Stats NZ has a new website.

For new releases go to

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Discussion of review findings

This chapter draws together themes that emerged from the review of the statistical standard and classification for iwi research and consultation. Below we present a summary of the findings in response to each of the three review aims. These findings informed a range of options for revising the standard, which were shown to the working group for feedback.

Aim 1: Evaluate whether the current concept and definition remain useful for measuring iwi and Māori identity groups

The existing standard defines iwi as the “economic and political unit of traditional Māori descent and kinship-based hierarchy of waka (founding canoe), iwi (tribe), hapū (sub-tribe), and whānau (family)” (Statistics NZ, nd.b). At the time the statistical standard and classification for iwi was developed, iwi was considered the most appropriate unit for statistical measurement.

Statistical standards and classifications guide the collection of data across the OSS (Statistics NZ, nd.a). Currently, however, there is no agreement on how to reflect statistics for and about Māori in the New Zealand data ecosystem. The concepts and definitions used in the current statistical standard have received ongoing criticism for not reflecting contemporary New Zealand society because evolving and emerging Māori groups (including emerging iwi and non-kinship groups) are not captured.

A revised statistical standard and classification might better reflect a personal sense of belonging for iwi and Māori identity groupings. Exploring whether the current concept and definition of iwi remains relevant, or whether it needs to change, is important for enabling the provision of meaningful and useful data, for both the Crown and Māori.

Hapū and marae are an important part of identity

 Iwi affiliation is most commonly used to identify Māori groups. Yet iwi affiliation is merely one part of the larger whole that is Māori identity (Statistics NZ, 2016b). For many Māori, the concepts of iwi, hapū, and marae are fluid expressions of identity (Statistics NZ, 2016b). Findings from submissions (Statistics NZ, 2016c) and from academic literature (Statistics NZ, 2016b) show connections to iwi, hapū, marae, whānau, and location are important to Māori and help to provide a wider understanding of whakapapa.

Hapū: The current classification for iwi allows for the collection of hapū. However, hapū responses are coded to an iwi where possible, and are not output as a separate category. The value of collecting hapū as a distinct and separate concept, in addition to iwi, was emphasised through the review process (Statistics NZ, 2015; Statistics NZ, 2016c). Hapū received wide-ranging support in submissions and was the most frequently requested identity marker (Statistics NZ, 2016c). Government agencies are also in favour of collecting hapū information (Statistics NZ, 2015).

Marae: Some submitters consider marae to be an important identity marker for Māori (Statistics NZ, 2016c). Similarly, academic literature shows marae is an important physical marker, connecting a group of related people (Statistics NZ, 2016b).

Although hapū and marae are an important part of Māori identity, the working group expressed concern with Statistics NZ collecting this data. The feasibility of collecting hapū and marae was questioned. In addition, there were concerns relating to data access, confidentiality, and ownership. The benefits and costs of collecting hapū and marae are discussed in detail under Aim 2.

See Confidentiality, accessibility, and ownership.

Non-kinship cultural affiliation groups

The relevance of the current standard for understanding contemporary New Zealand society came up as a theme, with discussions focusing on the evolving nature of Māori groupings – particularly the emergence of non-kinship cultural affiliation groups.

For some Māori, non-kinship groups form an important part of belonging and identity (Statistics NZ, 2016c). A literature review of Māori groupings indicates urban Māori – living away from their tribal lands and other group members – have adapted to build non-traditional tribes based on residential location rather than whakapapa (Statistics NZ, 2016b). While urban Māori may feel a connection to their ancestors, they may affiliate more with physical markers (eg an urban marae, church, or club) to fulfil their need for collective belonging (Statistics NZ, 2016b).

We need to better understand and capture non-affiliated Māori. Government agencies identified urban Māori authorities as important, particularly in representing urban Māori interests (Statistics NZ, 2015). Government agencies have criticised the statistical information available, for its limited flexibility and for not reflecting the range of Māori groups that government works with (Statistics NZ, 2015). Some submitters also identified information on non-kinship groups as important (Statistics NZ, 2016c).

The ability to identify non-kinship groups is important for understanding and providing information for Māori and the Crown in relation to aspects such as economic development, social well-being, and housing. Excluding non-kinship groups may hinder the ability to inform policymaking and funding decisions relating to these areas. There was a strong feeling among government agencies that “official statistics need to be more responsive to the requirements of emerging Māori entities and groups” (Statistics NZ, 2015). If the data is solely about a traditional non-urban, marae-based population, Crown responsiveness to non-kinship and emerging Māori groups is limited.

