Approved researchers can now use 2018 Census data at unit record level in the Stats NZ Data Lab, Stats NZ said today.
Two anonymised datasets are available – individual and dwelling. These are subsets of the datasets used within Stats NZ and most variables from the individual and dwelling census forms are included. The exceptions are those variables that cannot be used because the data is very poor quality (iwi affiliation, absentees, and families and households).
The individual dataset has a variable that identifies the source of the record, either from an individual form response, a partial response, or from admin enumeration. The 2018 Census incorporated significantly more non-response mitigation methods than previous censuses. Item source variables on both individual and dwelling datasets explain where the information for given variables was sourced from, either 2018 Census forms, 2013 Census data, administrative data or statistical imputation.
The 2018 Census data in the Data Lab is stand-alone microdata only. 2018 Census microdata will be linked within the Integrated Data Infrastructure (IDI) following the December 2019 refresh.
Apply to use microdata for research explains how to apply to access the Data Lab.
About the 2018 Census dataset
We combined data from the census forms with administrative data to create the 2018 Census dataset, which meets Stats NZ’s quality criteria for population structure information.
We added real data about real people to the dataset where we were confident the people should be counted but hadn’t completed a census form. We also used data from the 2013 Census and administrative sources and statistical imputation methods to fill in some missing characteristics of people and dwellings.
An independent panel of experts has assessed the quality of the 2018 Census dataset. In its Initial report of the 2018 External Data Quality Panel, the panel endorsed the overall methods Stats NZ used for the 2018 Census and assessed the quality of some of the key variables. The second report 2018 Census External Data Quality Panel: Assessment of variables assesses the quality of 30 more variables.
Data quality ratings for 2018 Census variables summarises the quality rating and priority levels for 2018 Census variables.
If you have any enquiries, please email [email protected].
