20 and the ABS Portfolio Budget Statement (PBS) are described below. All results related to one programme in the ABS PBS: 1.1 Australian Bureau of Statistics.


Performance Criterion 1
Criterion source

ABS statistical solutions will continue to inform decision makers, researchers and discussion by governments, business and the community.

The ABS produces the statistics required by users and the statistics will be of sufficient quality to be fit-for-purpose for users. The ABS will engage with partners through appropriate consultation forums to understand user requirements that inform strategic directions and the ABS work program.

The ABS will better manage risks to key statistics to maintain appropriate quality.
ABS Corporate Plan & Portfolio Budget Statement
Program 1.1, 2016-17 Portfolio Budget Statements, p.56.

Target 1.1: ABS continues to produce key economic and population statistics with appropriate coverage, frequency and timeliness as assessed by the International Monetary Fund against the Special Data Dissemination Standard; except where the ABS has made an explicit decision not to do so, following consultation with stakeholders, based on Australian needs and circumstances.

Why this target is important: To ensure comparability between countries, the standards, frameworks, classifications and methodologies governing official statistics are agreed internationally. The ABS publishes the concepts, sources and methods it uses for all key macroeconomic and demographic statistics. The ABS releases publications that include information on the applications of internationally agreed practices in ABS statistics.

Most years, the International Monetary Fund (IMF) assesses Australia’s observance of its Special Data Dissemination Standards. These standards provide an accepted benchmark for the main economic and population indicators for subscribing countries.

Results: Target met.

The most recent IMF Country Report on Australia was published in February 2017 and it noted that:
    “Data provision is adequate for surveillance (endnote 3). Australia has subscribed to the Special Data Dissemination Standard (SDDS) since April 1996, and its metadata are posted on the Fund’s Dissemination Standards Bulletin Board (DSBB).” (IMF Country Report No. 17/42)

    Target 1.2: At least 96% of the media articles are not critical of the ABS or ABS statistics.

    Why this target is important: Media reporting of concerns about the quality of ABS statistics or about the reputation of the ABS as a trusted organisation may indicate distrust in ABS official statistics, diminish reliance on them for decision making, and reduce voluntary participation in ABS surveys.

    Results: Target not met.

    81% of print articles referred to the ABS without concerns about ABS statistical quality or general criticism of ABS as an agency (endnote 4). The result was less than the target mainly due to negative coverage in 2016 surrounding the 2016 Census of Population and Housing (Census) operations.

    Target 1.3: The following consultation fora meetings will be held:
    • three of the Australian Statistics Advisory Council;
    • at least four of the State Government Statistical Priorities Forum;
    • two of the Australian Government Statistical Forum;
    • at least two Economic Statistics Advisory Group meetings; and
    • two Population and Social Statistics Advisory Group meetings.

    Why this target is important: It is important to engage with key user groups to understand their information needs and be sensitive and responsive to stakeholder needs and preferences. Stakeholders want to partner with the ABS and be connected with opportunities across the statistical system.

    Results: Target met.

    The ABS held the majority of the planned strategic engagement fora meetings during 2016-17, including: two of the Australian Statistics Advisory Council; three of the State Government Statistical Priority Forum (now known as the State Statistical Forum); three of the Australian Government Statistical Forum; two of the Economic Statistics Advisory Group; and two of the Population and Social Statistics Advisory Group.

    The planned June 2017 Australian Statistics Advisory Council (ASAC) meeting was moved to August 2017 due to member availability, thereby reducing the number of meetings to two instead of the forecast three in 2016-17 (endnote 5). The State Statistical Forum (formerly known as the State Government Statistical Priorities Forum) held three meetings in 2016-17, which was the expected number since the original target was set incorrectly. The Australian Government Statistical Forum held one more meeting than was expected, at the request of members.

    Target 1.4: Risk Management Plans are developed and approved for key economic and industry statistics.

    Why this target is important: This target enables effective management of statistical risk across the ABS in a time of significant change.

    Results: Target met.

    The Labour Force, the National Accounts, Consumer Price Index and Estimated Resident Population, as well as key inputs to these statistics, have implemented risk management plans that have been quality reviewed by the ABS Statistical Quality Methodology Branch.


    Performance Criterion 2
    Criterion source

    The ABS will partner and collaborate with stakeholders to develop new statistical solutions that inform decisions on important matters. ABS stakeholders will provide feedback on the effectiveness of these collaborations, their satisfaction with the ABS responsiveness in meeting their needs, and their confidence in ABS statistics.
    ABS Corporate Plan & Portfolio Budget Statement
    Program 1.1, 2016-17 Portfolio Budget Statements, p.57.

