Māori And Iwi Population Concepts In 2023 Census

This paper gives users of 2023 Census data an overview of the variables that measure Māori and iwi populations, including information on the questions asked in the census, the alternative data sources used to fill in missing information, and the categories that are output.

The differences between Māori descent and Māori ethnicity are discussed, as well as when to use each variable, and the quality of the data.

Ethnicity, Māori descent, and iwi affiliation data from the 2023 Census can be compared with published data from previous censuses. The definitions of these concepts have not changed. However, data users should be aware that the introduction of alternative data sources means that the way these concepts are measured has changed.

The paper includes points to consider when doing time series comparisons across censuses, including a reminder that Stats NZ did not release official statistical counts of iwi from the 2018 Census. Therefore published time series tables for the 2023 Census will only include iwi affiliation data from the 2023 and 2013 Censuses.

Ngā ariā taupori Māori, taupori iwi hoki, Tatauranga 2023

Ko tā tēnei pepa he hoatu māramatanga ki ngā kaiwhakamahi o Tatauranga 2023 mō ngā tāupe e inea ai ngā taupori Māori, iwi hoki, tae atu ki ētehi mōhiotanga mō ngā pātai ka tāpaetia ki te tangata i roto i te tatauranga, me ngā puna raraunga tāpiri hoki e whakamahia ana hei whakakī i ngā āputa mōhiotanga, me ngā kāwai ka noho hei putanga.

Ka kōrerotia ngā rerekētanga o te uri Māori, me te momo iwi Māori, ā, hei āhea whakamahia ai tētahi, tētahi, me te kounga o ngā raraunga.

Ka taea te whakatairite ngā raraunga momo iwi, uri Māori, piringa ā-iwi hoki mai i te 2023 ki ngā raraunga mai i ō mua tatauranga kua oti te whakaputa. Kāore anō ngā tautuhitanga o ēnei ariā kia panonitia. Ahakoa tērā, kia mārama ngā kaiwhakamahi nā te urunga mai o ētahi puna raraunga tāpiri kua panonitia ngā tikanga ine i ēnei ariā.

Kei roto i te pepa ētehi mea hei whiriwhiri ina mahia he whakatairite houanga puta noa i ngā tatauranga, me tētahi tauanga ā-iwi mai i te Tatauranga 2018. Nā reira heoi anō nei ngā ripanga raraunga houanga ka puta mō Tatauranga 2023 ko ngā raraunga piringa ā-iwi mai i ngā Tatauranga o 2023 me 2013.

On this page

What data was collected and what is being output?
Overview of concepts and variables
Difference between Māori descent and Māori ethnicity
When to use each variable?
Quality ratings
Time series comparisons
Further data

What data was collected and what is being output?

The 2023 Census collected information relating to many concepts, including ethnicity, Māori descent, and iwi affiliation. Māori and iwi population information was collected through three questions on the individual form:

• Ethnicity - all respondents were asked "Which ethnic group do you belong to?".
• Māori descent - all respondents who were usually resident in New Zealand at the time of the census were asked "Are you descended from Māori (that is, did you have a Māori birth parent, grandparent or great-grandparent, etc)?".
• Iwi affiliation - all respondents who were usually resident in New Zealand at the time of the census, and who responded that they are of Māori descent or that they don't know if they are of Māori descent, were asked "Do you know the name(s) of your iwi (tribe or tribes)?". If they selected "Yes", they were then asked to "give the name(s) and region(s)/rohe of your iwi".

Design of forms for the 2023 Census summarises how Stats NZ developed these three questions for the 2023 Census.

The 2023 Census used a combined model, where data from census responses and alternative data sources was combined to produce the highest quality data possible about the populations living in Aotearoa New Zealand.

Using a combined census model for the 2023 Census has more information.

When a census form was not completed or when a response was not provided to any of the three questions above, the following alternative data sources were used:

From this data, several variables were produced that measure Māori and iwi populations, including:

• Ethnicity (total responses)
• Ethnicity (single/combination)
• Māori ethnicity indicator
• Māori descent (output)
• Māori descent (electoral)
• Iwi affiliation

The following sections describe these concepts and variables in more detail.