The working group expressed concern with integrating the concepts of kinship and non-kinship groups within a statistical standard (eg a Māori grouping standard), noting fundamental differences between these groups. For example, kinship groups are based on whakapapa and genealogical connection, whereas Māori can choose to belong, or not, to a non-kinship group. Non-kinship groups are more about helping Māori, rather than identity. There was concern that a Māori groupings standard might define Māori identity, because of its influence on the OSS.

The working group identified the fluidity of non-kinship groups as concerning, with the existence of many of these groups dependent on funding and resourcing. Non-kinship groups are more likely to come and go over time. The working group questioned the ability to structure a classification around such a fluid concept.

It is evident there are diverging perspectives regarding the integration of kinship and non-kinship groups within a standard or classification. It is important to consider the primary purpose of the standard, its scope, and what the options are for measuring kinship and non-kinship groups. The purpose of a standard is to provide guidelines for collecting statistical information (Statistics NZ, nd.a), and government agencies and submitters have expressed the need for data on non-kinship groups. At the same time, there is strong opposition to integrating kinship and non-kinship concepts. As part of the decision-making process, we need to consider the broader implications for measuring Māori identity and the need to be responsive to Māori data needs.

Statistical versus legislative purposes

The current classification for iwi was introduced in 1991, in part to be responsive to Māori and in part to assist with the provision of information in relation to Treaty and Fisheries settlements. At that time, the limited information about iwi was a major obstacle to Māori development. The iwi classification was included in the 1991 Census of Population and Dwellings to aid the Iwi Transition Authority to determine groups to negotiate with and to determine the scope of settlements.

New Zealand is now moving towards a post-Treaty settlement environment, with many historical claims settled or in the process of settlement (Statistics NZ, 2015). As progress towards a post-Treaty settlement environment continues, data needs are changing and evolving (Statistics NZ, 2015; Statistics NZ, 2016c).

Both Māori and the Crown have a greater need for official statistics to inform decision-making and policymaking. As such, the rationale in the current standard, which emphasises Waitangi Tribunal settlements and resource allocation, is outdated. The purpose of the revised statistical standard for iwi / Māori groups needs to move beyond traditional data requirements and better reflect the statistical needs of contemporary New Zealand society.


Aim 2: Explore whether different types of information (eg hapū, marae, or location), in addition to iwi, could improve the quality and use of iwi statistics

It is crucial, and expected, that Statistics NZ collects information relevant to Māori development, to support decision-making and policymaking.

Statistics about and for Māori are required

Traditionally, iwi statistics have been driven by government’s interest in understanding the Māori population relative to the New Zealand population, and in the context of engaging with Māori in Treaty settlement and post-settlement development (Statistics NZ, 2014a). These data needs have resulted in a range of statistics about Māori (Statistics NZ, 2014a).

With many historical Treaty claims now settled or in the process of settlement, there is an increasing requirement for statistics for Māori (Statistics NZ, 2014a). Post-settlement iwi are now able to play an important role in improving the well-being of their people. To plan and implement services effectively, these groups need high-quality and detailed statistical information.

The review process shows there is agreement that statistics both about Māori and for Māori are important, particularly as the Crown and Māori increasingly engage and collaborate for mutual benefit (Statistics NZ, 2015; Statistics NZ, 2016c). The effective planning and provision of services requires a wide range of statistics. Submitters cited education (with particular reference to kōhanga reo), health, location, economic and social measures, and demographics as important current and future data needs (Statistics NZ, 2016c).

The statistical standard was originally designed to enable the provision of iwi data relating to population, social, and household measures. However, feedback from submissions and from government agencies indicates iwi data is now also required for economic purposes (Statistics NZ, 2015; Statistics NZ, 2016c). It is worth considering whether two types of standards are needed to capture these different types of information.

Collection of rohe/location

Currently the collection of rohe alongside iwi occurs in Statistics NZ’s questionnaires, but this concept does not represent true rohe boundaries. In the iwi classification, rohe is similar to regional council boundaries. As such, Statistics NZ’s interpretation of rohe does not necessarily align with the physical markers Māori identify with. These differences have led to confusion and misunderstanding among respondents. A few submitters expressed dissatisfaction with these differences (Statistics NZ, 2016c).

Rohe is not directly comparable to other geospatially enabled boundaries or Te Puni Kōkiri boundaries (Te Puni Kōkiri, nd), limiting comparability across regions. Given the limited utility rohe information currently provides, the working group recommended removing rohe from our questionnaires. However, a number of iwi groups share the same name across different locations. In these cases, rohe or other location data is integral in enabling the coding of iwi to the correct area. The utility of rohe or location data needs to be balanced against the relevancy of the geographic measures for Māori.

Collection of hapū and marae

As part of this review, the option of creating additional response-based category sets, to reflect contemporary New Zealand society, are being explored. Feedback from government agencies (Statistics NZ, 2015) and submissions (Statistics NZ, 2016c) indicates a requirement for hapū and marae data. These identity markers may increase knowledge and understanding of whakapapa connections, improve data quality, and help with analyses.