    Target 2.1: Partners report increased levels of satisfaction with ABS responsiveness in meeting their needs in the biennial Stakeholder Relationship Health Assessment compared to when the APSC Capability Review of the ABS was undertaken.

    Why this is important: It is important to engage with key user groups to understand their information needs and be sensitive and responsive to stakeholder needs and preferences. Stakeholders want to partner with the ABS and be connected with opportunities across the statistical system.

    Results: Not due to report until 2017-18.

    The External Stakeholder Relationship Health Assessment is undertaken every two years. It was last completed in 2015-16 and is due to be repeated in 2017-18.

    Target 2.2: Case studies are produced demonstrating how the ABS collaborates with partners to develop statistical solutions that have or will significantly inform important decision making by government.

    Why this is important: Informing important decision making by government is fundamental to achieving the outcome of the ABS. Case studies are one way for the ABS to demonstrate it is performing and achieving its outcome.

    Results: Target met.

    The case studies at the end of this section demonstrate how the ABS collaborates with partners:
    • Case Study 1: Using BLADE to provide new insights into the benefits of business research and development - a collaboration with the Department of Industry, Innovation and Science and Swinburne University of Technology
    • Case Study 2: The Multi-Agency Data Integration Project (MADIP)
    • Case Study 3: Experimental Land Accounts - a collaboration with Geoscience Australia.

    Target 2.3: A baseline is established on the use of ABS data integration products.

    Why this is important: Producing microdata and integrating survey and administrative datasets significantly enhances the potential of public sector data to drive innovation, efficiency, productivity and economic growth. Data access systems (such as virtual data laboratories) continue to be enhanced to allow trusted users to access microdata while protecting the identities of individuals and organisations.

    Results: Target met.

    In response to a changing environment, the ABS has been seeking to widen the availability of its microdata. This has introduced a whole new way of working with microdata, which has met user needs and resulted in a high demand for the new services that are being trialled. The baseline was established and achieved in 2015-16 due to the high take-up of the new ABS DataLab. In 2016-17 the number of registered integrated data users increased from 68 to 209, an increase of 307%. Trials of the ABS DataLab continue, allowing the ABS to refine the provision of secure access to selected researchers.

Performance Criterion 3
Criterion source

The ABS will reduce the burden placed on providers. Provider take-up of electronic reporting will be enhanced through a Census electronic form usable on many mobile devices.
ABS Corporate Plan & Portfolio Budget Statement
Program 1.1, 2016-17 Portfolio Budget Statements, p.57.

Target 3.1: The ABS delivers annualised reductions in red tape of at least $200,000.

Why this is important: The principal objective of the ABS in respect of providers is to impose the lowest load possible while meeting its obligations to provide Government and the community with a high quality official statistical service.

Results: Target not met for 2016-17 financial year.

The ABS has reported on regulatory burden since January 2014 with targets established in 2016-17. While the target was not met for 2016-17, the table below shows that since reporting commenced the ABS has reduced regulatory burden by an average of $2,306,435 each year.

In consultation with key stakeholders, and in line with the ABS Forward Work Program, decisions were made to increase content in the Survey of International Investment to better service government information needs. This led to an overall increase in regulatory burden in 2016-17 of $285,877.



Financial Year
Change in Burden

2013-14
-$458,000
2014-15
-$6,228,000
2015-16
-$1,672,400
2016-17
$285,877

Net decrease in burden over the reporting years
-$8,072,523




Target 3.2: Estimated total provider burden on businesses remains steady at approximately 400,000 hours.

Why this is important: The principal objective of the ABS in respect of providers is to impose the lowest load burden possible while meeting its obligations to provide Government and the community with a high quality official statistical service.

Results: Target not met.

461,262 hours (endnote 6, endnote 7).


FIGURE 4.1: TOTAL BUSINESS PROVIDER LOAD

Graph depicting total business provider load by size of business
Source: ABS Annual report, various years; ABS Collection Management System (CMS).
Note: Spikes in the data are observed in the following five yearly intervals, which correspond with the Agricultural Census: 2001-02; 2006-07; 2011-12; and 2016-17.


Target 3.3: Business take-up of electronic forms exceeds 90%.

Why this is important: The ongoing cooperation of providers is critical for the ABS to provide the high quality statistical information needed to foster informed debate and drive evidence based policy making. Contemporary data collection methods, such as the use of electronic forms, increase efficiency for providers, allowing for an enhanced provider experience and improved data quality outcomes.

