2023 Post-enumeration Survey

The 2023 Post-enumeration Survey (PES) was the sixth to be undertaken in New Zealand since the inaugural PES in 1996. The 2023 PES, like its predecessors, was a sample survey of individuals in private dwellings. Survey collections were carried out in two waves due to extended census collections in the Auckland area.  

• Wave 1: Collection took place in all regions except for Auckland between 2 June and 13 August 2023. 

• Wave 2: Collection took place in the Auckland region between 1 July and 31 August 2023. 

Around 16,500 dwellings were included which made up around 37,500 individual responses.

Frequency: Five-yearly

^en-NZ

The main purpose of the post-enumeration survey (PES) is to measure the level of coverage (undercount and overcount) in the Census. Coverage measures are used as key performance indicators for the 2023 Census, with the PES constituting a major part of the evaluation component for the 2023 Census. PES results are a key input to population estimates and projections produced after the census. The 2023 PES increased the sample size from 2018 by ten percent to approximately 16,500 dwellings, along with other planned quality improvement initiatives (e.g improvements to the dwelling frame selection), to enable more robust evaluation of census. See 2023 Post-enumeration Survey: Standard design for coverage estimation | Stats NZ for a more comprehensive discussion of improvements to 2023 PES.

^en-NZ

Survey Objectives

The survey objectives of the 2023 PES were to measure the level of coverage by the 2023 Census of Population and Dwellings of permanent private dwellings and of New Zealand residents in New Zealand on census night.

The PES was designed to produce estimates of gross undercount, gross overcount and net coverage for the following key population groups:

• total usually resident population.

• gender (male, female, people of another gender).

• age (0–14 years, 15–29 years, 30–44 years, 45–64 years, 65-74, 75+ years).

• ethnicity (the European, Māori, Pacific peoples, Asian, MELAA (Middle Eastern, Latin American, and African) and Other groupings of ethnicities, together with Chinese, Indian and Samoan, which contribute coverage measures for population projections).

• birthplace (New Zealand born, overseas born) and the Māori descent population.

• geographical area in two groupings: Regional Council (RC), and Territorial Authority Local Board (TALB).
  PES in New Zealand sets out to:

• reach most people who were randomly missed by census.

• look for missing people via other methods such as through administrative data (data collected by other government agencies which some international statistical agencies use to improve their coverage estimates).

• reach some people who have a lower likelihood of being found.

However, PES was not specifically designed to locate people who intend to remain invisible and take sufficient steps not to be found.

Main users of the data

The main users of the PES data are those concerned with the performance of 2023 Census in counting people who usually reside here in New Zealand.

For Stats NZ, PES data is of high importance for creating the 2023 base Estimated Resident Population (ERP) and for assessing some of the key performance indicators of the 2023 Census.

More widely, this data is of interest to population planners, researchers and other government agencies.

2023 PES Key Performance Indicators

The PES has its own KPI targets, intended to focus our design and resources in the areas that are critical for meeting both survey objectives and project outcomes. Two KPI targets set for 2023 PES data collection were the survey response rate, and the achieved sample rate.

Survey Response Rate

The survey response rate is calculated as the percentage of eligible dwellings that respond to the survey.

		Eligible responding dwellings
Response rate	=	___________________
		Eligible dwellings

In 2023, the PES response rate was 89 percent, just below the 90 percent target. In comparison, the 2018 PES response rate was 91 percent. 

Achieved Sample Rate (ASR)

The achieved sample rate (ASR) is calculated as the number of eligible dwellings responding as a proportion of all dwellings sampled.

		Eligible responding dwellings
Achieved Sample Rate	=	____________________
		Sampled dwellings

The Achieved Sample Rate (ASR) which measures how many eligible dwellings responded to the survey was 84 percent, thus achieving the KPI target (82 percent). In comparison, the 2018 PES ASR was 82 percent.

Description of sampling methods

A summary of the final 2023 PES sample design and sample selection is described as follows. The Stats NZ national framework of Primary Sampling Units (PSUs) were grouped in 112 strata based on broad regions, a mix of urban and rural living, the 2023 Census enablement strategy, and socio-economic characteristics captured in the NZ Deprivation Index. See Field collection strategies and operations for the 2023 Census(linked in the related documentation section) for further information on 2023 Census data collection strategies.

Sample selection across TALBs was implicitly achieved through the selection process which aimed to maximise the probability of selecting at least one PSU in each TALB. The 2023 PES sample design does not include any explicit over-sampling of any demographic group. Instead, we achieved the necessary over-sampling implicitly through the strata design. The national over-sampling rates, in the final PSU selection, for key groups of interest were:

• Māori descent 3.3 percent

• Māori ethnicity 2.7 percent

• Pacific peoples ethnic group 0.4 percent

• Asian ethnic group 0.6 percent

^en-NZ