```{r message=FALSE}
setwd("C:/Users/Alain/Documents")
options(repos = c(CRAN = "https://cloud.r-project.org/"))
requiredPackages <- c("sf","RColorBrewer", "spatstat.geom", "spatstat.utils", "ggplot2",
"spatstat", "maptools", "httr","ggspatial", "viridis", "dplyr", "readxl",
"forcats","tidyr", "stringr")
for(i in requiredPackages)
{if(!require(i,character.only = TRUE)) install.packages(i)}
```Adolescent health disparities remain a major concern in sub-Saharan Africa. In the Democratic Republic of the Congo (DRC), limited healthcare infrastructure and uneven service delivery have contributed to spatially heterogeneous outcomes among youth. Recent studies emphasize the importance of spatial analytics for understanding and addressing geographic inequality in health services. This study builds upon such literature by mapping and modeling the spatial distribution of adolescents affected by health conditions in the DRC, with a special focus on gender disparities and regional intensity.
Geospatial layers: Administrative boundaries (provinces, communes, and districts), railway infrastructure, and protected areas were sourced from geoBoundaries, Natural Earth, and OpenStreetMap datasets.
Health data: A dataset from “ADO & JEUNES 2023” provided counts of affected adolescents disaggregated by gender (male/female) and province. Estimated population sizes by province were used to standardize reported values.
Data preprocessing involved renaming variables, converting to numeric types, eliminating missing values, and reshaping the data into long format for gender-disaggregated analysis. Each province was assigned a unique code, and simulated centroid coordinates were generated for visualization and point pattern analysis.
Health service delivery for adolescents in the DRC is marked by disparities in access and outcome, influenced by geography and gender. Yet, few studies have comprehensively mapped this inequality.
Map the absolute and standardized burden of adolescent health cases by province.
Identify provinces with significant gender disparities.
Evaluate spatial distribution patterns (random, clustered, or dispersed).
Determine the intensity of health needs by combining population ratios and gender proportions.
Administrative boundaries were transformed to a consistent CRS (EPSG:3857) and joined with the health dataset. Key visualizations included:
Choropleths showing total affected adolescents per province.
Faceted maps to compare male vs. female distributions.
Bivariate maps combining intensity and gender proportion.
Simulated coordinates allowed for:
Construction of unmarked and marked point patterns using the spatstat package.
Nearest-neighbor distance (NND) and pairwise distance analysis to assess spatial clustering.
Ratios of affected adolescents per 1,000 inhabitants and female proportions per province were computed. These allowed for standardized comparisons and composite metrics for policy prioritization.
This study reveals marked spatial disparities in adolescent health conditions across the DRC. Provinces such as Kinshasa, Kasaï-Oriental, and Nord-Kivu show both high absolute numbers and significant gender imbalances. Spatial point analysis indicates non-random clustering of affected cases, suggesting underlying regional factors such as urban density, health infrastructure, and sociopolitical context.
The findings underscore the necessity for spatially targeted interventions, especially in provinces with both high intensity and low female representation. Future work should incorporate temporal dynamics and more granular demographic data (e.g., age groups, urban/rural segmentation).