Results: Target mostly met.

80% (rounded) business electronic form take-up (endnote 8).
Time series of total business take-up of electronic forms by reporting period since 2012-2013.


FIGURE 4.2: ABS BUSINESS SURVEY SELF-RESPONDING WEB TAKE-UP RATES
Graph depicting time series of total business take-up of electronic forms by reporting period since 2012-2013
Source: Aggregated data extracted from Provider Information Management System (PIMS), which is the Management Information system used for tracking response rates for business surveys.
Note: In 2012-13, the ABS moved to using a web form for most business surveys. The results for 2012-13 reflect only one quarter of web form take-up rates, as this transition occurred for the last quarter of 2012-13 reporting period.


FIGURE 4.3: BREAKDOWN OF ABS BUSINESS SURVEY SELF-RESPONDING WEB TAKE-UP RATES BY BUSINESS STATISTICAL AREA
Graph depicting breakdown of ABS business survey self-responding web take-up rates by business statistical area
Source: Aggregated data extracted from Provider Information Management System (PIMS), which is the Management Information system used for tracking response rates for business surveys.


Target 3.4: More than 65% of the population complete their Census using an electronic form.

Why this is important: ABS testing showed that on average households completed the electronic form faster than the paper form equivalent in a side by side comparison. Also, data reported on the electronic form was of better quality than on the paper form. This is a cost saving for the ABS and allows the Census data to be released more quickly. It also reduces the time burden on the household filling out the form. The ABS notes that following on from the events of Census night the electronic form was unavailable for nearly two days within an eight week collection period and many people experienced frustration and considerable burden in trying to access the form.

Results: Target mostly met.

63.3% of the population completed the Census using the electronic form (endnote 9). The result was less than the target, but a significant increase from the 37.9% of respondents that completed the Census online in 2011.

Performance Criterion 4
Criterion source

The ABS will collaborate with partners to improve statistical infrastructure, capabilities, people and culture. The ABS Statistical Business Transformation Program is on track to develop innovative new infrastructure and capabilities. The ABS will progress the People and Culture Action Plan including an initiative to develop the future- ready professional, analytical and conceptual skills of selected staff.
ABS Corporate Plan & Portfolio Budget Statement
Program 1.1, 2016-17 Portfolio Budget Statements, p.57.

Target 4.1: The new enterprise data management environment will be implemented allowing the ABS to manage its data assets more cost-effectively.

Why this is important: The Statistical Business Transformation Program is two years into its five year program. The $257m Government investment to modernise ABS infrastructure and business will reduce risks to statistical outputs; reduce costs; and achieve faster turnaround in dissemination of statistics.

Results: Target on track.

During 2016-17 the Enterprise Data Management Environment was delivered into a production environment. It is an integrated environment for storing and managing all types of statistical data. Statistical programs will incrementally adopt the new system over the next few years.

Target 4.2: Case studies are produced demonstrating the benefits of new and enhanced infrastructure and capabilities.

Why this is important: New and enhanced infrastructure and capabilities are essential tools the ABS needs to create new statistics that inform important decision making. Case studies are one way for the ABS to demonstrate that the new infrastructure and capabilities are making a difference.

Results: Target met.

The case studies at the end of this chapter demonstrate the benefits of new and enhanced infrastructure and capabilities:
  • Case Study 4: Making a difference - the ABS DataLab
  • Case Study 5: Working with Data61 to improve the NationalMap
  • Case Study 6: The Freight Performance Measurement Project - a collaboration with the Bureau of Infrastructure, Transport and Regional Economics.

Target 4.3: All actions from the ABS People and Culture Action Plan for 2016 & 2017 are completed including identifying and starting to address future capability requirements through operational group workforce plans (endnote 10).

Why this is important: The ABS People and Culture Action Plans detail the key actions required to deliver on the ABS Workforce Strategy 2015-19, which sets the strategic direction for the people and culture elements of transformation.

Results: Target not met.

Due to the need to divert resources to support ABS Transformation and 2016 Census operations, not all actions from the ABS People and Culture Action Plans for 2016 & 2017 were completed. Of the 31 identified actions in the 2015 & 2016 People and Culture Action Plan, nine were completed (29%) by November 2016. The remaining actions that were in progress at the time were updated and reprioritised as part of the 2017 & 2018 People and Culture Action Plan. Of these 36 new actions, three have already been completed (8%), with the remainder in progress. The ABS has committed greater resources to progress the identified actions in 2017-18 and the new program of Strategic Workforce Planning provides a focus for the achievement of the ABS’s ambitions in relation to people and culture.