Overview of concepts and variables

Ethnicity

Ethnicity is the ethnic group or groups that people identify with or feel they belong to. Ethnicity is a measure of cultural affiliation, as opposed to race, ancestry, nationality, or citizenship. Ethnicity is self-perceived, and people can affiliate with more than one ethnic group.

An ethnic group is made up of people who have some, or all, of the following characteristics:

• a common proper name
• one or more elements of common culture, which need not be specified but may include religion, customs, or language
• unique community of interests, feelings, and actions
• a shared sense of common origins or ancestry, and/or
• a common geographic origin.

Ethnicity can be output in a variety of ways, including total response ethnicity, ethnic group indicators, and single and combination ethnicity.

Alternative data sources of historical census data, admin data, and statistical imputation were used for missing census responses and responses that could not be classified or did not provide the type of information asked for. This ensured that everyone in the 2023 census usually resident population count, and the 2023 census night population count, was coded to one or more ethnic groups (ie, no one was coded to 'Not elsewhere included').

Ethnicity (total responses)

Total response ethnicity includes all people who stated each ethnic group, whether as their only ethnic group or as one of several ethnic groups. Where a person reported more than one ethnic group, they have been counted in each applicable group. Therefore the total number of responses in a table will be greater than the number of respondents.

The ethnicity (total responses) variable uses a 4-level hierarchical classification. Level 1 categories are listed below:

• 1  European
  2 Māori
  3 Pacific Peoples
  4 Asian
  5 Middle Eastern/Latin American/African
  6 Other Ethnicity
  9 Not elsewhere included

Ethnicity (single/combination)

Total response ethnicity is also output as single/combination variables. The single-combination response classifications distinguish between those who responded with a single ethnicity, and those who responded with multiple ethnicities.

Respondents are counted once in the single or combination category that applies to them. For example, the Ethnicity (detailed single/combination) variable has 45 output categories in total.

Listed below are the 19 categories that are used for people who identify with Māori ethnicity:

• 112 Māori only
   211 European/Māori
  216  Māori/Pacific Peoples
  217  Māori/Asian
  218  Māori/Middle Eastern/Latin American/African
  219  Māori/Other Ethnicity
  311  European/Māori/Pacific Peoples
  312  European/Māori/Asian
  313  European/Māori/Middle Eastern/Latin American/African
  314  European/Māori/Other Ethnicity
  341   Māori/Pacific Peoples/Asian
  342 Māori/Pacific Peoples/Middle Eastern/Latin American/African
  343 Māori/Asian/Middle Eastern/Latin American/African
  344 Māori/Asian/Other Ethnicity
  345 Māori/Middle Eastern/Latin American/African/Other Ethnicity
  351   Pacific Peoples/Māori/Other Ethnicity
  411   Four ethnic groups
  511     Five ethnic groups
  611     Six ethnic groups

Māori ethnicity indicator

For each level 1 ethnic group in the total response ethnicity classification, an indicator is derived.

The Māori ethnicity indicator counts all people who identify with the Māori ethnic group. Individuals are counted once: either they only identify with the Māori ethnic group, they identify with the Māori ethnic group and at least one other ethnic group, or they do not identify with the Māori ethnic group.

The Māori ethnicity indicator variable has four output categories:

00 Non-Māori
01    Māori only
02 Māori and at least one other ethnic group
99 Not elsewhere included

Ethnicity - 2023 Census: Information by concept provides detailed information about the ethnicity concept and associated variables.

Editing, data sources, and imputation in the 2023 Census provides detailed information about data sources and methodology used for ethnicity in the 2023 Census.

Māori descent

A person is of Māori descent if they are descended from Māori. Māori descent is based on a genealogical or biological concept. For the purposes of the Māori descent classification, having Cook Island Māori ancestry is not classified as being of Māori descent.