The usefulness of iwi data is reliant on the accuracy of the data, in terms of both collection and coding. Data from the 2013 Census shows over 10,000 responses to the iwi question were unable to be coded.

Collecting hapū- and marae-level information, in addition to iwi, might improve coding of iwi affiliation, and subsequently the quality of data. Given all marae are geographically located, the addresses could be used to create an administrative dataset or be based off a property frame/address register. The inclusion of additional response fields (eg hapū and marae) would also allow those with limited knowledge of their whakapapa to complete as many fields as they can with the knowledge they have.

There is an opposing view that hapū and marae will not necessarily improve the accuracy and quality of the data. The ability to code hapū- and marae-level information is constrained by current survey processing systems. The working group also notes that coding data might be difficult in cases where there are multiple hapū and marae with similar names.

The working group was concerned that hapū and marae data might be difficult to collect from Māori with limited knowledge of their whakapapa. Findings from Te Kupenga – a survey of Māori well-being – show 55 percent of respondents know their hapū and 71 percent know their marae (Statistics NZ, 2014b). In comparison, 89 percent of respondents know their iwi. Responding to questions on hapū or marae might require additional support and guidance from family or kaumātua. There was also concern that the inclusion of additional hapū and/or marae questions will contribute to respondent burden, without necessarily any additional benefit.

Confidentiality, accessibility, and ownership

Iwi and government agencies have indicated that access to iwi data cross-tabulated with other variables is important for informing decision-making and policymaking. However, there are currently constraints around the ability to access cross-tabulated iwi data. Confidentiality, accessibility, and ownership emerged as dominant themes through the review process. Most of these findings are, however, outside the scope of the review of the statistical standard.


The working group was concerned that if Statistics NZ collect hapū and marae data it creates the unrealistic expectation that the data is available for release. They note confidentiality thresholds currently restrict output of some iwi data, particularly at demographic and geographic levels. Consequently, the working group questioned whether hapū- and marae-level data could be disseminated beyond basic classification counts. There is a strong feeling that if government are collecting granular-level data, then disseminating the data back to iwi and Māori groups should be a requirement.

Changing the confidentiality rules is outside the scope of the review of the standard and classification. However, the Statistics Act 1975 is currently under review, which presents an opportunity to highlight some of the points raised by users of iwi statistics. Work to address the confidentiality rules and to explore options for releasing lower-level data to iwi is in progress.

Accessibility and capability

Stakeholders observed that access to data is constrained by capability and capacity (Statistics NZ, 2015; Statistics NZ, 2016c). While there is acknowledgement that iwi statistics are currently available, there is converging evidence to show that these statistics are not widely known about, nor understood, especially by Māori (Statistics NZ, 2015; Statistics NZ, 2016c). Feedback from submitters indicates the ability to know what data is available and how to access it is challenging and can be costly (Statistics NZ, 2016c). The working group also noted that capability and capacity are more limited at hapū- and marae-level.

These issues extend beyond the scope of the review of the standard. However, Statistics NZ can work to create greater awareness and increase transparency of the information held about Māori and for Māori.


The review has raised questions about who owns, and who should have access to, iwi and Māori data (Statistics NZ, 2016c). The working group felt that if the Crown collects low-level information (eg hapū or marae); they are receiving more information than necessary to improve coding. Feedback from a few submitters also highlights that access to information via an Official Information Act request was of concern, as this allows third parties to access iwi data without iwi consent (Statistics NZ, 2016c).

Currently iwi and Māori data fall under the same jurisdiction as other data. Māori are, however, seeking a platform with the Crown to have in-depth conversations regarding the treatment of data as a taonga. Statistics NZ may wish to consider whether iwi/Māori data should be treated uniquely to reflect obligations under the Treaty, recognising the uniqueness of Māori as tangata whenua.


Government agencies are engaging with multiple Māori groups and entities (Statistics New Zealand, 2015). However, submissions (Statistics NZ, 2016c) and interviews with government agencies (Statistics NZ, 2015) show there is limited consistency in the concepts, definitions, and lists used to collect iwi and/or Māori grouping information. We also identified that data is aggregated across the government sector in different ways. Stakeholders indicated that the ability to develop accurate time-series information would increase the utility of iwi data (Statistics NZ, 2015; Statistics NZ, 2016c).

If the revised statistical standard is widened to support the use of iwi Māori statistics from an economic sense, then developing a standard that takes into account the multitude of Māori entities the government is engaging with is important. This would enable better integration of data now and in the future (Statistics NZ, 2015; Statistics NZ, 2016c). In addition, it would be important to work collaboratively with representatives across government to develop and promote the revised standard.

