Stats NZ, New Zealand’s national statistical office and leader of the Official Statistics System, has been delivering New Zealand's most important official statistics for 120 years. It is governed by and administers the Statistics Act 1975. Under Section 3 of the Act, Statistics NZ holds a census every five years.
The census is one of New Zealand’s most important data sources and much of the census data is released for re-use. Statistics NZ released aggregated, confidentialised data from the 2006 Census including information on age, ethnicity, income, workplace, dwelling size and other variables.
Data included summary statistics, aggregate data in the NZ.Stat tool available via the Statistics NZ website (previously Infoshare and Tablebuilder), and a meshblock dataset. Data is available in comma delimited (.csv), Excel (.xls), Access (.mdb), Table Query (.tqx) and Beyond 20/20 (.ivt) formats.
The 2006 Census meshblock dataset contains counts at the meshblock level for selected variables from the 2006, 2001 and 1996 Census of Population and Dwellings, rebased to 2006 Census boundaries. This dataset also contains counts for area units, wards, territorial authorities, and regional council areas. The meshblock dataset can be integrated with other databases or mapping systems to assist analysis and decision making.
Releasing data for re-use is part of the census programme. Under the Statistics Act, release of official statistics must be balanced with the protection of personal information. A dedicated project team develops products and data for release, including output systems, tools, and supporting metadata. For each new census, changes and improvements to this range of products are made, based on user feedback.
Confidentiality and privacy - Information Stats NZ collects can only be used for statistical purposes, must be kept secure to prevent unauthorised access, and must not be released where it could lead to disclosure of individuals' details. Where necessary, Stats NZ adjusts data to make sure that no individual or business can be identified, using techniques such as random rounding, collapsing categories, and suppressing cells. Under specific conditions, researchers can access anonymised unit record data.
Data quality - Census data must meet the quality criteria set out in the Principles and Protocols for Producers of Tier 1 Statistics. The census aims to meet information needs with data that is fit-for-purpose. Of fundamental importance are the accuracy of the population count and statistics on the characteristics of small populations. Significant emphasis is given to ensuring the quality of core census variables including age, sex, ethnicity, and location. The 2006 Census achieved a coverage rate of 98 percent and the response rate was 94.8 percent. A post-enumeration survey measures the undercount, which was 2 percent in 2006.
Release of data for re-use, as part of the core census budget, does not incur additional cost.
Census data is released in a phased manner following each census.
Note that the 2011 Census was deferred until March 2013 due to the 2011 Canterbury earthquakes. 2013 Census results will be released from December 2013.