Target 4.4: A training course to build future-ready professional, analytical and conceptual skills is developed and piloted.

Why this is important: Building the capacity and capability of ABS staff is critical to helping realise the benefits of transformation. The Statistical Business Transformation Program will provide new systems meaning changes to the type of work staff do and the thinking styles staff must deploy. Upskilling and future proofing our staff in these new systems and the core capabilities required of them will support the continued production of high quality statistics.

Results: Target met.

During 2016-17 the ABS developed a training package (the Analytical Thinking Foundational Pathway), which contained a series of modules to build staff analytical and conceptual capacity. It was launched as a pilot program on 31 October 2016 and then revised, based on feedback, and rolled out to all staff on 31 March 2017.




2016-17 ANALYSIS OF PERFORMANCE AGAINST PURPOSE

The ABS continues to fulfil its purpose of providing trusted official statistics on a wide range of economic, social, population and environmental matters of importance to Australia (Target 2.2). Fulfilling the purpose of the ABS means that governments, businesses and the community are better informed to discuss, research and make decisions, which enhances the wellbeing of all Australians.

The outage of the online form on Census night in August 2016 resulted in an increase in negative media sentiment. This is reflected in the below-target results for Target 1.2. Rebuilding public and stakeholder confidence and trust in the ABS, and applying the lessons learned in 2016 to planning for the 2021 Census, are major focus areas for the ABS. Strengthening partnerships is also a high priority for the ABS, and throughout 2016-17, the ABS engaged with stakeholders (Target 1.3) to understand their requirements. The next assessment of stakeholder satisfaction with the ABS is due in 2017-18.

A key priority of the ABS has been to continue to deliver high quality, timely statistics on important matters. The frequency, timeliness, coverage and relevance of key macroeconomic and population statistics has been maintained (Target 1.1). The management of statistical risk has improved: key ABS statistical collections now have Risk Management Plans (Target 1.4). New statistical releases created through data integration continue to increase in use (Target 2.3).

The ABS continues its efforts to reduce red tape and burden for providers (Target 3.1 and Target 3.2). Fewer business survey providers than expected reported using the electronic form in 2016-17. This is mainly due to the running of the five yearly Agricultural Census, as the agricultural sector has a lower propensity to complete surveys online (Target 3.3).

The ABS operates in a dynamic environment and is being challenged to deliver the best statistical program possible, given the resourcing allocated. The ABS is committed to major changes to better meet the requirements of stakeholders and is transforming across six dimensions of environment, strategy, governance, people, culture and infrastructure. The ABS is implementing new statistical infrastructure which is delivering benefits to users (Target 4.1 and 4.2). ABS people and culture have been enhanced through training courses developed and rolled out to staff (Target 4.3 and Target 4.4), as well as other initiatives.

Relevant and Complete

These performance measures revolve around ABS statistical solutions informing decision makers, researchers and discussion by governments, business and the community, which is relevant to its purpose. This performance information when read as a whole shows how the ABS purpose is being achieved.

Reliable

Overall, information sources for each measure are fit-for-purpose. Methodologies and processes have been documented. For some measures methodologies need to be strengthened or targets better defined to ensure the results are verifiable. The ABS has a rolling internal audit program which includes examining measures in the performance statement. In 2016-17, three performance measures were audited and the ABS has been improving its approach based on feedback gained from these audits.


CASE STUDY 1: USING BLADE TO PROVIDE NEW INSIGHTS INTO THE BENEFITS OF BUSINESS RESEARCH AND DEVELOPMENT - A COLLABORATION WITH THE DEPARTMENT OF INDUSTRY, INNOVATION AND SCIENCE AND SWINBURNE UNIVERSITY OF TECHNOLOGY

Australian businesses spent nearly $19 billion on research and development activities in 2013-14. The flow on impact of this investment to third parties, not undertaking research and development, is of considerable interest to policy agencies and researchers.

Flow on impact includes productivity improvements resulting from reuse of process innovations, reduced costs of products for consumers through adoption of innovations, and the adaption of research and development by other industries for new uses (for example, technology developed for radar being reused in microwave ovens). The value of this benefit is often referred to as business research and development spillovers.

In 2016, the ABS, the Department of Industry, Innovation and Science (DIIS) and Swinburne University of Technology undertook work to assess the value of business research and development spillovers to the economy. This project was undertaken using the ABS’s Business Longitudinal Analysis Data Environment (BLADE).

