Country-and-Region-Metadata

Overview

• Purpose: To create a useful metadata solution for all sectors—both for upstream input and as a reference for verification
• It is flexible, easy to collaborate on, and helps avoid duplicate work
• Produces modularised output (based on regional grouping) that all sectors can use; a combined output file is also saved
• Each regional grouping has one CSV output, one script, and one raw input folder—all named according to the regional grouping
• All outputs follow the same format and use the same codebase as the data warehouse

Setup

We aim to align all the coding with the data warehouse.

• raw_data: Contains the original input files for producing the "output", folder names should follow the regional_grouping, e.g., "UNICEF_PROG_REG_GLOBAL"
• output: Contains files for each regional grouping (or parent), named by the regional_grouping, e.g., "UNICEF_PROG_REG_GLOBAL.csv"
• script for output: Contains the scripts to produce the output files, e.g., "UNICEF_PROG_REG_GLOBAL.R"
• R: Contains general functions used by each script, e.g. "general_functions.R"

Output

The output files from this project serve as inputs for downstream processes. The data is in a long format, where each region is mapped to all the ISO3Code values belonging to it.
Format:
Regional_Grouping: uppercase, only connected by underscore, no hyphen or space, e.g., "UNICEF_PROG_REG_GLOBAL"
Region_Code: same code as used by the data warehouse, uppercase, only connected by underscore, no hyphen or space, and led by Regional_Grouping. e.g., "AUREC_UMA"

All the csv outputs into one

The combined file for all regions is also saved in the "output" folder:
https://raw.githubusercontent.com/unicef-drp/Country-and-Region-Metadata/refs/heads/main/output/all_regions_long_format.csv

It can also be created easily using an existing function:

library(data.table)
source("R/general_functions.R")
dt_all <- bind.all.output()

Easily reshaped into a wide format

Reshape the long-format output using ISO3Code ~ Region_Code. Then every row represents an ISO3Code/Country, and every region becomes a column.

dt_unicef_prog <- fread("output/UNICEF_PROG_REG_GLOBAL.csv")
dt_unicef_prog_wide <- data.table::dcast(dt_unicef_prog, ISO3Code ~ Region_Code, value.var = "Region")
head(dt_unicef_prog_wide)

Update

2025

add African Union, UNICEF programme region, SDGRC (Regional Commissions)
add WB income regions and WB regions
add UNICEF reporting (incl Kosovo, 203 countries in total)("output/UNICEF_REP_REG_GLOBAL.csv")
add SDG M49 regions ("output/UNSDG_REGION_GLOBAL.csv")