Māori descent (output)

The Māori descent variable (also referred to as the Māori descent indicator) has four output categories:

01     Māori descent
02 No Māori descent
04 Don't know
99 Not elsewhere included

"Don't know" is an important category in the census output Māori descent variable. It is a response option on the census form that recognises that not all respondents will know if they are of Māori descent or not. It is not treated as a residual category.

Alternative data sources of historical census data, admin data, parental data, deterministic derivation, and statistical imputation were used for missing census responses and responses that could not be classified or did not provide the type of information asked for. This ensured that no one in the 2023 census usually resident population count was coded to 'Not elsewhere included'.

Māori descent - 2023 Census: Information by concept provides detailed information about the Māori descent concept and variable.

Data sources and imputation for Māori descent in the 2023 Census provides detailed information about data sources and methodology used for the Māori descent (output) variable in the 2023 Census.

Māori descent (electoral)

The Māori descent (electoral) variable has two output categories:

01 Māori descent
02   No Māori descent

Under the Electoral Act 1993, census counts are used to calculate electoral populations and determine the number of electorates.

Following legal advice, the Government Statistician decided for the 1996 and following censuses that those not answering Yes or No to the question on Māori descent should be allocated to one of these two categories for electoral purposes (previously those not answering Yes or No were all effectively treated as answering No). This ensures the Māori descent population is appropriately represented at the geographic levels needed for electoral purposes.

The methodology for deriving the Māori descent (electoral) variable has changed over time. For 2023, alternative data sources of historical census data, admin data, parental data, deterministic derivation, and statistical imputation were used to ensure everyone in the census usually resident population count was coded to 'Māori descent' or 'No Māori descent'.

Māori descent (electoral) is used for electoral purposes only. It is not used to determine the Māori descent census usually resident population count, which conceptually recognises that some people do not know if they are of Māori descent.

Māori descent electoral - 2023 Census: Information by concept provides detailed information about the Māori descent (electoral) concept and variable.

Data sources and imputation for Māori descent in the 2023 Census provides detailed information about the methodology for the Māori descent (electoral) variable in the 2023 Census.

The mathematics of electorate allocation in New Zealand based on the outcome of the 2023 Census and Māori Electoral Option outlines how the number of electorates in New Zealand is calculated, as specified in the Electoral Act 1993.

Iwi affiliation

An iwi (a Māori tribe) is one of the largest kinship groupings and is generally made up of several hapū that are all descended from a common ancestor. Hapū are clusters of whānau (families) where the whānau is usually an extended family grouping consisting of children, parents, often grandparents, and other closely related kin.

For statistical purposes, Stats NZ defines iwi as a whakapapa-based kinship grouping that generally has several hapū, one or more active marae, and a recognised structure that represents the interests of the iwi, such as a rōpū whakahaere, committee, or board.

Iwi affiliation

The census measures self-identified iwi affiliation(s), as opposed to tribal registers, which require proof of whakapapa. When answering the iwi question, people provide a broad range of responses. Some of these responses belong to other concepts, for example confederations or waka. For this reason, the iwi affiliation classification has a range of categories to capture an individual's responses.

The iwi affiliation variable uses a 2-level hierarchical classification.

Level 1 of the classification lists 18 broad categories. These include:

• 12 categories that group iwi into geographical regions for statistical purposes
• a category for confederations and waka, where the iwi is not named
• a category for when iwi are named but the region is not known
• a category for when hapū are affiliated to more than one iwi
• a category for when the region is known but the iwi is not named
• a category for people who 'Don't know' their iwi
• a category for 'Not elsewhere included' categories.

Level 2 of the classification lists iwi and iwi-related categories:

• within the 12 geographical regions in level 1 there are 139 'Fully coded iwi' categories
• within the four level 1 categories for 'Confederations and waka', 'iwi named, region not known', 'Hapū affiliated to more than one iwi', and 'Region known, iwi not named' there are a total of 45 'Partially coded iwi' categories
• there is a category for 'Don't know'
• the 'Not elsewhere included' category contains the residual categories 'Refused to answer', 'Response unidentifiable', 'Response outside scope', and 'Not stated'.