Keywords: DRC, adolescents, health disparities, spatial analysis, gender, choropleth, point pattern, spatstat
```{r}
banderies <- st_read("C:/Users/Alain/Downloads/province26/provinces26/Province26.shp")
banderies_nad83 <- st_transform(banderies, crs = "+proj=longlat +datum=NAD83")
library(rnaturalearth)
rdc <- ne_states(country = "Democratic Republic of the Congo", returnclass = "sf")
rdc_83<- st_transform(rdc, crs = 3857)
ggplot() +
geom_sf(rdc_83, mapping = aes(geometry=geometry))
```Reading layer `Province26' from data source
`C:\Users\Alain\Downloads\province26\provinces26\Province26.shp'
using driver `ESRI Shapefile'
Simple feature collection with 26 features and 8 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 1358728 ymin: -1501941 xmax: 3481773 ymax: 596454.8
Projected CRS: World_Mercator
```{r}
provinces <- st_read("C:/Users/Alain/Downloads/geoBoundaries-COD-ADM2-all/geoBoundaries-COD-ADM2.shp")
provinces_nad83 <- st_transform(provinces, crs = "+proj=longlat +datum=NAD83")
ggplot() +
geom_sf(provinces_nad83, mapping = aes(geometry=geometry))
crds <- st_centroid(st_make_valid(provinces_nad83))
head(crds)
plot(st_geometry(provinces_nad83)) # rysuje kontur / plotting the contour
plot(st_geometry(crds), pch=21, bg='red', add=TRUE)
```Reading layer `geoBoundaries-COD-ADM2' from data source
`C:\Users\Alain\Downloads\geoBoundaries-COD-ADM2-all\geoBoundaries-COD-ADM2.shp'
using driver `ESRI Shapefile'
Simple feature collection with 189 features and 5 fields
Geometry type: POLYGON
Dimension: XY
Bounding box: xmin: 12.20566 ymin: -13.456 xmax: 31.30522 ymax: 5.386098
Geodetic CRS: WGS 84
Simple feature collection with 6 features and 5 fields
Geometry type: POINT
Dimension: XY
Bounding box: xmin: 23.7547 ymin: 0.7354574 xmax: 30.70698 ymax: 4.471862
Geodetic CRS: +proj=longlat +datum=NAD83
shapeName shapeISO shapeID shapeGroup shapeType
1 Aba <NA> 63176286B20411552692002 COD ADM2
2 Aketi <NA> 63176286B77771623677962 COD ADM2
3 Ango <NA> 63176286B24678106401017 COD ADM2
4 Ariwara <NA> 63176286B55306571020797 COD ADM2
5 Aru <NA> 63176286B14719092717616 COD ADM2
6 Bafwasende <NA> 63176286B50174012793828 COD ADM2
geometry
1 POINT (30.23818 3.858422)
2 POINT (23.7547 2.98004)
3 POINT (26.07273 4.471862)
4 POINT (30.70698 3.136862)
5 POINT (30.53241 3.065347)
6 POINT (26.99695 0.7354574)
```{r}
library(rnaturalearth)
rdc <- ne_states(country = "Democratic Republic of the Congo", returnclass = "sf")
rdc_83<- st_transform(rdc, crs = 3857)
``````{r}
unique(rdc$name)
Kasai_Oriental <- rdc[rdc$name == "Kasaï-Oriental", ]
Kasai_Oriental <- rdc[grepl("Kasaï-Oriental", rdc$name, ignore.case = TRUE), ]
nrow(Kasai_Oriental)
Kasai_Oriental_proj <- st_transform(Kasai_Oriental, crs = 32733)
Kasai_Oriental_geom <- st_geometry(Kasai_Oriental_proj)[[1]]
Kasai_Oriental_win <- as.owin(Kasai_Oriental_geom)
``` [1] "Équateur" "Bandundu" "Kinshasa City" "Bas-Congo"
[5] "Orientale" "Sud-Kivu" "Katanga" "Nord-Kivu"
[9] "Kasaï-Occidental" "Kasaï-Oriental" "Maniema"
[1] 1```{r}
plot(Kasai_Oriental_win, main = "Observation window - Kasaï-Oriental")
ggplot() +
geom_sf(Kasai_Oriental_proj, mapping = aes(geometry=geometry))
Kasai_Oriental_proj <- st_transform(Kasai_Oriental, crs = 32733)
st_bbox(Kasai_Oriental_proj)
``` xmin ymin xmax ymax
1268838 9113480 1757329 9804526 
#st_bbox(banderies_nad83)
```{r}
crds <- st_centroid(st_make_valid(Kasai_Oriental_proj))
head(crds)
plot(st_geometry(Kasai_Oriental_proj))
