• Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Recommendations from the 2016 review of the statistical standard for iwi

Purpose

Recommendations from the 2016 review of the statistical standard for iwi has been approved by the Government Statistician.

Statistics New Zealand, in collaboration with Māori group representatives and government agencies, is reviewing the statistical standard and classification of iwi. This report details the recommendations, which are an outcome of the review.

Aims of review

The primary aim of the review was to investigate the suitability of the existing statistical standard and classification for iwi for meeting current and future data needs of users. The focus of the review was on whether and what changes need to be made.

Feedback was sought from Māori, iwi, government, and the public on concepts, definitions, rationale, criteria, and procedures, with a specific focus on understanding:

  1. Whether the current concepts and definitions remain useful for measuring iwi and Māori identity groups.
  2. Whether different types of information (e.g., hapū, marae, or location), in addition to iwi, could improve the quality and use of iwi statistics.
  3. How to decide which groups, and which types of groups, are included in our classification(s) (list of groups).

Findings specific to each of the three review aims are presented in the discussion paper entitled Discussion of findings from the 2016 review of the statistical standard for iwi, which should be read in conjunction with this recommendations report.

Recommendations

This section presents the recommendations, which are an outcome of the review of the statistical standard and classification of iwi.

1. Purpose and rationale for the standard

Recommendation 1:
Change rationale of the standard to emphasise a statistical focus. Remove reference to Treaty settlements, Waitangi Tribunal, and resource allocation, as these are no longer relevant in a post Treaty settlement environment. Replace with rationale that reflects Treaty principles and acknowledges commitments as a Treaty partner.

1.1 The primary purpose of a statistical standard for iwi is to enable the production of statistics, both about and for Māori. We recommend making this statistical focus explicit in the purpose and rationale of the statistical standard.

1.2 The current statistical standard for iwi was developed in 1994 and reflects the Crown’s preference for engaging with Māori in Treaty of Waitangi and Fisheries claims at the time. This approach is losing relevance in a post Treaty settlement environment. Remove references to Treaty of Waitangi settlements and Waitangi Tribunal decisions (land ownership, fishing rights, resource allocation, etc.) to better reflect evolving data needs in a post Treaty settlement environment. 1

1.3 To acknowledge Crown commitments as a Treaty partner, incorporate Treaty principles (partnership, protection, and participation) in rationale. Māori are seeking a platform with the Crown to have the wider, deeper conversations regarding the treatment of data as a taonga.

2. Structure of Māori groupings classifications and standards

Recommendation 2:
a. Update the iwi standard and refresh the classification.
b. Introduce new concepts and create separate standards and classifications for each identity collection under a wider framework using a staged approach (eg stage one iwi, stage two hapū / marae, and stage three non-kinship groupings).

2.1 Discontent with the current iwi standard and classification is evident. We recommend prioritising the update of the statistical standard for iwi and a refresh of the classification.

2.2 Hapū, marae, and non-kinship groups are also important identity markers for Māori. Users require more data and statistical information on these Māori identity groupings than is currently captured.

2.3 There is opportunity to introduce a suite of standards and classifications under a wider framework and expand the underlying concepts to include hapū, marae, and/or non-kinship.

2.4 Statistics New Zealand acknowledges there are concerns with integrating kin (whakapapa) and non-kin (non-whakapapa) concepts. Creating separate standards and classifications for each identity group will make clear the Māori identity group being considered.

2.5 Users expressed concern that a standard might define, by virtue of its influence on the Official Statistics System (OSS), Māori identity. Mitigating this risk requires education and raising awareness of the purpose of a statistical standard.

2.6 Developing a framework requires further work to assess the feasibility. The implementation of Recommendation 2 requires further work to develop new standards / classifications for hapū, marae, and non-kinship data (refer to recommendations four to six). A staged approach (ie iwi as the first stage, followed by hapū, marae, and non-kinship groups) allows time to plan and budget the work.

3. Iwi classification criteria

Recommendation 3:
Replace the current iwi classification criteria with new criteria to better reflect a post Treaty settlement environment and a contemporary New Zealand society. Change the process to create a more collaborative and inclusive partnership with Māori to facilitate updating and maintenance of the classification.

3.1 The current iwi classification criteria have been criticised for being too rigid, as it does not allow applicant groups to be included easily. Changing the criteria to be more inclusive and allowing recognition of applicant iwi groups will allow Statistics New Zealand to be more responsive to existing data needs.

3.2 To allow applicant groups to be recognised more easily, changes to the criteria should include removal of the statement, “In addition, Statistics New Zealand will consult with the larger Iwi to obtain its view on the group’s position in relation to the five criteria”. A risk of removing this statement is the potential alienation of some large iwi.

3.3 Changing the criteria might affect time-series data (because of changing the scope of population), particularly for the Census of Population and Dwellings. Developing a strategy to educate users about potential effects to time-series should be considered. The risk to changes in time-series needs to be balanced against the benefits of creating an iwi classification that is more inclusive and reflective of the real world.