Multiple responses could be provided to the iwi affiliation question. People reporting more than one iwi affiliation are counted in each category they stated. Therefore the total number of responses in a table will be greater than the number of respondents.

Data is only published at level 2 of the classification. The geographic regions provide structure to the long list of iwi and are not intended for reporting purposes.

Alternative data sources for iwi affiliation were introduced for the first time in the 2023 Census. Alternative data sources of parental data, historical census data, and admin data were used for missing and residual census responses and responses that could not be classified or did not provide the type of information asked for.

Statistical imputation was not used for iwi affiliation. As a result, 2.5 percent of the 2023 Māori descent census usually resident population count were coded to a residual category.

Iwi affiliation - 2023 Census: Information by concept provides detailed information about the iwi affiliation concept and variable.

Data sources and methodology for iwi affiliation in the 2023 Census provides detailed information about data sources and methodology used for the iwi affiliation variable in the 2023 Census.

New Zealand literature review of Māori groupings discusses the challenges faced when defining tribal groupings such as iwi.

Statistical standard for iwi and iwi related groups V1.0.1 provides advice on collecting, processing, and outputting iwi affiliation information.

Difference between Māori descent and Māori ethnicity

Māori descent refers to whether a person is biologically descended from Māori, while Māori ethnicity refers to a person's self-identified cultural affiliation to the Māori ethnic group.

Some people of Māori descent do not identify with the Māori ethnic group, and some people who identify with the Māori ethnic group are not of Māori descent.

The 2023 census usually resident population count was 4,993,923. Out of these people, 978,246 were of Māori descent, and 887,493 identified with the Māori ethnic group. There were 857,652 people who were both of Māori descent and identified with the Māori ethnic group.

One of the reasons why understanding the difference between Māori ethnicity and Māori descent is important is because the iwi affiliation data that Stats NZ publishes is only based on people of Māori descent.

When to use each variable?

Ethnicity (total responses) data should be used when wanting to analyse results for people who identify with ethnic groups in the classification: for example, when data users want to look at the total number of people who identify with the Māori ethnic group.

Ethnicity (single/combination) data should be used to analyse results for people who affiliate with a single ethnic group, and those who affiliate with different combinations of multiple ethnic groups: for example, when data users want to look at people who identify with the Māori ethnic group and other specific ethnic groups.

Māori ethnicity indicator data should be used when wanting to analyse results for people who only identify with the Māori ethnic group, people who identify with the Māori ethnic group and at least one other ethnic group, or people who do not identify with the Māori ethnic group.

Māori descent (output) data should be used when wanting to analyse results for people of Māori descent, people not of Māori descent, and people who do not know if they are of Māori descent. The Māori descent (electoral) variable is only intended for electoral purposes. All other users interested in outcomes for people of Māori descent should use the Māori descent (output) variable. This is the default variable used for tables in Aotearoa Data Explorer and other 2023 Census outputs.

Māori descent (electoral) data is used by Stats NZ to calculate electoral populations and to determine the number of electorates.

Iwi affiliation data should be used when wanting to analyse results for people of Māori descent who affiliate with the iwi in the classification.

Quality ratings

Stats NZ produced quality ratings for all key output variables from the 2023 Census.

Data quality assurance in the 2023 Census provides more detail.

The quality ratings for the key output variables covered in this paper are:

• Ethnicity - High
• Māori ethnicity indicator - High (metric 1)
• Māori descent (output) - Very high
• Māori descent (electoral) - Very high
• Iwi affiliation - Moderate

Ethnicity and Māori descent variables received quality ratings of high or very high. Significant reasons for this include the use of alternative data sources, and the availability of, and consistency with, comparative data.

