Page 318 - Special Topic Session (STS) - Volume 3
P. 318
STS547 John D. et al.
New population estimation methods: New
Zealand and Ireland
2
John Dunne , Patrick Graham
1
1 CSO, Ireland
2 Stats NZ, New Zealand
Abstract
Like many countries, New Zealand and Ireland are researching new systems of
population estimates compiled using administrative data. Neither Ireland or
New Zealand have a Central Population Register from which the estimates can
be compiled. Projects in both countries involve first creating a Statistical
Population Dataset (SPD) to represent the population and then adjusting for
coverage errors.
In Ireland, the SPD is compiled such that by design it will only have
undercoverage. The SPD counts are then adjusted using Dual System
Estimation (DSE). An extension of DSE, Trimmed Dual System Estimation
(TDSE), is then used to verify that no overcoverage exist, or if overcoverage in
the SPD is present then those parts of the SPD identified with overcoverage
are removed prior to an application of traditional DSE methods to compile
population estimates. While the use of TDSE in official statistics is new, this
system is also novel in that it uses an administrative data list as list B in the
application of DSE methods.
In New Zealand, the SPD is compiled to represent the population as close as
possible and as such is expected to contain both undercoverage and
overcoverage. The New Zealand approach is novel in that it uses Bayesian
methods and a single sample from the target population to correct for both
undercoverage and overcoverage at the same time.
This short paper will describe both the Irish and New Zealand approaches as
well as discuss the strengths and weaknesses of each approach.
Keywords
Bayesian, TDSE, overcoverage, coverage, SPD
1. Introduction
Statistical agencies in many countries are investigating methods for
replacing traditional census based population estimation systems. Ireland and
New Zealand are two such countries. Neither country has a Central Population
Register (CPR) but both countries have invested significant resources into the
exploitation of administrative data sources for statistical purposes (Dunne,
Corresponding Author: John Dunne, Central Statistics Office, Ireland; E-mail:
John.Dunne@cso.ie.
307 | I S I W S C 2 0 1 9