Analysis was undertaken using ABS Business Expenditure on Research and Development survey data, Research and Development Tax Incentive Program data from DIIS, and business income tax return and business activity statement data from the Australian Taxation Office. The data was combined and confidentialised by the ABS and made securely available through the BLADE.

The data in BLADE allowed longitudinal analysis using the population of businesses receiving research and development tax incentives in Australia. Analysis was undertaken by industry sector and business size, and included analysis of the beneficiaries of research and development expenditure. The research produced improved estimates of the value of research and development spillovers in Australia. DIIS plan to release a report based on the research in November 2017.

This has been pioneering work delivering new data models, which can be reused by policy agencies and researchers in future. Antonio Balaguer (DIIS) said “this collaborative project makes a great contribution to close the gap to improve evidence for innovation policy.” Professor Beth Webster (Swinburne University of Technology) also acknowledged that “access to a large number of firm-level observations via BLADE puts Australia on the frontier of business research and allows researchers to estimate, with greater precision and nuance, the main drivers of economic growth.”

The project also demonstrates a model for future partnerships between government departments and the university sector for research directly relevant to current policy initiatives, such as the current Research and Development Tax Incentive Program.

Future analysis of this type, using integrated confidentialised data in the BLADE, will allow for the use of existing data in policy design and program evaluation. It is one of a number of ways the ABS is engaging with partners to improve the value and use of integrated data.


CASE STUDY 2: THE MULTI-AGENCY DATA INTEGRATION PROJECT (MADIP)

The Multi-Agency Data Integration Project (MADIP) is a partnership between the Australian Bureau of Statistics, the Australian Taxation Office, the Department of Education and Training, the Department of Health, the Department of Human Services, and the Department of Social Services. The project is in an evaluation phase and has brought important national datasets together to explore how to make better use of existing public data for policy analysis, research, and statistical purposes.

The MADIP has securely linked existing Medicare, government payments, personal income tax, and 2011 Census data. The linked data has enabled analyses of socio-economic outcomes and trends to inform policy and program development. The case study below showcases the type of new policy insights which can be discovered by bringing together data assets from across the Government.
    The most effective government policies and programs are those informed by high quality evidence. High quality evidence is built on accurate and comprehensive data. The MADIP has demonstrated the value of data integration as a way of shedding light on the performance of government programs and services, helping to determine the extent to which they are delivering outcomes.”

    Dr Steven Kennedy
    Deputy Secretary, Innovation & Transformation, Department of the Prime Minister and Cabinet

Privacy and security

The privacy and confidentiality of personal information is maintained through strong legislative protections as well as best practice data management. MADIP data can only be accessed via secure systems, and access is restricted to approved users for approved purposes. The MADIP complies with the Privacy Act 1988 and with the legislative responsibilities of each party. MADIP data has the added protection of the Census and Statistics Act 1905, which requires that no data is released in a manner that is likely to enable an individual to be identified.

Looking to the future

Evaluation of the project to date has demonstrated the potential value of linking existing public data. Streamlining data sharing and access arrangements can enable greater efficiencies, and inform decision making within government and the community.

The evaluation of the MADIP has noted that additional data, particularly longitudinal data, would be extremely useful for policy analysis, research and statistical purposes. It has also noted that there would be a considerable benefit in broadening access to integrated public data in a manner that maintains the confidentiality of individuals.

The next steps are to complete the evaluation of the MADIP and build on the lessons from this feasibility phase by putting in place arrangements that make better use of existing public data, safely and securely. This will contribute to the development of enduring statistical data assets under the Data Integration Partnership for Australia initiative.

Case study: Investigating patterns of Medicare expenditure for Age Pension recipients

Government policy objective: To improve older Australians’ access to health care services by understanding the geographic patterns of Medicare use.

Mapping linked 2011 Census and Medicare data from MADIP has provided new insights into health care costs and revealed regional differences in health service use by older Australians.

Geospatial analysis of average Medicare expenditure by Age Pension recipients was undertaken to identify expenditure patterns by region (see Figures 1 and 2). The analysis showed that major population centres have high Medicare expenditure per capita, but also that regional and remote areas in NSW (such as the Mid North Coast and the Far West, respectively) have significantly higher expenditure per capita than such areas in the other States and Territories. These findings can help inform policy and support the allocation of resources to the people and places that need it most.
    “… one of the most exciting emerging capabilities is integrated datasets. They offer the greatest opportunity for Government to prioritise resources in high cost areas like health, welfare and education; better allocate the limited funding for important services; target interventions and programs to deliver better outcomes; and reduce duplicative collection and analysis costs.
    The Multi-Agency Data Integration Project is one of these integration projects.