plot(st_geometry(crds), pch=21, bg='red', add=TRUE)
```Simple feature collection with 1 feature and 121 fields
Geometry type: POINT
Dimension: XY
Bounding box: xmin: 1496407 ymin: 9491290 xmax: 1496407 ymax: 9491290
Projected CRS: WGS 84 / UTM zone 33S
featurecla scalerank adm1_code diss_me iso_3166_2
3789 Admin-1 states provinces lakes 5 COD-1896 1896 CD-KE
wikipedia iso_a2 adm0_sr name name_alt name_local
3789 <NA> CD 1 Kasaï-Oriental East Kasai|Kasai East <NA>
type type_en code_local code_hasc note hasc_maybe region region_cod
3789 Province Province <NA> CD.KR <NA> <NA> <NA> <NA>
provnum_ne gadm_level check_me datarank abbrev postal area_sqkm sameascity
3789 9 1 0 3 <NA> KR 0 -99
labelrank name_len mapcolor9 mapcolor13 fips fips_alt woe_id
3789 6 14 4 7 CG04 <NA> 2344980
woe_label woe_name latitude
3789 Kasai-Oriental, CD, Democratic Republic of Congo Kasaï-Oriental -4.40511
longitude sov_a3 adm0_a3 adm0_label admin
3789 24.0842 COD COD 2 Democratic Republic of the Congo
geonunit gu_a3 gn_id gn_name
3789 Democratic Republic of the Congo COD 214138 Province du Kasai-Oriental
gns_id gns_name gn_level gn_region gn_a1_code region_sub sub_code
3789 -2046899 Kasai-Oriental 1 <NA> CD.04 <NA> <NA>
gns_level gns_lang gns_adm1 gns_region min_label max_label min_zoom
3789 1 fra CG04 <NA> 6 11 6
wikidataid name_ar name_bn name_de name_en
3789 Q80953 كاساي الشرقية কাসাই-ওরিয়েন্টাল Kasaï-Oriental Kasaï-Oriental
name_es name_fr name_el name_hi name_hu
3789 Kasai Oriental Kasaï-Oriental Κασάι-Οριεντάλ कासाइ-पूर्वी प्रान्त Kelet-Kasai
name_id name_it name_ja name_ko name_nl
3789 Kasai-Oriental Kasai Orientale 東カサイ州 카사이오리앙탈 Oost-Kasaï
name_pl name_pt name_ru name_sv name_tr
3789 Kasai Wschodnie Kasaï-Oriental Восточное Касаи Kasaï-Oriental Doğu Kasai
name_vi name_zh ne_id name_he name_uk
3789 Kasai-Oriental 东开赛省 1159311343 קאסאי-אוריינטל Східне Касаї
name_ur name_fa name_zht FCLASS_ISO FCLASS_US FCLASS_FR
3789 کاسائی-مشرقی کاسای- اورینتال 东开赛省 <NA> <NA> <NA>
FCLASS_RU FCLASS_ES FCLASS_CN FCLASS_TW FCLASS_IN FCLASS_NP FCLASS_PK
3789 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
FCLASS_DE FCLASS_GB FCLASS_BR FCLASS_IL FCLASS_PS FCLASS_SA FCLASS_EG
3789 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
FCLASS_MA FCLASS_PT FCLASS_AR FCLASS_JP FCLASS_KO FCLASS_VN FCLASS_TR
3789 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
FCLASS_ID FCLASS_PL FCLASS_GR FCLASS_IT FCLASS_NL FCLASS_SE FCLASS_BD
3789 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
FCLASS_UA FCLASS_TLC geometry
3789 <NA> <NA> POINT (1496407 9491290)
```{r}
railways <- st_read("C:/Users/Alain/Downloads/congo-democratic-republic-latest-free.shp/gis_osm_railways_free_1.shp")
railways_nad83 <- st_transform(railways, crs = "+proj=longlat +datum=NAD83")
plot(railways$geometry)
district <- st_read("C:/Users/Alain/Downloads/district/District.shp")
district_nad83 <- st_transform(district, crs = "+proj=longlat +datum=NAD83")
plot(district$geometry)
```Reading layer `gis_osm_railways_free_1' from data source
`C:\Users\Alain\Downloads\congo-democratic-republic-latest-free.shp\gis_osm_railways_free_1.shp'
using driver `ESRI Shapefile'
Simple feature collection with 1900 features and 7 fields
Geometry type: LINESTRING
Dimension: XY
Bounding box: xmin: 13.43202 ymin: -13.49397 xmax: 30.71915 ymax: 3.36139
Geodetic CRS: WGS 84
Reading layer `District' from data source
`C:\Users\Alain\Downloads\district\District.shp' using driver `ESRI Shapefile'
