Census participation lowest in five surveys: Causes problems for data collection and delays

The number of individual responses received from the 2018 Census was lower than expected. Photo / Bevan Conley

Around 10 per cent of New Zealanders didn't fill out their Census forms this year, resulting in the lowest participation nationally for the past five surveys.

Statistics NZ's interim calculations, released today, show that full or partial information for only 90 per cent of individuals was received this year, compared with 94.5 per cent for the 2013 Census.

The low participation has caused a delay in the release of information collected, which was due to be released in October - but will now be available in March 2019.

In order to make up for missing data, Stats NZ will also have to use statistical methodology to compensate - this involves using other sources of information and different imputation approaches.

"There is a long-term, international trend of declining census response rates. Because of this we have made a strategic decision to use more administrative data to improve the quality of census data," Stats NZ said.

The 2018 Census was the first time Stats NZ used a digital-first approach, which encouraged respondents to complete the census online.

Stats NZ said they chose to go digital to improve data quality, while modernising.

Their aim was to make the census as cost effective as possible, while ensuring it delivered
high-quality data to its customers.

"We optimised the online experience so it should be quick and easy, with
functionality built in to make the forms as intuitive and accurate as possible," it said.

Interim figures showed the organisation surpassed their online participation target of 70 per cent, however, some aspects were more challenging and the number of individual responses received was lower than expected.

Lower response rates could have downstream effects on population estimates and projections, and some key data areas that make the census unique, including family and household structure.

It could also affect small population areas and rural settlements.

Due to this year's results, Stats NZ said they were undertaking analysis on how to improve data for small populations, subgroups, and small geographies.

Three types of imputation methods will be used, including; unit imputation, item imputation and donor imputation.

Imputation is the replacement of missing information with what is expected to be true information when looking at other questionnaire responses.

For the 2013 Census, four variables were imputed where a valid response was missing: age, sex, usual residence, work and labour force status.

"The new date for our first release will give us time to develop revised methodology for processing and analysing census data," Stats NZ said.

"We are committed to delivering a high-quality and accurate dataset."

What is the Census?

The census is the official count of New Zealand's population and dwellings.

Every five years, Stats NZ asks everyone in the country to provide information about themselves and their dwellings.

The data from the census is used by Government, iwi, businesses, and community groups across the country to make informed decisions.

Census data enables accurate population projections and estimates. It underpins decisions about policy development, funding allocation, and service provision, and determines how billions of dollars of Government funding is spent.