The experimental administrative population census (APC)
| Date | Tuesday 2 May 2023 |
|---|---|
| Time | 12pm - 1pm |
| Where | K.G.O9 or via Zoom: https://waikato.zoom.us/j/95492778502 |
| Presenter | Hannes Diener - Statistics NZ |
| Contact | Renae Dixon |
| Contact email | renae.dixon@waikato.ac.nz |
| Website | tengira.waikato.ac.nz/te-ngira-seminar-series |
| Admission Cost | Free |
ABSTRACT:
Like many other NSOs, Stats NZ is looking at ways we can produce more relevant and timely statistics while also reducing respondent burden by (re-) using administrative data. The NZ Census has already adopted a combined model that uses administrative data to fill in any gaps left in the field enumeration. This approach was first used in 2018 when the field collection was faced with lower-than-expected response rates, and it is included in the 2023 Census by design, to ensure the final data is as high quality as possible. The experimental administrative population census (APC) is taking the idea one step further and examines the potential for census information to be produced from administrative data only.
The second iteration of the APC was released in October 2022. It is a time-series spanning 2006—2021 and covers demographic (age, sex, geography, number of children born), identity (detailed ethnicity, Māori descent, birthplace and years since arrival in NZ), as well as education, income, and work variables. The APC is designed to be released iteratively and provides data users the opportunity to provide feedback and be involved in its further development.
In this talk we will cover various technical aspects of the APC, highlight research opportunities it provides, describe areas of ongoing development, but also outline information needs that are not well covered in administrative data.
Hannes is a senior design analyst in the Research and Development Hub which is part of Stats NZ's Method and Design group.
He received a PhD in mathematics from the University of Canterbury in 2008 and then spent a decade as a researcher working on topics in the intersection of logic, analysis, and the foundation of computer science.
In 2020 he changed interests with starting an MSc in (Bayesian) Statistics at the University of Canterbury, which was completed just after joining Stats NZ in 2021. Since then he has worked on the use of admin data in deriving statistics and has been leading the work on the experimental administrative population census (APC).