17 January 2024: This page has been updated with the links below to provide access to the feedback received on the consultation, and to the methodologies paper that sets out how gaps in census responses will be managed.
The consultation for gender and sex concepts methodology for the 2023 Census closed in January 2023.
Thank you to the submitters who provided feedback on the planned approach to using alternative data sources to fill gaps in census responses for the four gender and sex concepts: gender, sex at birth, sexual identity, and variations of sex characteristics.
Gender and sex concepts 2023 Census methodology consultation - feedback summarises the feedback we received.
Stats NZ has finalised the methodologies for the 2023 Census.
Methodologies for filling gaps in gender and sex at birth concepts for the 2023 Census sets out how gaps in census responses will be managed for the 2023 Census.
Combined census model
For the 2023 Census, to give Aotearoa New Zealand the most complete census possible, we will be adding existing data from other sources to the data we collect from the census. We call this a 'combined census model'.
The combined census model involves collecting data as part of the 2023 Census and using alternative data sources, including administrative (admin) data, historical census data, and statistical imputation, to fill data gaps, if required and where the data exists. This is called a 'combined methodology' because it involves combining different datasets.
Admin data is data that is collected by government agencies or other organisations while they are conducting their normal business, such as delivering a service or recording an event. However, there are gaps in admin data for key population groups and variables, so participation in the census field collection remains the main way of making sure we gather quality census data.
Gender and sex concepts methodology
As part of the combined census model, we may use alternative data sources on a variable-by-variable basis to fill gaps in census responses and improve data quality. There are options for how we use these alternative data sources.
Because we have made changes to the sex and gender variables in the 2023 Census, we would like feedback on our planned approach to the use of alternative data sources for these variables. The four sex and gender concepts to be covered as part of this consultation are: gender, sex at birth, sexual identity, and variations of sex characteristics.
The 2023 Census content and the form design and questions have been decided by the Deputy Government Statistician, Census and Collection Operations, on behalf of the Government Statistician. For more information, please see:
We have three main methods for filling in gaps in census responses. Where available, we can use:
- historical census responses (for sex at birth and gender this potentially includes 2013 or 2018 Census responses to the question, 'Are you Male/Female?' (with no clarification of what that meant). Previous censuses did not ask about sexual identity or variation of sex characteristics.)
- admin data (information taken from an admin data source)
- statistical imputation (We use the term 'imputation' to refer to the replacement of missing information with values from a statistical process, in contrast to our methods of sourcing real values from admin or historical census data).
Key questions for you to consider in your submission
Gender
When we need to fill in the gaps in census responses to do with gender, we intend to use all three methods described above.
- What is your preferred order for using the three methods - historical (2013 and 2018) census responses, admin data, and statistical imputation?
- If we use historical census data, should we use both 2013 and 2018 Census?
- How much do you think gender is likely to change over time?
Sex at birth
We have similar options for sex at birth as we do for gender. The sex question in previous censuses did not specify sex at birth, so there may be some variation in how respondents interpreted the question. We are considering whether it appropriate to use these historical responses when we do not have a 2023 Census response for sex at birth.
- Should we be using historical census data for sex at birth?
- If, after using admin data and/or historical census responses, we still have gaps, should we fill these in using statistical imputation or leave gaps in the data?
- What is your preferred order for using the selected methods?
Sexual identity
We do not have historical census data as this question is new, and we have limited coverage from admin data sources. We could consider using statistical imputation.
- Should we attempt to fill in missing sexual identity information using statistical imputation?
Variations of sex characteristics
We do not have historical census data as this question is new, and we have no admin data sources. Our only option would be to use statistical imputation.
- Should we attempt to fill in missing variations of sex characteristics information using statistical imputation?
For more information
Email [email protected]
