2940.0.55.002 - Information Paper: Measuring Overcount and Undercount in the 2016 Population Census, Jul 2016  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 01/07/2016  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All


The PES interview determines whether each person in the sample should have been counted in the Census, and the linking and matching process determines how many times each person was actually counted in the Census. The PES estimation process combines and weights these data to produce an estimate of the number of people who should have been counted in the Census. This difference between this and the number of people who were actually counted in Census is the estimate of net overcount or net undercount.


Weighting is the process of adjusting results from a sample survey to infer results for the total in-scope population. The weight can be considered an indication of how many people or dwellings are represented by those in the sample.

To derive estimates for the entire population in the scope of the PES, expansion factors (weights) are applied to the sample responses. The weighting method ensures that estimates conform to known population totals by age, sex and geographic area. It considers population information such as the number of dwellings of different characteristics that were located in Census and the number of people with different characteristics that were counted in the Census. This reduces sampling variability and compensates for any under-enumeration or non-response in the survey.

The underlying principle adhered to in the weighting is that any person in the sample will receive the same weight, regardless of whether or not they were counted in the Census. In a simplistic example the weights of PES people that were counted in the Census should be adjusted so that the sum of the weights is equal to the number of people counted in the Census. The weights applied to PES sample respondents who were not counted in Census should be adjusted in the same way as the weights of those that were counted. The sum of the weights of all PES respondents will then be an estimate of the number of people that should have been counted in the Census.

The weighting classes classify the population by:

  • state or territory and geographic region;
  • sex;
  • age;
  • country of birth;
  • marital status;
  • Aboriginal and/or Torres Strait Islander origin and
  • whether selected from a Discrete Community.

The weight adjustment applied to a person depends only on the information reported in the PES.

As a final step in weighting, weights are adjusted so that the PES estimates represent people in non-private dwellings as well as private dwellings. This final step uses weighting class categories of state or territory, region, age and sex only, as information on other items is not considered reliable for non-private dwellings.

The method used for adjusting weights to meet population totals is the Prediction Regression (PREG) estimator. PREG estimation is able to take into account people counted multiple times in the Census and people who give different responses between PES and Census. A detailed description of the PREG estimator can be found in Research Paper: An Estimating Equation Approach to Census Coverage Adjustment, May 2007 (cat. no. 1351.0.55.019).