Simple feature collection with 48 features and 8 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 1358728 ymin: -1501941 xmax: 3481773 ymax: 596454.8
Projected CRS: World_Mercator
```{r}
parc <- st_read("C:/Users/Alain/Downloads/parc/Parc.shp")
parc_nad83 <- st_transform(parc, crs = "+proj=longlat +datum=NAD83")
table(parc$NOM)
plot(parc$geometry)
```Reading layer `Parc' from data source `C:\Users\Alain\Downloads\parc\Parc.shp' using driver `ESRI Shapefile'
Simple feature collection with 44 features and 6 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 12.35354 ymin: -11.19897 xmax: 30.20467 ymax: 5.388336
Geodetic CRS: WGS 84
Domaine de chasse de Bili-Uere Domaine de Chasse de Bombo Lumene
1 1
Domaine de Chasse de Bomu Domaine de Chasse de Bushimaie
1 1
Domaine de Chasse de Gangala-na Bodio Domaine de Chasse de Luama-Katanga
1 1
Domaine de chasse de Luama-Kivu Domaine de chasse de Lubudi Sampwe
1 1
Domaine de chasse de Maika-Penge Domaine de chasse de Mangai
1 1
Domaine de chasse de Rubi-Tele Domaine de chasse de Rutshuru
1 1
Domaine de Chasse de Swa-Kibula Domaine de chasse de Tshangalele
1 1
Massif d'Itombwe Parc de la N'Sele
1 1
Parc Marin des Mangroves Parc National de Kahuzi-Biega
1 1
Parc National de Kundelungu Parc National de l'Upemba
1 1
Parc National de la Garamba Parc National de la Maiko
1 1
Parc National de la Salonga Parc National des Virunga
2 1
Réserve de Abumonbazi Réserve de biosphère de la Lufira
1 1
Réserve de biosphere de la Luki Réserve de biosphère de Yangambi
1 1
Réserve de Bomu Réserve de Epi
1 1
Réserve de faune à okapis Réserve de Lomami-Lualaba
1 1
Réserve de Mai Mpili Réserve de Maniema
1 1
Réserve de Shaba Elephant Réserve du Lac Tumba-Lediima
1 1
Réserve du Mont Kabobo Réserve du Sud Masisi
1 1
Réserve du triangle de la Ngiri Réserve forestière de Lomako-Yokokala
1 1
Reserve Naturelle de Kisimba Ikobo Reserve Naturelle de Tayna
1 1
Réserve Scientifique de Luo
1 
```{r}
plot(st_geometry(st_transform(district, crs = st_crs(provinces_nad83))))
plot(st_geometry(provinces_nad83[provinces$shapeName == "Kinshasa", ]),
add = TRUE,
col = "red")
```
```{r}
b0 <- st_read("C:/Users/Alain/Downloads/CD-KN-b7bd0eae-20250601-fr-gpkg/data/boundary-polygon.gpkg")
b2 <- st_read("C:/Users/Alain/Downloads/CD-KN-b7bd0eae-20250601-fr-gpkg/data/boundary-polygon-lvl2.gpkg")
b4 <- st_read("C:/Users/Alain/Downloads/CD-KN-b7bd0eae-20250601-fr-gpkg/data/boundary-polygon-lvl4.gpkg")
b7 <- st_read("C:/Users/Alain/Downloads/CD-KN-b7bd0eae-20250601-fr-gpkg/data/boundary-polygon-lvl7.gpkg")
b8 <- st_read("C:/Users/Alain/Downloads/CD-KN-b7bd0eae-20250601-fr-gpkg/data/boundary-polygon-lvl8.gpkg")
b9 <- st_read("C:/Users/Alain/Downloads/CD-KN-b7bd0eae-20250601-fr-gpkg/data/boundary-polygon-lvl9.gpkg")
target_crs <- 3857
b0 <- st_transform(b0, crs = target_crs)
b2 <- st_transform(b2, crs = target_crs)
b4 <- st_transform(b4, crs = target_crs)
b7 <- st_transform(b7, crs = target_crs)
b8 <- st_transform(b8, crs = target_crs)
b9 <- st_transform(b9, crs = target_crs)
boundaries <- rbind(b0, b2, b4, b7, b8, b9)
nb_col <- length(unique(boundaries$NAME))
palette <- colorRampPalette(brewer.pal(9, "Set3"))(nb_col)
```Reading layer `boundary-polygon' from data source
`C:\Users\Alain\Downloads\CD-KN-b7bd0eae-20250601-fr-gpkg\data\boundary-polygon.gpkg'
using driver `GPKG'
Simple feature collection with 234 features and 25 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 15.12706 ymin: -5.032536 xmax: 16.53412 ymax: -3.927611
