Quality Statement

Label
Usual residence five years ago - 2023 Census: Information by concept en-NZ
Definition

Usual residence five years ago address identifies an individual’s usual residence on 6 March 2018, which may be different to their current usual residence (7 March 2023). This variable reports an individual’s usual residence five years ago at the lowest possible geographic level.

Usual residence five years ago indicator determines if there is a difference between an individual’s usual residence and their usual residence five years ago. It provides information on the migration of people within New Zealand and of those who have either been born or arrived from overseas over the last five years.

en-NZ
Overall quality rating

Usual residence five years ago address: High quality
Usual residence five years ago indicator: High quality
Data quality processes section below has more detail on the rating.

en-NZ
Priority level

Priority level 2
A priority level is assigned to all census concepts: priority 1, 2, or 3 (with 1 being highest and 3 being the lowest priority).
Usual residence five years ago is a priority 2 concept. Priority 2 concepts cover key subject populations that are important for policy development, evaluation, or monitoring. These concepts are given second priority in terms of quality, time, and resources across all phases of a census.
The census priority level for usual residence five years ago remains the same as 2018.

The 2023 Census: Final content report has more information on priority ratings for census concepts.

en-NZ
Subject population

Census usually resident population count
‘Subject population’ means the people, families, households, or dwellings that the variable applies to.

en-NZ
How this data is classified

Usual residence five years ago address
The usual residence five years ago classification consists of a combination of geographic classifications that are ordinarily stored independently of each other. There is a hierarchical relationship between the New Zealand geographic classifications of meshblock, statistical area 2, territorial authority, and regional council. They are combined for the purpose of providing as much detail as possible for usual residence five years ago. Ideally usual residence five years ago would be defined at the meshblock level. Where this is not possible, the next most detailed geography is used.

Usual residence five years ago address is a flat classification. The standard codes are:

Usual Residence 5 Years Ago Classification V3.0.0 – level 1 of 1

  • Meshblock codes (7 digits)
  • Statistical area 2 codes (6 digits) prefixed by '9'
  • Territorial authority codes (3 digits) prefixed by '9999'
  • Regional council codes (2 digits) prefixed by '99999'
  • Country codes (4 digits) prefixed by '999'
  • 9999977 Response unidentifiable
  • 9999988 Response outside scope
  • 9999994 Unable to match to admin data
  • 9999995 Not born 5 years ago
  • 9999996 No fixed abode
  • 9999997 Country not stated
  • 9999998 New Zealand not further defined
  • 9999999 Not stated

Follow the link above the table to examine the classification in more detail.

The 2023 classification used for usual residence five year ago address is similar to 2018 in structure, however area codes have been updated to reflect the 2023 geographic pattern.

Usual residence five years ago indicator
Usual residence five year ago indicator is classified into the following categories:

Usual Residence 5 Years Ago Indicator Classification 2 V4.0.0 – level 1 of 2

Code Category
1 Same as usual residence
2 Elsewhere in New Zealand
3 Not born five years ago
4 Overseas
5 No fixed abode five years ago
9 Not elsewhere included

Usual residence five years ago indicator uses a 2-level hierarchical classification with level 1 presented in the table above. The level 1 residual category ‘Not elsewhere included’ contains the residual categories ‘Response unidentifiable’, ‘Unable to match to admin data’, ‘New Zealand not further defined’, and ‘Not stated’. However, for the 2023 Census, the only residual category used is ‘Unable to match to admin data’.

Follow the link above the table to examine the classification in more detail.

The 2023 Census classification for usual residence five years ago indicator is consistent with that used in 2018 Census.

Usual residence five years ago indicator can also be summarised by:

The summary variables are different than in the 2018 Census as they have been split into two variables to reflect the different geography levels.

Standards and classifications has information on what classifications are, how they are reviewed, where they are stored, and how to provide feedback on them.

en-NZ
Question format

There was no question on the 2023 Census form for usual residence five years ago. Instead, data for the subject population was sourced from the 2018 Census and admin data.

For more information on how this data was sourced, see Data sources.

en-NZ
Examples of how this data is used

Data-use outside Stats NZ:

  • to understand population mobility within New Zealand
  • to support long-term service planning (for example, housing and transport planning, public health planning).

