FEATURE ARTICLE 3: PROPOSED ABS TREATMENT OF INTERCENSAL ERROR FOR 2006-2011
The intercensal error for the 2006-2011 period is the highest on record, and the ABS acknowledges the challenge that improvements to the 2011 Census Post Enumeration Survey (PES) pose to the traditional rebasing methodology for ERP.
The results of the Automated Data Linking (ADL) Statistical Impact Study estimated that the introduction of ADL resulted in an undercount that was 247,000 less than the previous methodology would have produced. This challenges the traditional notion that intercensal error is largely the accumulation of error in the components of population change over the intercensal period, rather than imprecision in the base reference years.
PREVIOUS COMMENTARY ON THE ISSUE
This article follows on from the technical note The Impact of Improvements to the 2011 PES on Measuring Population Growth (2006-2011) that first appeared in the 20 June 2012 release of Australian Demographic Statistics (cat. no. 3101.0) and the feature article Advice on the use of 2011 Preliminary rebased ERP. For ease of reference, this article is repeated in this edition of Australian Demographic Statistics (cat. no. 3101.0).
PLANS FOR FINAL REBASING
The ABS is proposing to treat the 2006-2011 intercensal error in the final rebasing differently to the methodology used for preliminary rebasing, and historically for other rebasing periods.
The current proposal for final ERP (to be released 20 June 2013) is for the 2006-2011 intercensal error to be evenly distributed over a 20 year period back from June 2011 to September 1991. This will result in an historical revision to part of the ERP series at the national, state and territory, and sub-state levels.
The ABS treatment will be finalised in late 2012 with further details to be released in Australian Demographic Statistics (cat. no. 3101.0) on 18 December 2012.
The ABS encourages feedback on this proposed treatment please contact Bjorn Jarvis on (02) 6252 6411, Phil Browning on (02) 6252 6639, or email email@example.com.