Geodetic CRS: WGS 84
Reading layer `boundary-polygon-lvl2' from data source
`C:\Users\Alain\Downloads\CD-KN-b7bd0eae-20250601-fr-gpkg\data\boundary-polygon-lvl2.gpkg'
using driver `GPKG'
Simple feature collection with 1 feature and 25 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 15.12706 ymin: -5.032536 xmax: 16.53412 ymax: -3.927611
Geodetic CRS: WGS 84
Reading layer `boundary-polygon-lvl4' from data source
`C:\Users\Alain\Downloads\CD-KN-b7bd0eae-20250601-fr-gpkg\data\boundary-polygon-lvl4.gpkg'
using driver `GPKG'
Simple feature collection with 1 feature and 25 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 15.12706 ymin: -5.032536 xmax: 16.53412 ymax: -3.927611
Geodetic CRS: WGS 84
Reading layer `boundary-polygon-lvl7' from data source
`C:\Users\Alain\Downloads\CD-KN-b7bd0eae-20250601-fr-gpkg\data\boundary-polygon-lvl7.gpkg'
using driver `GPKG'
Simple feature collection with 24 features and 25 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 15.12982 ymin: -5.032536 xmax: 16.53412 ymax: -3.927611
Geodetic CRS: WGS 84
Reading layer `boundary-polygon-lvl8' from data source
`C:\Users\Alain\Downloads\CD-KN-b7bd0eae-20250601-fr-gpkg\data\boundary-polygon-lvl8.gpkg'
using driver `GPKG'
Simple feature collection with 197 features and 25 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 15.20606 ymin: -4.440094 xmax: 15.42763 ymax: -4.296267
Geodetic CRS: WGS 84
Reading layer `boundary-polygon-lvl9' from data source
`C:\Users\Alain\Downloads\CD-KN-b7bd0eae-20250601-fr-gpkg\data\boundary-polygon-lvl9.gpkg'
using driver `GPKG'
Simple feature collection with 11 features and 25 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: 15.21345 ymin: -4.398089 xmax: 15.37933 ymax: -4.320823
Geodetic CRS: WGS 84```{r}
ggplot(boundaries) +
geom_sf(aes(fill = NAME), color = "black", size = 0.2, show.legend = FALSE) +
scale_fill_manual(values = palette) +
labs(title = "Communes, quartiers et districts de Kinshasa",
subtitle = "Chaque entité est représentée par une couleur distincte") +
annotation_scale(location = "bl") +
annotation_north_arrow(location = "tl", which_north = "true",
style = north_arrow_fancy_orienteering()) +
theme_minimal() +
# 🏷️ Affichage des noms des entités
geom_sf_text(aes(label = NAME), size = 2.5, color = "black", check_overlap = TRUE)
```
```{r}
commune_zoom <- boundaries %>%
filter(NAME %in% c("Gombe", "Limete", "Kintambo", "Bandalungwa"))
# Create a manual color palette for the 4 targeted communes
palette_zoom <- setNames(c("red", "blue", "green", "orange"),
c("Gombe", "Limete", "Kintambo", "Bandalungwa"))
ggplot(commune_zoom) +
geom_sf(aes(fill = NAME), color = "black", size = 0.2, show.legend = FALSE) +
scale_fill_manual(values = palette_zoom) +
labs(
title = "Zoom on Selected Communes of Kinshasa",
subtitle = "Communes: Gombe, Limete, Kintambo, Bandalungwa"
) +
annotation_scale(location = "bl") +
annotation_north_arrow(location = "tl", which_north = "true",
style = north_arrow_fancy_orienteering()) +
theme_minimal() +
geom_sf_text(aes(label = NAME), size = 2.5, color = "black", check_overlap = TRUE)
```
```{r}
# 📥 Charger les données adolescents affectes par les Infections carabines
ado <- read_excel("C:/Users/Alain/Downloads/ADO&JEUNES 2023.xlsx")
str(ado)
```tibble [28 × 4] (S3: tbl_df/tbl/data.frame)
$ Indicateur: chr [1:28] NA "Somme de Valeur" "Étiquettes de lignes" "Bas-Uele" ...