Data-use by Stats NZ:

  • to measure and identify trends in internal migration
  • to contribute to the quality of subnational population estimates and projections
  • to construct matrices to examine inflows and outflows in more detail (for example, regional council area of usual residence five years ago (that is, 2018 address) by usual residence regional council area (that is, 2023 address)
  • for distinguishing movers and non-movers.
en-NZ
Data sources

Usual residence five years ago address
The table below shows the distribution of data sources for usual residence five years ago address data.

Data sources for usual residence five years ago address data, as a percentage of census usually resident population count, 2023 Census
Source of usual residence five years ago address data Percent
2023 Census response 0.0
Historical census 71.7
 2018 Census 71.7
 2013 Census 0.0
Admin data 26.2
Deterministic derivation 0.2
Statistical imputation 0.0
No information 1.9
Total 100.0
Note: Due to rounding, individual figures may not always sum to stated total(s) or score contributions.

This question was not asked on the form in the 2023 Census, so the majority of data for this variable has come from the 2018 Census. If 2018 Census data was not available, admin data was used.

Admin data was used to identify people aged under five, and therefore not born five years ago. The Department of Internal Affairs was prioritised above other admin data sources. The other admin data sources are:

  • Accident Compensation Corporation
  • Auckland City Mission
  • Early Start Programme
  • Inland Revenue
  • Ministry of Business, Innovation and Employment
  • Ministry of Education
  • Ministry of Justice
  • Ministry of Health
  • Ministry of Social Development
  • New Zealand Transport Authority
  • Oranga Tamariki
  • Stats NZ surveys.

Data from New Zealand Customs Service was used to identify people who were overseas five years ago.

For the remainder, admin address information from the following sources was used:

  • Accident Compensation Corporation
  • 2013 Census
  • Department of Internal Affairs
  • Inland Revenue
  • Kāinga Ora
  • Ministry of Education
  • Ministry of Health
  • Ministry of Social Development
  • New Zealand Transport Agency.

2013 Census addresses is used as an input to the models for determining usual residence address from admin data, but is not split out as a separate data source.

Predicting usual residence address from admin data in the 2023 Census describes this process in more detail.

Deterministic derivation was used for individuals aged 0 to 5 years where age was not able to be determined by admin data but where their usual residence five years ago was coded to ‘Not born five years ago’.

All remaining records where usual residence five years ago information could not be found, or there was conflicting information between age and usual residence five years ago data, were coded to ‘No information’.

Usual residence five years ago indicator
Usual residence address and usual residence five years ago address were used to derive the usual residence five years ago indicator. The data sources from each of the input variables can be found above and in the Usual residence address – 2023 Information by concept.

Editing, data sources, and imputation in the 2023 Census describes how data quality is improved by editing and how missing and residual responses are filled with alternative data sources (admin data and historical census responses) or statistical imputation. This webpage also contains a spreadsheet that provides additional detail on the admin data sources.

en-NZ
Missing and residual responses

Missing responses represent data gaps where it was not possible to find individuals’ usual residence five years ago address data in the Integrated Data Infrastructure.

Usual residence five years ago address
For usual residence five years ago address, ‘No information’ is the percentage of the subject population for whom their usual residence five years ago address was coded to residual categories ‘Unable to match to admin data’ and ‘Not stated’.

Percentage of ‘No information’ for the census usually resident population count:

  • 2023: 1.9 percent
  • 2018: 14.6 percent
  • 2013: 3.5 percent

Percentage of ‘New Zealand not further defined’ for the census usually resident population count:

  • 2023: 0.0 percent
  • 2018: <0.1 percent
  • 2013: 5.3 percent

Usual residence five years ago indicator
In 2023 and 2018 Censuses ‘No information’ is the percentage of the subject population for whom their usual residence five years ago indicator was coded to residual category ‘Unable to match to admin data’.

In 2013 Census, ‘No information’ is the percentage of the subject population for whom their usual residence five years ago indicator was coded to residual category 'New Zealand not further defined', 'Not stated', 'Response unidentifiable', and ‘Unable to match to admin data’.

Percentage of ‘No information’ for the census usually resident population count:

  • 2023: 1.9 percent
  • 2018: 14.6 percent
  • 2013: 9.0 percent
en-NZ
Data quality processes

Overall quality rating:
Usual residence five years ago address: High
Usual residence five years ago indicator: High

Data has been evaluated to assess whether it meets quality standards and is suitable for use.