4. Hapū

Recommendation 4:
Further investigation is required before developing a hapū standard / classification. Such work needs to address issues around collection, coding, and output of information. Aim to include hapū in a future survey (e.g., Māori wellbeing survey or a household sample survey) and potentially partner with Māori to collect their own data.

4.1 Further work is required before developing a hapū standard and / or classification (refer to Recommendation 2). Such work needs to assess the feasibility of collecting hapū information and address the concerns raised during consultation (eg data access, respondent burden, and data quality).

4.2 There is support to include hapū as an identity marker (in addition to iwi). Hapū connections are important for Māori and provide a wider understanding of whakapapa. Users also require hapū level information.

4.3 Statistics New Zealand acknowledges the data need for hapū but also notes there are concerns with the collection of hapū information. Collecting hapū information creates the expectation that the data will be available for release. However, hapū data might not be of sufficient quality for release and confidentiality thresholds might constrain access to the data. There are also concerns that collecting additional information will increase respondent burden. So far, limited cognitive testing indicates knowledge of and ability to report hapū varies considerably between respondents. Findings from Te Kupenga show 55% of Māori know their hapū.

4.4 Hapū information might be better suited to collection by Māori.

4.5 There is insufficient evidence to show hapū information will improve the coding of iwi. The ability to code and operationalise the collection of hapū data is also currently constrained by Statistics New Zealand’s current processing systems. Statistics New Zealand should not be collecting data that will not be accurate enough to disseminate (either as data or statistics), or that does not – to a substantial extent – support improving the accuracy of the iwi data that is released.

4.6 The final decision for including hapū in a survey will occur independently of the development of the standard or classification. This is because census (and other surveys) has an independent content determination process to determine the final variables for inclusion in the questionnaire.

5. Marae

Recommendation 5:
Further investigation is required before developing a marae standard / classification. Such work needs to address issues around collection, coding, and output of information. Aim to include marae in a future survey (e.g., Māori wellbeing survey or a household sample survey) and potentially partner with Māori to collect their own data.

5.1 Further work is required before developing a marae standard and / or classification (refer to Recommendation 2). Such work needs to assess the feasibility of collecting marae information and address the concerns raised during consultation (eg data access, respondent burden, and data quality).

5.2 There is support to include marae as an identity marker (in addition to iwi). Marae connections are important for Māori and provide a wider understanding of whakapapa. Users also require marae level information.

5.3 Statistics New Zealand acknowledges the data need for marae but also notes there are concerns with the collection of marae information. Collecting marae information creates the expectation that the data will be available for release. However, marae data might not be of sufficient quality for release and confidentiality thresholds might constrain access to the data. There are also concerns that collecting additional information will increase respondent burden. So far, limited cognitive testing indicates respondents can more readily report marae than hapū. Findings from Te Kupenga show 71% of Māori know their marae.

5.4 There is insufficient evidence to show marae information will improve the coding of iwi. The ability to code and operationalise the collection of marae data is also currently constrained by Statistics New Zealand’s current processing systems. Statistics New Zealand should not be collecting data that will not be accurate enough to disseminate (either as data or statistics), or that does not – to a substantial extent – support improving the accuracy of the iwi data that is released.

5.5 Marae information might be better suited to collection by Māori.

5.6 All marae are geographically located. Moving forward marae could be useful for creating an administrative dataset or be based off a property frame / address register.

5.7 The final decision for including marae in a survey will occur independently of the development of the standard or classification. This is because census (and other surveys) has an independent content determination process to determine the final variables for inclusion in the questionnaire.

6. Non-kinship groups

Recommendation 6:
Further investigation required before developing a non-kinship standard/classification. Aim to include non-kinship groups in a future survey (e.g., Māori wellbeing survey or a household sample survey) and potentially partner with Māori to collect their own data.

6.1 Non-kinship groups play an important part of urban Māori life and offer a sense of belonging, identity, and security. Information about these groups is important for facilitating decision-making and policymaking. Government agencies and other users (including Māori) expressed a need for statistical information about non-kinship Māori groups.

6.2 Despite the clear need for information on non-kinship groups, there is strong opposition to integrating the fundamentally different concepts of kin and non-kinship within a standard. Kin groups are based on whakapapa and genealogical connection, whereas Māori can choose to belong (or not) to non-kinship groups.

6.3 Collecting information on non-kinship groups is complex. Non-kinship groups are fluid and may come and go over time (eg due to funding and resourcing). Non-kinship groups include Urban Marae but also include groups such as kapa haka, church, or sporting groups.

6.4 There are concerns that capturing non-kinship information might increase respondent burden and that data might not be of sufficient quality to release. So far, limited cognitive testing indicates further work is required.

6.5 Though there is a data need for non-kinship information, the current environment does not support collecting this information. Non-kinship information might be better suited to collection by Māori. Further work is necessary before Statistics New Zealand collects non-kinship information or the concept is developed and operationalised as a standard and / or classification (refer to Recommendation 2).