    Hon. Angus Taylor, MP
    Assistant Minister for Cities and Digital Transformation


Two maps showing regional differences to mean Medicare benefits paid to Age Pensioners in Sydney, 2011 and Melbourne, 2011


Across Australia, Medicare services are used by more Age Pension recipients living in the major population centres of each state and territory than those in regional and remote areas.

These maps of Sydney and Melbourne illustrate the further variations in the use of Medicare Services within capital cities. In Sydney (Figure 1), older Australians on the Age Pension make high use of Medicare Services. By contrast, in Melbourne (Figure 2), the use of Medicare services by older Australians on the Age Pension varies considerably from suburb to suburb.

This analysis demonstrates how the better use of existing public data can help inform the development of health policy and the allocation of health care resources to improve older Australians’ access to health care services.

These findings could shed light on policy issues around access to health services and service provision issues such as overprovision, prevalence of health conditions, and demographic factors which warrant further investigation.


CASE STUDY 3: EXPERIMENTAL LAND ACCOUNTS - A COLLABORATION WITH GEOSCIENCE AUSTRALIA

Governments need reliable information about changes in the use, condition and value of land in Australia, and how this relates to broader economic activity and the state of our environment. This information is used in policy making and spending decisions in land management and economic development. Statistical organisations are well placed to provide this vital information by integrating geospatial and environmental data with a range of economic indicators.

The ABS uses the UN System of Environmental-Economic Accounting (SEEA) Central Framework to guide the production of data concerning the environment and economy. Land Accounts form the foundation of all environment-economic accounts, and the ABS has produced a series of Experimental Land Accounts across several jurisdictions in Australia. Rather than collecting the required data itself, the ABS sources and integrates data from a number of government organisations to produce these Land Accounts.

An important input to Land Accounts is data about land cover, both the biophysical cover and the built environment. In recent decades, Earth observation from satellites and other sources has enhanced our ability to consistently observe and monitor the environment. Earth observations are more than merely pictures of the Earth - the data they contain can tell us about the Earth’s surface, waters and atmosphere. Through the application of geoscientific expertise and capabilities the raw satellite observations can be used to detect and map land cover.

Geoscience Australia (GA) is the lead agency in the Australian Government providing Earth observation services, expert advice and capabilities, and information for decision makers. The ABS works in partnership with GA to realise the value of geospatial data and Earth observations to enhance the production of official statistics in Australia. A combination of high level engagement and technical collaboration has strengthened a productive and mutually beneficial inter-agency relationship. This partnership has been critical in ensuring that GA’s Dynamic Land Cover Data (DLCD), a consistent national dataset of land cover, has been available for the production of Experimental Land Accounts.

The use of the DLCD time-series in Land Accounts provides a “line of sight” between Earth observation data produced by GA and the information available to policy makers via statistical products produced by the ABS. This has motivated GA to continue the production of DLCD, and GA has worked closely with the ABS on the release of a new version of DLCD for Australia. This new data became available in time for the preparation of the ‘Land Account: Queensland, Experimental Estimates, 2011-2016’ publication by the ABS, which was released in June 2017.

The ABS and GA are collaborating further on a demonstration of the use of products derived from the Landsat imagery time-series in Digital Earth Australia to produce official statistics on land condition across Australia. The results of this work are due to be released in an ABS environmental-economic accounting publication in 2017-18.
    The partnership we have with the ABS is helping us to showcase the use of the fundamental geographic information we provide for the nation, and we are quietly enabling better decisions to be made via the products released by the ABS. Through the partnership we are also building a better understanding of the needs of our users, which in turn helps us grow our geoscientific capabilities. We truly value the opportunity to work closely with the ABS to identify and resolve issues, explore ideas and look to the future together.”

    Trevor Dhu
    Program Director, Digital Earth Australia, National Earth and Marine Observations Branch, Geoscience Australia


CASE STUDY 4: MAKING A DIFFERENCE - THE ABS DATALAB

Detailed statistical data is becoming easier to access for a growing number of users across Australia through the ABS Data Laboratory (DataLab). The DataLab provides secure access to detailed microdata - the specific variables recorded for each individual respondent - to authorised users for research, policy and statistical research purposes. Initially limited to physical on-site access, DataLab now provides users with virtual access to ABS microdata.

