Census data user guide

This section answers some of the common questions raised by data users about the collection and processing of census data.

What areas are covered by the census?

The geographic coverage of the census includes the North Island, South Island, Stewart Island, and the Chatham Islands, plus offshore islands including the Kermadec Islands, Three Kings Islands, Mayor Island, Motiti Island, White Island, Moutohora Island, Bounty Islands, Snares Islands, Antipodes Islands, Auckland Islands, and Campbell Island. Counts of people by sex on Ross Dependency are taken but are not included in the New Zealand population count.

How is everyone counted in the census?

An efficient and successful data collection process plays an important part in ensuring high-quality data can be produced from the census. The collections phase is a huge logistical exercise and there are several challenges: planning the operation, identifying and getting access to dwellings, locating respondents, and ensuring compliance from respondents. There are three main elements to a successful data collection:

• delivery of census forms to every person and occupied dwelling in New Zealand by census night
• high response rate from those present in New Zealand on census night
• efficient collection of census forms from every person and occupied dwelling.

Statistics NZ recruits a team of census collectors before each census. These collectors deliver and collect census forms to individuals and households across New Zealand. They work in their assigned location and are provided with full training and support.

The process of data collection begins with the delivery of census forms to dwellings by collectors. The collectors leave two types of forms: a dwelling form and an individual form for everyone who will be in the dwelling on census night. Collectors also provide an Internet Access Code for the dwelling.

Every effort is made to ensure that census forms are delivered to all dwellings. In addition to private dwellings, census forms are also delivered to non-private dwellings, such as hotels, motels, prisons, hospitals, camping grounds, and other places where people may be on census night.

In the past, inner-city apartments, especially in Auckland and Wellington, have posed a challenge to collectors. Knowing where apartments are located, gaining access to them, and establishing if the apartments are occupied all make the job difficult. To overcome these challenges, collectors have to use different strategies from those used for other dwellings. These strategies include identifying high-rise inner-city apartments in advance, networking with building managers to gain their cooperation and getting access to the apartments, and using extra resources to increase awareness among residents of these apartments. If known collection issues like these are not addressed, there can be significant impacts on data quality, particularly at the area unit or meshblock levels.

Individuals can complete their forms online or on paper. Doing the forms online is secure, quick, and easy. If everyone in a household completes their forms online then a collector may not need to call back.

Census forms can be completed in English or Māori using an English form or a Māori/English form. Māori/English forms are delivered to all dwellings in Northland, Whakatane, Gisborne, and the Chatham Islands. All other individuals can request a Māori/English form in three ways: from the collector, by ringing the toll-free census number, or through the census website. To help individuals fill census forms, Statistics NZ:

• provides guide notes together with the census form. Also available online, these guide notes help respondents answer some specific questions in the form. They also answer some frequently asked questions that individuals may have while filling in the census form.
• sets up a toll-free census helpline number well before the census date. This helpline assists the public with their phone enquiries about the census. This service is offered in eight languages: English, Māori, Samoan, Tongan, Mandarin, Cantonese, Korean, and Hindi.
• offers online help through the census website.

The collection of census forms from dwellings starts immediately after census day. Unless everyone in a dwelling has filled in their forms online, the collector will need to return to that dwelling to collect paper forms. When the collector is unable to make contact with dwelling occupants after three attempts, a freepost envelope is left for the forms to be returned in. A small but significant part of Statistics NZ's quality management plan is to get collectors to check (while collecting forms) that individuals have answered the key questions (foremost variables) on the first page of the individual form.

Filling in census forms is required by law and an individual who does not participate in the census or provides false information can be prosecuted. Statistics NZ recognises that the legal requirement to complete the census is not by itself enough to achieve the high response rate that is desired. Consequently, Statistics NZ launches a broad public communications programme – using advertising, news media, community programmes, and social media websites – to encourage people to take part in the census. The information campaign about the census aims to tell people why the census is important, when it is, and how to take part.

Who must fill in a census form?

Filling in a census form is required by law. Anyone who does not take part in the census or provides false information can be prosecuted.

Despite the legal requirement to complete the census form, and the best efforts made by Statistics NZ to get census forms to everyone in New Zealand, it is inevitable that some people will be missed and some counted more than once.

Statistics NZ has undertaken a post-enumeration survey (PES) after each census since 1996. The PES estimates how many people were missed or counted more than once in the census.

The PES showed that about 98.0 percent of New Zealand residents in the country on census night were counted in the 2006 Census, compared with 97.8 percent in the 2001 Census. In the international context, recent census coverage results in New Zealand are similar to those in Australia and Canada. For example, in Australia, 97.3 percent of people were counted in the 2006 census and 98.2 percent in 2001. Results from the PES are used to adjust the population base for deriving post-censal population estimates and demographic projections. However, census counts are not altered.

Individual census questions are also subject to non-response. These are discussed in the 'What is non-response' section in this guide.

How do topics get included in the census?

The questions asked in the census form the basis of the information available to data users. Although Statistics NZ has a minimal-content-change strategy to provide data time-series continuity, it also considers including new topics and changing existing census questions to reflect real world changes and emerging information needs. When including topics in the census, Statistics NZ is guided by these criteria:

• Is the census the most appropriate information source?
• Is there public acceptance of the topic?
• Will the topic produce quality information?
• Does the topic have significant community value?

However, there are also constraints to what can and cannot be included. These constraints are that:

• some topics are required by law
• the collection of data must be cost effective
• the length of the forms is limited
• respondent burden and resistance must be minimised
• continuity and relevance need to be balanced
• the need for data consistency must be considered.