6.6 The final decision for including non-kinship groups in a survey will occur independently of the development of the standard or classification. This is because census (and other surveys) has an independent content determination process to determine the final variables for inclusion in the questionnaire.

7. Rohe / location

Recommendation 7:
Investigate adding the word ‘region’ alongside rohe to improve clarity. Continue to investigate further options for change. Keep informed of Statistics New Zealand’s geospatial iwi boundary mapping project.

7.1 Rohe is currently collected to support coding of iwi as a number of iwi share the same name across different regions. If respondents record these iwi, but do not record rohe, their iwi response is coded into a residual category.

7.2 The rohe used within the current classification are similar to regional council boundaries and do not necessarily reflect the ‘true rohe’ (based on physical markers) Māori identify with. These discrepancies result in respondents misinterpreting the term ‘rohe’. Statistics New Zealand needs to clarify what they mean by the term rohe.

7.3 Rohe are not geospatially-enabled boundaries and therefore are not directly comparable with data collected using geospatially-enabled boundaries. If rohe is changed to a geospatially enabled location measure (i.e., statistical area 1, statistical area 2, territorial authority, or regional council); there is concern that it will not accurately measure location in a way that is meaningful for Māori. Statistics New Zealand’s geospatial team is currently investigating the mapping of iwi boundaries, but timeframe of completion is unknown.

7.4 There is some opposition to the collection of rohe, given the limited utility of the information. Statistics New Zealand acknowledges these concerns, but also notes rohe (or an alternative location identifier) is important for coding iwi. We recommend maintaining rohe to support the coding of iwi, and investigate adding the word ‘region’ to improve clarity. Discontinuing or changing the collection of rohe may affect time-series data.

7.5 Further work is required to explore whether there is a more suitable location marker.

8. Data access and capability

Recommendation 8:
Work with Māori to increase analytical capabilities to enable Māori groups to collect and analyse their own data. Review legislative constraints imposed within the Statistics Act 1975 around confidentiality.

8.1 Access to iwi data is constrained by capability and capacity (due to awareness, skills, and cost). Working with Māori to increase capability might result in a greater number of Māori users able to collect, analyse, and access data.

8.2 The collection of lower level information might be more suitable for Māori to collect.

8.3 Access to data, especially data cross-tabulated by other variables, might be constrained by confidentiality thresholds. Currently all data, including iwi and Māori data, are under the same jurisdiction. Work to address the confidentiality rules and to explore options of releasing lower-level data back iwi and Māori is in progress (as part of the review of the Statistics Act 1975).

9. Consistency

Recommendation 9:
Promote and encourage the adoption of the statistical standard and classification of iwi across government.

9.1 There is limited consistency in the collection of iwi statistics across government limiting the integration and comparability of data. Improving consistency will require compliance with the standard across the data ecosystem.

9.2 The Act already allows for the promotion of standards across the data ecosystem. This should be retained in the review of the Act.

10. Collection guidelines where Māori descent precedes iwi

Recommendation 10:
Where Māori descent precedes an iwi question, change the routing rules (for both paper and online surveys) so that ‘yes’ and ‘don’t know’ responses are routed to the iwi question. In addition, include iwi data in outputs where there is a non-response for the Māori descent question and an iwi is recorded (paper surveys only).

10.1 In questionnaires/surveys, a question on Māori descent often precedes the iwi question. Currently if respondents answer ‘no’, ‘don’t know’, or do not answer the Māori descent question, the respondent is routed away from the iwi question. A small number of respondents record an iwi after answering ‘no’, ‘don’t know’, or not responding to the Māori descent question. In these cases, iwi data is captured (as raw data) but is not included in outputs.

10.2 We recommend including all iwi responses in data outputs. This will provide a more accurate measure of iwi.

10.3 Guidelines for the output of iwi following a question on Māori descent will depend on the survey mode utilised. Those responding to an online survey will see only the questions relevant to them (automatic routing); we recommend routing those who respond ‘yes’ and ‘don’t know’ to the Māori descent question to the iwi question. When surveys are in paper form, respondents see all questions; we recommend responses to the iwi question are output in cases where respondents have selected ‘yes’, ‘no’, or have not responded to the Māori descent question.

10.4 Changing the processing and output of iwi data may affect time-series (eg through increased counts). All previous iwi data has been captured however, which allows Statistics New Zealand to identify this group in past censuses (ie with back-series).

11. Review cycle

Recommendation 11:
Review the standard/classification for iwi at a time when appropriate and at regular intervals to enable adoption by users across the data ecosystem.

11.1 The statistical standard for iwi was developed in 1994. Although the classification has previously been reviewed, the standard has not. Reviewing the standard more regularly enables Statistics New Zealand to be responsive to change and gives users confidence in the relevance of the standard.

Available files
Download and print the PDF from 'Available files' on the right-hand side of this page. If you have problems viewing the files, see opening files and PDFs.

Citation
Statistics New Zealand (2016). Recommendations from the 2016 review of the statistical standard for iwi. Retrieved from www.stats.govt.nz

ISBN 978-0-908350-78-0 (online)
Published 31 October 2016

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