Edouard-Legoupil
commented
10 months ago In the context of migration statistics, forced displacement is often analyzed with the prism of push and pull factors.
Though, traditional statistical data sources are often lacking sufficient geographically-fine-grained disaggregation to inform sub national scale approach and characterization. Alternative based on sophisticated index like Inform Colombia requires extensive expert consultations and might not fully reflect the important dimension to be reflected in the context of forced displacement and migration.
New sensors provide unique abilities to capture new flow of information from social medias (Anonymized data from Facebook platform) at subnational scale through grid level information. Satellite data can pick up signals of economic activity by detecting light at night, it can pick up development status by detecting infrastructure such as roads, and it can pick up signals for individual household wealth by detecting different building footprints and roof types.
In regard to the framework above, an initial selection of globally available layers includes:
-
Economics:
- Weighted Relative Wealth Index (Facebook-RWI)
- Social Connectedness Index (Facebook-SCI)
- Public Services Catchment area (OSM)
-
Environment:
- Agricultural drought frequency Index (FAO-ASI)
- Climatic Natural Risk - Flood and Cyclone (Prevention Web)
- Geologic Natural Risk - Earthquake and Volcano (Prevention Web)
-
Social:
- Population Dependency Ratio (Facebook)
- Movement Range data sets Facebook
- Violence (ACLED)
Information can be compiled and aggregated at admin level 2 in order to build composite Indicators. Different areas can be then grouped together based on the values from those composite indicators. The advantage of this approach are multiple: 1. Granularity: Optimal Level of granularity 2. Availibility: Data Consistently and freely available worldwide, simplicity to obtain information, ensor based indicators are potentially less sensitive to political pressure 3. Reproducibility: Can be used in multiple countries easily and Fully automated and audited through reproducible analysis script
Client - Validation
The current country template is pre-filled with the admin 2 code and name -
Would make it user friendly if the template was pre-populated per default a series of additional indicator to be compiled automatically from public source available at sub-national scale -
See some indicator exploration here: https://unhcr-americas.github.io/Area_Based_Approach/ --
Obviously the app user can decide to remove all those indicators as required
Dev - Tech
With some data extraction script to recycle from the same repo... https://github.com/unhcr-americas/Area_Based_Approach/blob/main/R/get_data.R