3405.0.55.001 - Discussion Paper: Assessment of Methods for Developing Experimental Historical Estimates for Regional Internal Migration, Dec 2011  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 22/12/2011  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All

SUMMARY


Internal migration data provide an important indicator of social, economic, and environmental issues and trends at the regional level.

Before experimental regional internal migration estimates were created as part of this project, the only source of sub-state regional migration was the five-yearly Census of Population and Housing.

The ABS uses data on change of Medicare enrolment address to estimate movements of people across state/territory boundaries. The data have not previously been deemed suitable for smaller regions mainly because they are supplied to ABS by postcode, which is not directly compatible with the ASGC.

The ABS has developed postcode to ASGC correspondences, based on the estimated population distribution across postcodes, which can be used to convert postcode level data to SLA.

Medicare and Defence data provide movement data at the postcode level, which can be converted to ASGC regions. This discussion paper presents the limitations of the method of converting postcode based internal migration data to create migration estimates at the SLA level. The regional internal migration estimates have been improved by applying adjustments to take account of the movers that may be missed when postcodes are split across SLAs and by combining selected SLAs to improve the quality of converted data.

Adjustments taking into account the following factors can be made to the Medicare and Defence data.

  • State/territory expansion factors: already calculated by ABS for the estimation of interstate migration, these factors can be applied to the postcode-based Medicare movement data to account for known under-reporting of moves made by people, based on age and sex.
  • Missed Movers Index: adjustments can be applied to estimate the number of moves missed between a pair of SLAs in the postcode-based migration data, due to part or all of those SLAs being within the same postcode.
  • Concordance Confidence Index: if there is little confidence in the quality of SLA-level data converted from a postcode, SLAs can be combined with neighbouring SLAs to provide a more confident conversion from that postcode for groups of SLAs.
  • Constraining to known state totals: the ABS already publishes estimates of interstate migration, which should be used as a control total for sub-state internal migration estimates.

Each of these methods contributes to improving the data series by accounting for some of the limitations of converting data from postcode to SLAs.

From 2012, the Australian Government Department of Human Services will provide Medicare data to the ABS at the SA2 level. The provision of data at this level should improve the quality of sub-state internal migration estimates by eliminating the quality issues associated with converting from postcode-based data. This includes removing the need to apply the Missed Movers Index and the Concordance Confidence Index.