Abstract
Based on original and recent household and telephone surveys in the Democratic Republic of the Congo (DRC), Haiti and Ghana, we estimate biases of mobility and migration statistics derived from mobile phone data (MPD).
We show that indicators of short- and long-term internal mobility differ greatly between mobile phone users and non-users, where phone users show higher mobility on average. These differences translate into biases when statistics are derived from mobile phone data and population inferences are made without adjustments.
These findings are not only relevant for mobility research but also have wider implications for the use of mobile phone data and other Big Data for analyses of mobility and migration.
Read the full abstract
Preferred citation
Flowminder Foundation, Hosner R. et al. (2025). Bias estimates for mobile phone data (MPD)-derived mobility statistics in low-and middle-income countries.
Authors
Roland Hosner1, Zachary Strain-Fajth1, Veronique Lefebvre1
1 Flowminder Foundation
This abstract was accepted and presented at Netmob 2025.
NetMob is the primary conference on the analysis of mobile phone datasets in social, urban, societal and industrial problems.