PREAMBLE


The Australian Bureau of Statistics (ABS) is Australia's official statistical agency and is committed to leading a high quality statistical service to assist and encourage quality informed decision making. As part of this service the ABS is committed to providing leadership in quality management of statistical processes.

National statistical agencies need to find new and improved ways of acquiring information about their target populations to reduce respondent burden, maximise available data usage and find efficiencies in the use of taxpayers' money. This has led to the use of information that is collected for purposes other than that of a statistical nature. This type of information is often obtained from records or transactional data from government agencies, businesses or non-profit organisations which use the information for administrative purposes. Examples of administrative data include:

• registration of births, deaths and marriages;
• customs data (e.g. import and export records);
• taxation data;
• financial data (e.g. automatic teller machine transactions);
• social security benefit applications;
• school enrolments; and
• hospital records, to name a few.

This paper provides information on managing the quality of statistics produced from administrative data. It focusses on some principles and best practices to assist in the management of the acquisition of administrative data. It will be useful for any agency considering entering into a relationship regarding the supply and access of administrative data, as well as those agencies already involved in this area. A future publication is intended for release to cover processing issues which are not discussed in this paper.

This paper is a continuation of the ABS' role in providing guidance on frameworks and best practice in the use of statistics, including those produced from administrative data. It can be used in conjunction with other previously released ABS papers in the quality management series:

ABS Data Quality Framework, May 2009, (cat. no. 1520.0); and

Quality Management of Statistical Risk Using Quality Gates, Dec 2010, (cat. no. 1540.0).

Another useful source for quality management information is the National Statistical Services' Data Quality Online (http://www.nss.gov.au/dataquality/) tool which provides detailed information on the application of the ABS Data Quality Framework.


INTRODUCTION


Administrative data consists of information collected about a population either directly or indirectly. It is collected by organisations for different administrative purposes including:

• for regulatory purposes (e.g. motor vehicle registration);
• for the administration of various government programs (e.g. social security benefits, education, and health);
• for legislative requirements (e.g. taxation declarations);
• for analysis (e.g. supermarket scanner data);
• as a by-product of transactions (e.g. electronic funds transfer at the point of sale (EFTPOS)).

The administrative records can also be used for statistical purposes and in doing so can offer important advantages over direct collection of data from the population concerned.

Some examples of data sourced from administrative systems which have been used in ABS publications include:

• Register of Births, Deaths and Marriages provides the ABS with data which is used in population estimates and projections. Data from this source can also be used to measure likely demand for services associated with population growth, such as housing, child-care and schools (Births, Australia - ABS (cat. no. 3301.0); Deaths, Australia - ABS (cat. no. 3302.0))
• Motor Vehicle Registries provide snapshots of their registers to be used as a population frame for the Survey of Motor Vehicle Use, (the data from the snapshot is published in Motor Vehicle Census - ABS (cat. no. 9309.0))
• Department of Education collects information on school enrolments (Schools, Australia - ABS (cat. no. 4220.0))
• Customs documentation completed for goods sent or received at an international sea or air port, provide the basis for overseas trade data and export and import statistics (International Trade in Goods and Services (cat. no. 5368.0))
• Local government authorities' collection of information on building approvals is used to construct an indicator of economic activity (Building Approvals, Australia - ABS (cat. no. 8731.0))
• The Australian Taxation Office collects information on personal and business taxation which is used by the ABS. Publications such as Wage and Salary Earner Statistics for Small Areas, Australia (cat. no. 5673.0.55.003), the National Regional Profile (cat. no. 1379.0.55.001) and Estimates of Personal Income for Small Areas, Australia (cat. no. 6524.0.55.002) are examples of some of the uses of this administrative data.

In many ways the use of administrative data for the production of statistics or statistical analysis is akin to the production of statistics from data collected directly from respondents, as is done with a census or survey. Many of the principles and methods of best practice for the various processes involved apply regardless of the source, and the approaches for managing quality are the same. However, there are important differences which include (but are not limited to) the addition of at least one extra party in the acquisition of data and the fact that the administrative data are not collected for the sole purpose of statistical analysis. These differences and others need to be taken into consideration when using administrative data.

This paper will define the major uses of administrative data for statistical purposes and the quality management techniques which are important in the production of statistics. While this paper focuses primarily on acquiring electronic administrative data the principles can be applied to other forms of administrative data such as hospital record forms. Later releases will examine the processing of administrative data throughout the whole statistical cycle.

The term "data custodian" will be used to represent those agencies or organisations who collect administrative records as part of their normal business process requirements. "Receiving agency" will refer to those agencies or organisations who utilise administrative data for statistical purposes. Receiving agencies may include national statistical agencies or other government agencies or businesses who use administrative data to produce statistical outputs. In some cases it is possible that the data custodian and the receiving agency are the same entity.