Page 310 - Special Topic Session (STS) - Volume 3
P. 310
STS547 Owen A.
Dual-system estimation by another name:
Population size estimation using weighting
classes
Owen Abbott
Office for National Statistics, Fareham, UK.
Abstract
This presentation will outline the application of a number of approaches to
population size estimation, applied to an administrative based system. The
Office for National Statistics (ONS) is exploring models for the future of
population statistics. The UK does not have a population register or a set of
coherent identifiers across administrative datasets held by government. The
current population statistics system is underpinned by the decennial Census,
which is expensive and is arguably becoming increasingly unwieldy as a source
of data in a rapidly evolving society and with ever increasing demands for
more timely, relevant statistics. The ONS is therefore researching how it can
transform its population statistics system within that context, and the most
important part is the estimation of population size.
The expected sources of data for estimation include administrative data
sources and a population coverage survey designed for the purposes of
estimating population size, much like a Post-enumeration Survey. Classic
capture-recapture estimators can be applied, but they rely on heavy
assumptions such as minimal over-coverage, which may not be the case for
the administrative datasets. Alternative estimators have been explored, as well
as ways of processing the administrative data in advance. Previous research
(Abbott et al, 2015) applied a weighting-class approach to the measurement
of coverage in a Census. This form of estimator can be thought of as a
modified dual-system estimation, with a different set of assumptions.
The presentation will discuss the different flavours of dual-system estimation
in use across National Statistical Institutes for estimating population size,
including the Bayesian approaches developed by Statistics New Zealand, and
where they differ in terms of the assumptions that underpin each application.
The research into exploring these methods in the UK context will be outlined,
with some early results.
Keywords
Administrative data; Official statistics; capture-recpture
1. Introduction
The Office for National Statistics (ONS) is exploring models for the future
of population statistics. The UK does not have a population register or a set of
coherent identifiers across administrative datasets held by government. The
299 | I S I W S C 2 0 1 9