$ (Tous) : chr [1:28] NA "Étiquettes de colonnes" "Féminin" "13993" ...
$ ...3 : chr [1:28] NA NA "Masculin" "10896" ...
$ ...4 : chr [1:28] NA NA "Total général" "24889" ...```{r}
ado_large <- ado %>%
rename(
Province = Indicateur,
Femme = `(Tous)`,
Homme = `...3`,
Total_general = `...4`
) %>%
filter(!is.na(Province)) %>%
mutate(
Femme = as.numeric(Femme),
Homme = as.numeric(Homme),
Total_general = as.numeric(Total_general)
) %>%
drop_na(Femme, Homme, Total_general) %>% # ✅ Supprimer les lignes avec au moins un NA
distinct() %>%
arrange(Province)
# Affichage final
print(ado_large)
glimpse(ado_large)
# Transformation en format long avec gestion des doublons
ado_long <- ado_large %>%
distinct() %>% # Élimine les doublons complets
pivot_longer(
cols = c(Femme, Homme),
names_to = "Gender",
values_to = "valeur"
) %>%
mutate(
Gender = as_factor(Gender) %>% # Convertir en facteur
fct_relevel("Femme", "Homme") %>% # Ordonner les niveaux
fct_recode("Female" = "Femme", "Male" = "Homme"), # Renommer en anglais
.after = Province # Placer la colonne Gender après Province
) %>%
distinct(Province, Gender, .keep_all = TRUE) %>% # Éviter les doublons par combinaison Province-Gender
select(Province, Gender, valeur, Total_general) # Sélection des colonnes
# 👀 Afficher le résultat
glimpse(ado_long)
View(ado_long)
ado_clean <- ado_long[complete.cases(ado_long), ]
```# A tibble: 25 × 4
Province Femme Homme Total_general
<chr> <dbl> <dbl> <dbl>
1 Bas-Uele 13993 10896 24889
2 Haut-Katanga 2435551 2124191 4559742
3 Haut-Lomami 111958 35086 147044
4 Haut-Uele 80677 54107 134784
5 Ituri 40843 36467 77310
6 Kasaï 139262 112649 251911
7 Kasaï-Central 78996 66261 145257
8 Kasaï-Oriental 520251 315834 836085
9 Kinshasa 207106 233067 440173
10 Kongo-Central 86683 77415 164098
# ℹ 15 more rows
Rows: 25
Columns: 4
$ Province <chr> "Bas-Uele", "Haut-Katanga", "Haut-Lomami", "Haut-Uele", …
$ Femme <dbl> 13993, 2435551, 111958, 80677, 40843, 139262, 78996, 520…
$ Homme <dbl> 10896, 2124191, 35086, 54107, 36467, 112649, 66261, 3158…
$ Total_general <dbl> 24889, 4559742, 147044, 134784, 77310, 251911, 145257, 8…
Rows: 50
Columns: 4
$ Province <chr> "Bas-Uele", "Bas-Uele", "Haut-Katanga", "Haut-Katanga", …
$ Gender <fct> Female, Male, Female, Male, Female, Male, Female, Male, …
$ valeur <dbl> 13993, 10896, 2435551, 2124191, 111958, 35086, 80677, 54…
$ Total_general <dbl> 24889, 24889, 4559742, 4559742, 147044, 147044, 134784, …```{r}
pop_total_Rdc <- c(
"Bas-Uele" = 1419000,
"Équateur" = 1856000,
"Haut-Katanga" = 5378000,
"Haut-Lomami" = 2842000,
"Haut-Uele" = 2614000,
"Ituri" = 4392000,
"Kasaï" = 3199000,
"Kasaï-Central" = 3743000,
"Kasaï-Oriental" = 3145000,
"Kinshasa" = 17071000,
"Kongo-Central" = 6365000,
"Kwango" = 2416000,
"Kwilu" = 6149000,
"Lomami" = 2842000,
"Lualaba" = 3138000,
"Mai-Ndombe" = 2482000,
"Mongala" = 2358000,
"Nord-Kivu" = 8103000,
"Nord-Ubangi" = 1482000,
"Sankuru" = 2478000,
"Sud-Kivu" = 6565000,
"Sud-Ubangi" = 2614000,
"Tanganyika" = 3561000,
"Tshopo" = 3113000,
"Tshuapa" = 1887000
)