Statistics NZ invites submissions from the public and special interest groups, and also consults data users, when considering changes to census content.

Following content decisions, the form-development phase involves redesigning the census forms, and balancing the requirements of content, printing, scanning, and the size of the forms.

The questions identified for redevelopment are usually those affected by data quality concerns from previous censuses (eg growing numbers of item non-response) or real world changes that require revisions to questions to reflect new legislation or changes in terminology.

Further details of the content and form design phases of the 2011 Census can be found in the 2011 Census Content Report and 2011 Census Form Design Report, which are available from the Statistics NZ website.

How good is census data for my purposes?

The primary purpose of the census is to provide social, economic, and demographic information on the people of New Zealand at a given point in time. Before releasing any census data, Statistics NZ ensures the data is ‘fit for use’.

Census variables are ranked by three 'quality levels' – foremost, defining, and supplementary. The basis for defining quality is 'fitness for use’. These quality levels (described below) are used as a guide throughout the census. A list of census output variables, together with the quality level assigned to each variable, is given in the Census variables by quality level chapter of this guide.

Level 1 – foremost variables are core census variables and their outputs are the main reason for conducting a census. These variables include age, sex, ethnicity, and location. Some of these variables produce the key outputs used for maintaining the accuracy of population estimates. Across all phases of the census, foremost variables are given the highest priority in terms of quality, time, and resources.

Level 2 – defining variables describe the key subject populations that the census provides measures for. They are important for policy development, evaluation, or monitoring. Defining variables are used frequently in cross-tabulations with foremost variables. They represent key sub-populations and the measures that are of high public interest, for example, iwi, birthplace, and labour force status. These variables are closely linked to the main purpose of the census and in the New Zealand context, may only be available in detail, for example, at the subnational level. These variables have second priority in terms of quality, time, and resources across all phases of the census.

Level 3 – supplementary variables do not fit directly with the primary purpose of the census but are important to some groups of data users. Examples include occupation, language, and religious affiliation. These variables have third priority in terms of effort and resources.

Although the level of quality checks that a variable is subject to is dependent on its quality level, minimum quality standards still need to be met before Statistics NZ releases any output data.

How does Statistics NZ ensure that census data is fit for use?

Statistics NZ produces census data that is fit for use. A comprehensive testing programme is undertaken before the census to ensure that its key processes and systems are working efficiently and to specification. The programme consists of different types of tests. These usually include form tests (cognitive and usability tests), field tests, system tests, integration tests, and a dress rehearsal.

Before a new question is included on the census form, or before a change is made to an existing question, these questions undergo a robust programme of cognitive testing. Cognitive testing shows how well a question meets the combined needs of respondents and data users. It confirms that the question is collecting good quality data.

Field tests ensure that the census forms collect reliable information and that field procedures are working effectively. The data processing system is thoroughly tested to ensure that the system treats the responses on the census form correctly.

Individual systems are tested, and then combined and tested as a group.

After the data has been processed and before it is published, the data is evaluated. The aim of the data evaluation phase is to ensure that the data provided by respondents has not been changed by the systems or processes used by Statistics NZ, that is, the data has not been subject to introduced errors.

The 2011 Census Testing Strategy adopted for the 2011 Census (which was deferred because of the February 2011 Christchurch earthquake) can be accessed from the Statistics NZ website.

What are the possible sources of error?

As the census covers the entire population of New Zealand, it is not subject to sampling error, which occurs when a sample, rather than the entire population, is surveyed. However, census data may be subject to non-sampling errors resulting from respondent errors, collection or processing errors, and undercounts. Statistics NZ strives to reduce each of these error types as much as possible and provide data that is fit for use.

Being self-administered, the census may be subject to errors made by individuals when filling in census forms. These could happen because individuals misunderstand the question, accidentally mark the wrong box, or give a partial response or no response at all to census questions that were relevant to them. To minimise these errors, census forms have been designed so that questions are as easy to understand and as simple to answer as possible. The introduction of online census forms has further helped minimise respondent errors. Built-in editing functionality directs individuals to the appropriate questions and ensures that their responses are valid.

To minimise purposeful distortion of information by individuals, the importance of the census is communicated through a variety of media channels – such as television, radio, the Internet, and newspapers – and through a community outreach programme. Guide notes (delivered with both paper and online census forms), other online help, and the toll-free census helpline number help individuals complete their census forms.

Collection errors, also called enumeration errors, are errors made by collectors when forms are delivered to or collected from dwellings. These could include assigning dwellings to an incorrect meshblock, misidentifying a dwelling as occupied or unoccupied, or incorrectly classifying a dwelling as private or non-private. Statistics NZ has checks and balances at different stages of the collection, processing, and evaluation phases to identify and fix these errors.

Examples of errors that can occur during data processing include incorrectly classifying responses, and misrecognition of written responses. Checks are made during data processing to identify possible errors and correct them if necessary. The data processing phase is then followed by a data evaluation phase, where further checks on the data are done to ensure that it meets quality standards and is fit for use.

What are derived variables?

Some census output variables are derived from responses to individual questions or from a combination of responses given to two or more questions on the census forms. These are called derived variables. Examples of derivations are shown in table 1. 

Table 1

In addition to the above derivations, aggregated family and household incomes are derived by grouping total personal income data as detailed in table 2.

Table 2

Why do I need to know if a variable is derived?

Derived variables are dependent on the quality of the input variables. Any errors or issues associated with the input variable(s) are likely to affect the data quality of the derived variable. The effect could be further compounded when two or more census questions feed into the derived variable. If such issues exist, Statistics NZ lists them in the Information by Variable metadata product. Data users need to know these issues before using derived variables.