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 Page Print All | ||
INDEPENDENCE FROM THE CENSUS
Population independence requires that people who were not counted in the Census are no more likely to be missed by the PES than people who were counted. In practice, some population subgroups may be less likely to respond to both the Census and the PES. DIFFERENCES BETWEEN THE CENSUS AND THE PES Householders are asked the PES questions using a personal interview by experienced, highly trained interviewers, whereas most Census forms are completed by a responsible adult on behalf of their household. The PES is also a much smaller scale operation (and hence easier to control) than the Census. These features enable the PES to deliver an accurate estimate of the percentage of people and dwellings missed or overcounted by the Census. Just as the Census misses some people and dwellings that PES finds, so too the PES misses some people and dwellings that Census finds. In some cases this is because PES excludes certain kinds of persons and dwellings from scope for practical reasons, such as homeless persons and non-private dwellings (hotels, motels, hospitals). In other cases, PES misses people through non-contact or refusal of the occupants of a selected dwelling. The PES uses weight adjustment during estimation to account for the dwellings and people missed in the PES but counted in the Census. LATE CENSUS FORMS For some people who have not returned a Census form, contact from the ABS following selection in the PES acts as a reminder and possible motivator to return a completed Census form. These late returns, if not identified, would result in the PES sample having a higher proportion of Census response than in the overall population. To protect against this, all Census forms received after the start of PES enumeration are deemed 'late' and can't be used to improve accuracy of PES estimates. CORRELATION BIAS On the assumption that the Census and the PES are independent, the estimate of the percentage of persons missed by the PES but found by the Census, and the percentage of persons missed by the Census but found by the PES, can be used to construct estimates of the percentage of persons missed by both the PES and the Census. Despite efforts to maintain independence, the likelihood of a person being missed in the PES may be related to whether they were missed in the Census. This would result in a 'correlation bias' in the PES estimates. To minimise this bias, sample weights can be adjusted during the PES estimation process to take account of the fact that different groups have a different likelihood of being missed. Document Selection These documents will be presented in a new window.
|