ACN - 101321555 Australasian Human Research Ethics Consultancy Services Pty Ltd (AHRECS)

Resource Library

Research Ethics MonthlyAbout Us

ResourcesHuman Research EthicsTe Mana Raraunga Statement on 2018 New Zealand Census of Population and Dwellings: A Call for Action on Māori Census Data

Australasian Human Research Ethics Consultancy Services Pty Ltd (AHRECS)

Te Mana Raraunga Statement on 2018 New Zealand Census of Population and Dwellings: A Call for Action on Māori Census Data

Published/Released on July 24, 2018 | Posted by Admin on July 24, 2018 / , , , ,

View full details | Download file

The five-yearly Census of Population and Dwellings is the flagship of the Official Statistics System (OSS) and is essential for many of the functions that underpin democracy. Te Mana Raraunga, the Māori Data Sovereignty Network, is concerned that Census 2018 may fail to deliver high quality Māori and iwi data. Te Mana Raraunga supports a comprehensive independent review of Census 2018 and calls for Māori governance of Māori data across the entire Official Statistics System.

National data collections, such as population censuses, are critical sources from which we reach conclusions about the nature of our society on a number of fronts. These perspectives are frequently used to develop and shape interventions (often) in the form of socio-economic policies that are used to promote social incentives and provide solutions to social problems such as poverty, and social inequality. But when data collections fall short in terms of level accuracy and completeness then there are serious consequences for the credibility of the processes referred to above. Poorly run censuses might not only yield compromised results but could also have an impact on public trust of government data and decision-making.

The failings of the 2018 New Zealand Census with respect to data on Maori presents obvious ethical challenges especially with regard to the incapacity to deal effectively with matters of social injustice and inequality so often suffered by indigenous peoples. Te Mana Raraunga has put together a press release and a more detailed technical statement on Census 2018. The two comments describe these concerns in more detail.

Australia’s recent experience with census collection demonstrates that poor data collection practices can completely undermine public trust in large scale e data collection.

Further discussion of Indigenous Data Sovereignty can be found here.

Problems with the 2018 Census and Stats NZ response

Interim figures for the 2018 Census released by Stats NZ indicate that full or partial information has only been received for about 90 percent of individuals, compared with 94.5 percent for the 2013 Census1. Given that a key goal of the census is to count all usual residents in the country on census night2, commentators are rightly concerned that up to ten percent of the population may be missing3. For Māori, the extent of the problem will inevitably be worse. Census 2018 may yet turn out to be the poorest quality enumeration of Māori in recent history.
But how poor? Stats NZ will not have a definitive answer for some months yet but the early signs are not positive. Let’s begin with the ‘full or partial’ information received by 90 percent of individuals. One might have the impression that ‘partial’ information means incomplete information on an individual’s census form. However, as used by Stats NZ, ‘partial’ information appears to mean a partial-response dwelling where there is no individual form but the dwelling form or household summary page has a list of people at the dwelling on census night4. We do not yet know what share of the 90 percent comprises partial-response dwellings, however we can gain some insight by considering the 2013 results. While recent Stats NZ releases report full or partial informaton was received for 94.5 percent5 of individuals for the 2013 Census, the total (or achieved) response rate was 92.9 percent6. The lower, and more informative, figure excludes all individuals in partly and completely missing households in 2013, as well as the 2.4 percent estimated national net undercount (coverage level) determined by the Post Enumeration Survey7 undertaken after the 2013 Census. Stats NZ has also noted that for Census 2018 ‘there are more households where no one has responded to the census than previous censuses’.
What does all of this mean for Census 2018? It means that the total response rate will inevitably be below 90 per cent. For Māori, the 2018 total response rate will be be significantly below 90 per cent. The crucial question is, how much lower? And at what point does this seriously compromise the quality and usefulness of the census data? To date Stats NZ has not provided any guidance on these important questions but needs to.
Constitutional and other implications for Māori
Recently Stats NZ announced that: ‘New Zealanders can be confident the 2018 Census will produce accurate and high-quality data which can be relied on by communities and decision-makers’9. We question whether this will be the case for Māori communities, iwi and Māori decision-makers.

Read the rest of this statement

Resources Menu

Research Integrity

Human Research Ethics