01 May 2023
Categories:
  • Project
Tagged:
  • Call Detail Records (CDR)
  • COVID-19
  • mobility analysis
  • Papua New Guinea
  • Population Displacement

Authors

The report was authored by Robert Eyre, with the contribution of Galina Veres, James Harrison, Sophie Delaporte, Xavier Vollenweider, and Caterina Irdi from Flowminder and Shohei Nakamura from the World Bank.

This report was made possible thanks to the aggregated and anonymised mobility indicators derived from Call Detail Records (CDR data) provided by Digicel Papua New Guinea (PNG).

Executive summary

Flowminder produced indicators of mobility for Papua New Guinea, using aggregated and anonymised data derived from Call Detail Records (CDR data) of the Mobile Network Operator Digicel Papua New Guinea.

These indicators were used to measure changes in mobility from December 2019 to May 2020, period during which two main sets of measures against the COVID-19 were introduced by the government:

  • Phase 1, 26 Jan 2020 - 23 March 2020: Travels from Asia and Indonesia are banned;
  • Phase 2, 24 March 2020 - 22 April 2020: Travels between districts within Papua New Guinea are also banned.

Key findings

Only the Phase 2 measures appear to have been effective in contributing to a reduction in mobility within Papua New Guinea.

All results are expressed as % change compared to the baseline period, i.e. normal conditions (Baseline from 2019-12-01 to 2020-01-25).

During Phase 1:

  • Urban areas were visited by subscribers from more district;
  • The daily count of unique subscribers in urban areas increased by up to 20% although it did decrease by 10 % in rural areas.

During Phase 2:

  • Subscribers travelled less than normally within the province where they live and even within the district where they live. As a result:
    • The daily count of unique subscribers in urban districts decreased by 10% on average with the largest decreases in unique subscribers in urban areas recorded in Central Papua New Guinea, between -30% to -50%;
    • There was less mixing of subscribers coming from various areas of the country (entropy indicators and population mixing indicators)
  • Subscribers travelled from areas where population population mixing was lower than during the baseline period, which hence likely reduced the risk of having them been exposed to the virus before their journey.

Read the full report (PDF)

Back to Publications & Reports