REVISIONS IN THE AUSTRALIAN LABOUR ACCOUNT
Revisions are a change in the value of a published estimate. Revisions arise from the correction of errors, the incorporation of more up-to-date data, reassessment of seasonal factors, and from time to time the introduction of new concepts or improved data sources and methods.
Revisions are an inevitable consequence of the process of producing the Australian Labour Account. Revisions reflect both the complexity of measurement, and the need to trade off some level of precision in order to provide timely estimates, to maximise their use in analysis of current economic conditions.
Sources of Revisions
- Updates to the Estimated Resident Population (ERP), usually affecting the latest eight quarters of data, resulting in quarterly revisions to the Labour Force Survey statistics on persons, jobs and hours worked;
- Revisions to QBIS statistics on filled jobs, arising from replacement of imputed data with actual responses following late receipt of survey questionnaires; and
- Revisions to previously published seasonally adjusted and trend series, which will be revised to incorporate the seasonal effects of the latest quarterly data.
Other Periodic Revisions
- Revisions which reflect the cumulative impact of previous revisions to quarterly data;
- Revisions to EAS statistics on filled jobs, arising from replacement of imputed data with actual responses following late receipt of survey questionnaires;
- Revisions to Compensation of Employees and GMI following annual benchmarking of National Accounts, usually affecting the latest three years of quarterly data; and
- Revisions to expenditure on recruitment services and training, following release of updated Input-Output Tables.
- Five yearly post-Census benchmarking of ERP, resulting in revisions to the household Labour Force Survey statistics on persons, jobs and hours worked; and
- Revisions to Compensation of Employees and GMI arising from scheduled National Accounts historical revisions, affecting quarterly data back to 1960.
- All data sources can be subject to revisions arising from the correction of errors. These can include data capture and compilation errors, mistakes in classification, or respondent misreporting; and
- Australian Labour Account data are also subject to revision arising from internal compilation errors.
ABS and international data quality assessment frameworks include revisions history as one of the indicators of quality. A revisions history assists users in assessing the probability and potential scale of change to published data. As part of an overall quality assessment program, the ABS intends to undertake annual analysis of revisions in order to inform users, and to help guide the continuous quality improvement process.