Michael Peters, Nicholas Cerimeli

Annie Nerger and Chelsea Doyle

Rnadom Forest

The following are steps undertaken for deidentifying NDMA data. The data is dis aggregated per county for all 23 counties - ASAL. The information covers the years of 2000 - 2020, where data prior to 2016 was stored in a different database (REWAS) and data from 2016 henceforth in the new database (DEWS). In each county data set workbook there are 6 different sheets: HHA REWAS, HHA DEWS, KIA REWAS, KIA DEWS, MUAC REWAS, MUAC DEWS The process involves inspecting individual sheets for each data set, dropping P.I.I columns, and then writing all the sheets to a single workbook - Kitui.