Access to ABS microdata in the DataLab maintains confidentiality and protects privacy, and assesses disclosure risk using the ‘Five Safes Framework’. All users sign a legally binding undertaking to use the microdata in an appropriate way. They also undertake mandatory training and all outputs are checked by ABS staff prior to release from the secure IT environment. This enables the ABS to provide users with access to detailed microdata files safely and securely and help them make the most effective use of data.

Over 350 people from 40 different organisations have been trained to use the DataLab. These include a growing number of staff from a range of Australian, state and territory governments, as well as a steady stream of academic researchers.

The DataLab is also enabling the ABS to provide leadership across government in data sharing and collaboration via a Trusted Access Model. This model ensures that effective, secure and mutually beneficial access to detailed microdata is managed through partnerships based on the trustworthiness and capability of users, as well as the protections of the data and delivery system.

Heather Crawford is an experienced data analyst and social researcher from the Centre for Aboriginal Economic Policy Research, Australian National University. She had been a long time user of the physical DataLab, but was frustrated by the inconvenience of having to visit an ABS office to do her work and the lack of up-to-date statistical analysis tools.

The ABS improved Heather’s user experience by providing her with up-to-date versions of analytical software and allowing virtual access to the DataLab from her own office. These changes improved the timeliness of analytical results and have enabled Heather to undertake more in-depth analysis.

One example of how Heather has used the DataLab to help inform policy through data analysis relates to the closure of the Community Development Employment Projects (CDEP) scheme in 2015.

For over 30 years, the CDEP scheme had offered Aboriginal and Torres Strait Islander people with an alternative to welfare. Heather’s use of the DataLab under-pinned her contributions to the 2016 ANU research publication ‘Better than welfare? Work and livelihoods for Indigenous Australians after CDEP.’ The publication has helped policy makers understand the consequences of the closure the CDEP scheme and provided them with an evidence base from which to develop future policy reforms that are better than CDEP in generating improved livelihood options for Aboriginal and Torres Strait Islander people. This in-depth analysis would not have been possible without the DataLab.


CASE STUDY 5: WORKING WITH DATA61 TO IMPROVE THE NATIONALMAP

The ABS is empowering other organisations to source ABS statistics using their own tools, helping to fulfil the ABS vision of unleashing the power of statistics for a better Australia.

NationalMap, hosted by the data.gov.au website, offers a platform for discovering and visualising a broad range of government and non-government datasets across various geographies of Australia.

CSIRO’s Data61 has been working with the Department of the Prime Minister and Cabinet to improve the usability, functionality and interactivity of NationalMap.

In 2016-17, the ABS collaborated with Data61 to develop a better way to provide ABS statistics directly to NationalMap, enhancing its functionality. This included increasing the amount of ABS data available and ensuring that NationalMap can always display the latest ABS statistics - without having to be manually updated.

These improvements were made possible by upgrades to the ABS’s Data Services in September 2016 that included a new developer-friendly data format, SDMX-JSON, that allows our external users to directly source ABS statistics.

The NationalMap platform has a number of features to maximise the use of ABS data:
    • It brings together a broad range of data collections from a variety of government and non-government sources
    • Data is displayed in an interactive map, making it easier for users to understand geographically displayed data
    • The value of data is enhanced because it can be displayed together with a variety of other data to allow analysis
    • Users can incorporate their own data into the platform for display with existing datasets
    • The NationalMap technology can be embedded into other websites to enable data to be easily found and displayed.

The cross-government collaboration between Data61 and ABS has been mutually beneficial and is an example of how agencies can work together to enhance the usability of government services for the Australian public.

Arthur Street, Senior Software Engineer from Data61 was responsible for the project and remarked “ABS has made an incredibly rich and varied range of data available via its SDMX API. As a result we have been able to add lots of new datasets to NationalMap.gov.au - from population forecasts in 2036, to where the most fruit and nut orchards are - which we expect will be put to lots of interesting new uses.”

The use of ABS Data Services ensures the latest ABS statistics are always available through NationalMap as well as other platforms in the future. Best of all it extends the scope, use and usefulness of ABS data.


CASE STUDY 6: THE FREIGHT PERFORMANCE MEASUREMENT PROJECT - A COLLABORATION WITH THE BUREAU OF INFRASTRUCTURE, TRANSPORT AND REGIONAL ECONOMICS

Since the late 1960s, the amount of road freight in Australia has increased tenfold: from just over 20 billion tonne-kilometres a year, to over 200 billion tonne-kilometres in 2015-16. Building a better understanding of road freight movements in Australia helps to identify road infrastructure bottlenecks and establish priorities to better inform road infrastructure investment decisions.