Māori descent

In some surveys, collection of iwi affiliation follows a question on Māori descent. There are, however, respondents who record an iwi after leaving the Māori descent question blank, or responding “no” or “don’t know” to the question. At present, we do not output iwi data in these instances.

A secondary objective of the review was to explore whether iwi outputs should include all those who report an iwi (Statistics NZ, 2016a). We recommended all those who report an iwi be counted, even those who report “no” or “don’t know” to the Māori descent question. However, potential effects to time series warrant consideration.


Aim 3: How to decide which groups, and which types of groups, are included in our classification(s)

The final aim of the review centres on understanding how to, and who should, decide on the groups to include in the classification. This aim was in part driven by ongoing dissatisfaction with the current iwi classification, as it prevents the inclusion of new groups.


A criticism of the current classification is that it does not reflect the multitude of Māori entities government agencies are engaging with (Statistics NZ, 2015). Feedback from submissions indicates the current inclusion criteria are too rigid, and most submitters are seeking change (Statistics NZ, 2016c). The current criteria were criticised for not allowing applicant groups (eg hapū acquiring or petitioning for iwi status) to be included (Statistics NZ, 2016c). For groups not currently reflected in the iwi classification, the application and usability of iwi data is limited.

The current criteria to classify iwi take into account:

  • whether the group has been separately categorised in earlier iwi or tribal classifications 
  • whether the group has been identified by respondents in previous surveys or censuses 
  • whether there is a history of the group operating as a separate iwi in a business or resource management capacity, with legal and/or administrative recognition as such 
  • whether historical and genealogical tradition identify the group as distinctive
  • whether the group (as hapū of a larger iwi) is moving to acquire or petition for iwi status (Statistics NZ, nd.b).

However, the operationalisation of these criteria are constrained by an additional criterion stating, “Statistics NZ will consult with the larger iwi to obtain its view on the group's position in relation to the five criteria listed above” (Statistics NZ, nd.b). Larger iwi therefore have the capacity to veto the inclusion of applicant groups into the classification.

In making a decision about whether the additional criterion should remain or be removed, the costs and benefits of the decision should be considered. Removing the criterion may result in the potential alienation of some larger iwi groups, however, continuing to include it limits Statistics NZ’s ability to be responsive to Māori and to reflect contemporary New Zealand society.

If the criteria do change (and as a result, the scope of the population), we need to consider the potential implications to time-series data. Raising awareness and educating users of potential changes to times series will be important.

Groups not in the classification

Statistics NZ has received ongoing petitioning from applicant groups who want to be recognised and included in the iwi classification. Often driving this is applicant groups’ desire to have access to their own data. Currently, iwi included in the classification have access to their data, whereas those not in the classification either do not have access to their data or can only access their data at a cost.

There is also a perception the census guide notes reflect a definitive (legal) list of iwi, which they do not. This perception indicates there is a need to raise awareness that the primary purpose of a standard is for producing official statistics and that the guide notes are there to assist respondents fill in their forms. This statistical focus should be explicit in the rationale and messaging of the statistical standard.

Standard and classification review cycle

Minor reviews of statistical standards and classifications generally occur on a four- to five-yearly cycle, and major reviews on a 10-yearly cycle (Statistics NZ, nd.c). In the context of the fluid nature of both kinship and non-kinship Māori groups (Statistics NZ, 2016b), the cyclical review process adds constraints to maintaining a relevant classification. In addition, the need to fit within the census development timelines affects the ability to deliver change.

The classification criteria need to be less restrictive to enable the ongoing inclusion of groups. Statistics NZ’s new tool, Ariā, will enable dynamic updating of concepts and category sets. Ariā will also enable the grouping of data to support different views. For example, iwi and other users can create and maintain their own output data views for Fisheries, Treaty settlements, boundaries, or for other purposes.

Statistics NZ’s role

Determining which groups to include in the revised iwi classification is a contentious issue. Statistics NZ’s role (implied or otherwise) in deciding which groups to include or exclude from the classification has been questioned in the literature (Statistics NZ, 2016b), by government agency representatives (Statistics NZ, 2015), and by submitters (Statistics NZ, 2016c). Increasing the transparency of the processes used to develop classification criteria should be considered.

If Statistics NZ remains in the position of ‘decision maker’, tensions and dissatisfaction are likely to be ongoing (Statistics NZ, 2015). Statistics NZ is exploring the option of relinquishing the custodianship of other standards to appropriate agencies or subject matter experts. For example, the Ministry of Business, Innovation and Employment is now the custodian of the research standard. In the case of the standard and classification of iwi, there is no one agency with the authority or expertise to take responsibility for cross-data ecosystem governance. Statistics NZ is therefore likely to retain custodianship of this standard.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+