The iwi affiliation variable received a quality rating of moderate. When determining this rating it was noted that iwi counts are higher, sometimes considerably higher, than in previous censuses. A number of factors have contributed to this, such as changes in how the variable has been collected, the incorporation of alternative sources of data, improvements in response rates for Māori in the 2023 Census, more people with multiple iwi affiliations, and potentially an increased awareness of whakapapa. Also, while alternative data sources are now being used to fill in some missing and residual responses, the accuracy ratings of those sources are lower than data collected directly from current census responses, and there is still a proportion of people with no iwi affiliation data (2.5 percent).

To understand quality aspects for each variable in this paper, and to see the quality ratings for other variables in the 2023 Census, please see their respective 'information by concept' publication, accessible through Data quality ratings for 2023 Census variables.

Data sources for 2023 Census provides a summary of the proportion of alternative data sources used for other variables in the 2023 Census.

Time series comparisons

Ethnicity, Māori descent, and iwi affiliation data from the 2023 Census can be compared with published data from previous censuses. This is because the definitions of these concepts have not changed. However, data users should be aware that the way these concepts are measured has changed.

In previous censuses (ie, 2013 and prior) all the data came from census form responses. This meant that all the data was collected around a single point in time. By comparison, the 2018 and 2023 censuses were a combined model, where alternative data sources were used to fill in missing data. As a result, not all data was collected around the same point in time, and not all data was collected in the same context.

Stats NZ is confident in the quality of counts for ethnicity, Māori descent, and iwi affiliation, and that these can be compared over time. Alternative data sources are used to fill in missing data in a way that best represents what an individual would have provided on a census form. This means that, even though alternative data sources will not always be a perfect replacement for a census response, the overall quality of outputs is increased, and can be considered to represent the population at a given point in time.

Stats NZ acknowledges that the inclusion of alternative data sources may add some complexity to the interpretation of the results for some variables. For example, people may answer an iwi affiliation question differently at different points in time. In the 2013 Census, all iwi affiliation data came from census form responses, so 2013 iwi counts clearly represent iwi affiliations on census night for people of Māori descent who responded to the census.

In 2023, parental data, historical census data, and admin data were used to fill in missing and residual responses. This means the complete data includes information collected in a variety of ways. To mitigate this complexity, Stats NZ measures the quality of each data source and has used these alternative data sources in a way that aims to best reflect what would have been provided through 2023 Census forms. As a result, 2023 iwi counts are considered to be a good representation of iwi affiliations on census night.

However, this change in the methodology has had an impact on the results. For example, the number of people counted in the Māori ethnic group, and the number of people of Māori descent, increased substantially between the 2013 Census and the 2018 Census. While there was a real-world population increase during this period, the increase in count was largely due to the introduction of alternative data sources to fill in missing data for these variables in the 2018 Census, reducing the undercount that had been present in historical collections.

In a similar way, between the 2013 Census and the 2023 Census there was an increase in all iwi affiliation counts. In addition to a real-world population increase, the numbers reflect the introduction of alternative data sources for iwi affiliation, and changes to the iwi affiliation question in the online form.

Data users should take this into consideration when comparing results from different censuses. Note: Stats NZ did not release official statistical counts of iwi from the 2018 Census. This was because of the very poor quality data, primarily due to the level of missing iwi affiliation data. Published tables for the 2023 Census will only include iwi affiliation data from the 2023 and 2013 Censuses.

It is recommended that the data be treated as a whole, regardless of the data source. For example, while the iwi affiliation variable is composed of information from census forms, parental data, historical data, and admin data, Stats NZ does not recommend separating the results by data source. As a reminder, data from census responses and alternative data sources was combined to produce the highest quality data possible about the populations living in Aotearoa New Zealand.

Further data

Statistical imputation was intentionally not used as an alternative data source for the official iwi affiliation results. However, Stats NZ has created a separate variable that uses statistical imputation to fill in the remaining missing iwi affiliation responses, so that this can be used for experimental analysis. Iwi will be able to request this experimental iwi data via Te Ara Takatū: Census data for iwi and iwi-related groups, or researchers working in partnership with iwi can apply directly to the Integrated Data Infrastructure (IDI).

Enquiries

Census communications

ISBN 978-1-991307-46-0