In collaboration with the practitioners, all stakeholders selected 13 data types to consider per saving groups during the mapping exercise. Those data types include: SG location (Province, District and Sector), SG name, SG year of creation, SG total membership (female & male members), International NGO affiliation, Local NGO affiliation, SG status (graduated or supervised), SG saved amount as of December 2016 and SG outstanding loan as of December 2016.
All practitioners compiled their data in the provided excel template and uploaded their data through cartix platform. Cartix is embedded with machine learning capabilities whereby it assisted practitioners by applying auto-correct or flagging fields that needed to be revised. This feature allowed practitioners to fast-track the data collection, compilation & data validation phases simultaneously.Other financial datasets were also considered, these datasets include: data on financial institutions (banks, MFIs & SACCOs), data on mobile agents (Telco agents & bank agents) and finally we used FinScope data (2012 & 2016 surveys). This allowed us to contrast the informal financial sector with the formal financial sector to get more insight on the current outlook from a data perspective.
The mapped datasets were collected between 2010 and 2016.