In this project, we seek to expand the existing work that has been conducted by Jean et al predicting poverty. We use the pytorch replication done by Jatin Mathur. We explore new countries in West Africa and approach the problem from the angle of education, a topic which is intricately linked with global poverty.
The stark lack of reliable data in developing countries is a major obstacle to sustainable development. This data drought should be addressed if we hope to inform policy decisions and help direct humanitarian efforts. The use of satellite data may be a way to do this, as remote satellite imagery is becoming increasingly available and inexpensive.
We use the same data sources as the original projects: The World Bank, Google Maps, and Night time lights.
sect3_plantingw3.csv contains education related indicators (ie. dropout rate, high education level obtained, etc.)
cons_agg_wave3_visit2.csv contains the average education consumption aggregate; the variable is ‘edtexp’
nga_householdgeovars_y3.csv contains the geovariables of households as variables 'LAT_DD_MOD' and 'LON_DD_MOD'.
All three files contain a column ‘hhid’ that is the household identification number so you can cross-reference data across different files. The state data dictionary can be found here (https://microdata.worldbank.org/index.php/catalog/2734/data-dictionary/F41?file_name=sect2_harvestw3) under STATE CODE.
Complete clean dataset: Schools-with-lat-long.csv
Check out our other repo https://github.com/raphaelletseng/datallite-site, and site in progress.
Contact our team with any inquiries through our email, [email protected].
Follow us on Linkedin at https://www.linkedin.com/company/datallite/about/