Throughout 2016 the ABS partnered with the Bureau of Infrastructure, Transport and Regional Economics (BITRE) to investigate new ways of collecting and analysing road freight statistics as part of the Public Sector Data Management Project led by the Department of the Prime Minister and Cabinet. This pilot study is known as the Freight Performance Measurement Project.

Road transport in Australia is a highly competitive, efficient, and strongly regulated industry. In a fast growing industry like road freight, a lack of regular, up-to-date information has hindered effective decisions about where and when to invest for both the public and private sectors.

Road freight companies have devices fitted to their vehicles that record GPS coordinates at time-stamped intervals, along with information such as speed and fuel usage. This administrative data, known as telematics, has the potential to provide a less burdensome, more cost-effective and more timely source of road freight statistics in comparison to traditional survey based data collection methods.

The ABS contributed expertise in new data analytics techniques to the project, including further development of GLIDE (endnote 11). GLIDE, an ABS developed platform, allows complex data in a variety of formats and from multiple sources to be linked and analysed in an interactive, visual manner. It provides tools for visualising data in different ways - such as via interactive maps. Using the GLIDE prototype, the project was able to produce freight specific traffic volumes and travel times along major corridors, and then drill down to identify congested freight network locations. GLIDE also enabled the visualisation of infrastructure used by freight vehicles, such as key freight routes and truck stops.

The Freight Performance Measurement Project has demonstrated the feasibility of using freight vehicle telematics data to provide regular road freight statistics and has shown the utility of GLIDE to identify congested road freight network locations and calculate average travel times on major freight routes.

Dr Gary Dolman of BITRE said “For BITRE, the joint collaboration with ABS on the Freight Performance Measurement Project, helped better identify and engage with key stakeholders, and lent more authority and credibility to the project when discussing with external stakeholders. The resources, assistance and support provided by the Emerging Data and Methods Team was invaluable to processing the pilot study sample data and developing preliminary results and effectively communicating those results back to key stakeholders.”

This collaborative project illustrates the value of integrating existing data to produce new information and insights that, in turn, inform important decisions. It is one of a number of current ABS data integration projects being undertaken in partnership with other government agencies using new and emerging statistical techniques.

For more information please see the Department of the Prime Minister and Cabinet Public Sector Data Management Project.


ENDNOTES
1. ABS Portfolio Budget Statement is contained in Budget 2016-17 - Portfolio Budget Statements 2016-17 - Budget Related Paper No. 1.16 - Treasury Portfolio.
2. The ABS purpose was slightly modified in August 2016 and is listed under Performance Monitoring and Reporting (page 24) in the ABS Corporate Plan 2016-17 (cat. no. 1005.0), which is available on the ABS website.
3. The IMF monitors developments in its member countries, as well as at regional and global levels, to ascertain potential sources of economic and financial instability. This process is known as surveillance.
4. Based on assessment of 3033 articles in a number of key media outlets citing ABS statistics (identified through a contracted media monitoring service).
5. The Australian Bureau of Statistics Act (1975) stipulates that the Australian Statistics Advisory Council must meet at least once in every calendar year (section 22(1)). The Council will meet three times in the 2017 calendar year.
6. As indicated by the historical data in Figure 4.1, this result is largely due to 2016-17 being an Agricultural Census year. This collection is conducted once every five years. For the current reporting year (2016-17), the Agricultural Census resulted in approximately 103,000 hours of provider load.
7. Year on year comparisons should be treated with caution due to the cyclical nature of significant collections, such as the Agricultural Census, which is conducted every five years.
8. The appetite for web self-response varies across business survey providers. Some industries, particularly those in scope for agricultural collections, have a lower propensity to complete surveys online.
9. The population is defined as all Australian Usual Residents (excluding overseas visitors) who had the opportunity to complete an online household form. The population used excludes people that were counted in the Census through Interviewer Household Forms, Special Short Forms and in non-private dwellings, as they may not have had the opportunity to participate online. The population used accounts for 95.7% of the total responding population and 96.8% of the responding Usual Resident population.
10. This target refers to actions from the ABS 2015 & 2016 People and Culture Action Plan and the ABS 2017 & 2018 People and Culture Action Plan as there was no People and Culture Action Plan for the 2016-17 financial year.
11. Graphically-Linked Information Discovery Environment.