Intercensal modelled population estimates for Papua New Guinea

Project lead: Andy Tatem, Attila Lazar

Team: Hal Voepel, Amy Bonnie, Chris Nnanatu, Heather Chamberlain, Duygu Cihan, Tim O’Riordan, Alexandra Frosch

Funding: UNFPA

Start: Aug 2021
Completion: Jul 2023

Papua New Guinea’s (PNG) fourth National Population and Housing Census was conducted in July 2011. Although preparations for the fifth census are advanced, because of the Covid-19 pandemic the Government of PNG deferred the census possibly until 2024. To enable robust and accurate planning and development decisions to be made, intercensal subnational demographic data on populations throughout the country were needed.

Our team provided technical assistance for estimating population numbers and distribution using the available geolocated household survey (DHS, Urban structural listing and Malaria survey), 5m resolution settlement data (Planet) and other geospatial datasets in a Bayesian statistical model that provided interim population estimates for national decision making until the fifth housing and population census is completed. Such small area estimates are also valuable in supporting updated planning efforts for the forthcoming census.

In this project we:

  • Developed and tested methods and procedures for predicting population numbers at sub-national scales, with age and sex breakdowns and associated uncertainty measures.
  • Produced sex- and age-disaggregated population estimates using the methods developed.
  • Trained Papua New Guinea’s National Statistics Officers and other relevant officials to ensure that methods are understood and that the outputs are effectively used within the government.