Two quality metrics contributed to the overall quality rating:

  • data sources and coverage
  • consistency and coherence.

Accuracy of response did not contribute to the overall quality rating as usual residence five years ago was not a question on the 2023 Census form.

The lowest rated metric determines the overall quality rating.

Data quality assurance in the 2023 Census provides more information on the quality rating scale.

Data sources and coverage:

  • Usual residence five years ago address: High quality
  • Usual residence five years ago indicator: High quality

The quality of all the data sources that contribute to the output for the variable were assessed. To calculate the data sources and coverage quality score for a variable, each data source is rated and multiplied by the proportion it contributes to the total output.

The rating for a valid census response is defined as 1.00. Ratings for other sources are the best estimates available of their quality relative to a census response. Each source that contributes to the output for that variable is then multiplied by the proportion it contributes to the total output. The total score then determines the metric rating according to the following range:

  • 0.98–1.00 = very high
  • 0.95–<0.98 = high
  • 0.90–<0.95 = moderate
  • 0.75–<0.90 = poor
  • <0.75 = very poor.

Usual residence five years ago address
The use of 2018 Census data, alongside the introduction of admin data to fill in missing information for usual residence five years ago address, resulted in a score of 0.95, leading to the quality rating of high.

Data sources and coverage rating calculation for usual residence five years ago address data, census usually resident population count, 2023 Census
Source for usual residence five years ago address data Rating Percent Score contribution
2018 Census 1.00 71.70 0.72
Admin data 0.90 26.20 0.24
Deterministic derivation 1.00 0.19 <0.01
No information 0.00 1.92 0.00
Total 100.00 0.95
Note: Due to rounding, individual figures may not always sum to the stated total(s) or score contributions.

Usual residence five years ago indicator
The usual residence five years ago indicator variable has a score of 0.97, leading to the quality rating of high.

Data sources and coverage rating calculation for usual residence five years ago indicator data, census usually resident population count, 2023 Census
Source for usual residence address data Source for usual residence five years ago address data Rating Percent Score contribution
2023 Census response Historical census 1.00 66.84 0.67
2023 Census response Admin data 0.95 20.92 0.20
2023 Census response Deterministic derivation 1.00 0.18 <0.01
2023 Census response No information 0.50 1.75 0.01
Admin data Historical census 0.95 4.74 0.05
Admin data Admin data 0.90 5.22 0.05
Admin data No information 0.45 0.02 <0.01
CANCEIS(1) nearest neighbour imputation Historical census 0.80 0.11 <0.01
CANCEIS nearest neighbour imputation Admin data 0.75 0.06 <0.01
CANCEIS nearest neighbour imputation Deterministic derivation 0.80 0.01 <0.01
CANCEIS nearest neighbour imputation No information 0.30 0.15 <0.01
Total 100.00 0.97
1. CANCEIS = imputation based on CANadian Census Edit and Imputation System
Note: Due to rounding, individual figures may not always sum to stated total(s) or score contributions.

Consistency and coherence:

  • Usual residence five years ago address: High quality
  • Usual residence five years ago indicator: High quality

Usual residence five years ago address and indicator data are consistent with expectations across nearly all consistency checks, with some minor variation from expectations that are in line with real-world change and incorporation of other sources of data.

The use of admin data has improved the quality of the 2023 Census usual residence five years ago data.

en-NZ
Recommendations for use and further information

Usual residence five years ago data can be used in a comparable manner to the 2018 and 2013 Censuses.

There has been a change in methodology for usual residence five years ago in the 2023 Census, where admin data was introduced for this variable. When comparing time series for usual residence five years ago, it is recommended to use proportion of total stated (excluding residuals), due to the decrease in missing and residual responses in the 2023 Census compared with the 2018 Census.

Data users are recommended to familiarise themselves with the changes in methodology between each census.

Data users should also be aware that:

  • new classifications for the summary variables, regional council and territorial authority and local board (usual residence five year ago summary RC and usual residence five year ago summary TALB)were introduced in 2023. These summary variables are unable to be compared to the 2018 summary variable.
en-NZ
Information by variables from previous censuses

To assess how this concept aligns with the variables from the previous census, use the links:

Contact our Information centre for further information about using this concept.

en-NZ

Information

History

View Full History
Revision Date Responsibility Rationale
36 26/09/2024 10:00:58 AM