Upon receiving access to the unblinded dataset, it became evident that small number of issues in the replication package had to be resolved for the replication code to work properly with the unblinded data. These corrections do not affect the estimation strategy, the construction of variables, or the inclusion/exclusion of data from the study. 


./00B-MASTER.R
Line 146: Incorrect file name. Changed to "Commune-Level Treatment Assignment.dta". 

./Code/Rcode - reproducible package/01-09 - Data Cleaning and Preparation/
02 - preparation - CBO group preparation for analysis.R
Line 206: Added the variable 'treatment' to CBO_groups_final_IND. This variable had been omitted in the blind analysis. 

./Code/Rcode - reproducible package/01-09 - Data Cleaning and Preparation/
09 - corrections after unblinding.R
### This script has been added on after unblinding. It fixes two variables that are missing/have the wrong data type in the unblinded data. 

DM_final$region_commune <- paste(DM_final$region, DM_final$commune, sep="_")
DM_final$treatmentcommune <- as.numeric(DM_final$treatmentcommune==1)

DM_finalx2$region_commune <- paste(DM_finalx2$region, DM_finalx2$commune, sep="_")
DM_finalx2$treatmentcommune <- as.numeric(DM_finalx2$treatmentcommune==1)

saveRDS(DM_final, file = here(datatype, "Final", "DM_final.RDS"))
saveRDS(DM_finalx2, file = here(datatype, "Final", "DM_finalx2.RDS"))

message("**09 completed")