# 1. Converting Vector to Data Frame in R
pop_data <- data.frame(
Province = names(pop_total_Rdc),
pop_total_Rdc = pop_total_Rdc,
row.names = NULL
)
View(pop_data)
# 2. Data Merging with ado_clean
# Adding the "Adolescents per 1,000 Inhabitants" Column
``````{r}
pop_data <- pop_data %>%
mutate(Province = str_trim(str_to_title(Province))) # Standardizing Names in the Dataset
ado_clean <- ado_clean %>%
mutate(Province = str_trim(str_to_title(Province)))
# Name Harmonization
#Merging the Two Data Frames on the "Province" Column
merged_data <- full_join(pop_data, ado_clean, by = "Province") %>%
distinct() # Supprimer les doublons
# Data Structure Verification After Merging
glimpse(merged_data)
# Displaying a Data Preview
head(merged_data)
View(merged_data)
```Rows: 53
Columns: 5
$ Province <chr> "Bas-Uele", "Bas-Uele", "Équateur", "Équateur", "Haut-Ka…
$ pop_total_Rdc <dbl> 1419000, 1419000, 1856000, 1856000, 5378000, 5378000, 28…
$ Gender <fct> Female, Male, Female, Male, Female, Male, Female, Male, …
$ valeur <dbl> 13993, 10896, 12622, 16048, 2435551, 2124191, 111958, 35…
$ Total_general <dbl> 24889, 24889, 28670, 28670, 4559742, 4559742, 147044, 14…
Province pop_total_Rdc Gender valeur Total_general
1 Bas-Uele 1419000 Female 13993 24889
2 Bas-Uele 1419000 Male 10896 24889
3 Équateur 1856000 Female 12622 28670
4 Équateur 1856000 Male 16048 28670
5 Haut-Katanga 5378000 Female 2435551 4559742
6 Haut-Katanga 5378000 Male 2124191 4559742```{r}
province_code <- c(
"Kinshasa" = 10, "Kongo-Central" = 20, "Kwango" = 302, "Kwilu" = 303, "Mai-Ndombe" = 305,
"Tshuapa" = 402, "Mongala" = 403, "Nord-Ubangi" = 405, "Sud-Ubangi" = 406, "Équateur" = 407,
"Tshopo" = 502, "Haut-Uele" = 503, "Bas-Uele" = 504, "Ituri" = 505, "Haut-Lomami" = 61,
"Lomami" = 62, "Kasaï" = 63, "Kasaï-Central" = 704, "Kasaï-Oriental" = 705, "Sankuru" = 706,
"Maniema" = 707, "Tanganyika" = 802, "Haut-Katanga" = 803, "Lualaba" = 804, "Sud-Kivu" = 904,
"Nord-Kivu" = 903
)
# Updating merged_data with Province Codes
#Adding the "code" Column Using cbind()
merged_data <- cbind(merged_data, code = province_code[merged_data$Province])
#Removing Rows with NA Values to Prevent Errors
merged_data <- merged_data[complete.cases(merged_data), ]
``````{r}
ado_ratios <- merged_data %>%
# Step 1: Keeping Only Unique Province-Gender Combinations
group_by(Province, Gender) %>%
filter(row_number() == 1) %>%
ungroup() %>%
group_by(Province, Gender) %>%
mutate(
ratio_per_1000 = (valeur / first(Total_general)) * 1000
) %>%
ungroup()
stopifnot(
"There are still duplicates" = !any(duplicated(ado_ratios[, c("Province", "Gender")]))
)
``````{r}
glimpse(ado_ratios)
View(ado_ratios)
sum(is.na(ado_ratios))
banderies_nad83 <- banderies_nad83 %>% mutate(code = as.numeric(CODE_INS)*1000)
```Rows: 44
Columns: 7
$ Province <chr> "Bas-Uele", "Bas-Uele", "Équateur", "Équateur", "Haut-K…
$ pop_total_Rdc <dbl> 1419000, 1419000, 1856000, 1856000, 5378000, 5378000, 2…
$ Gender <fct> Female, Male, Female, Male, Female, Male, Female, Male,…
$ valeur <dbl> 13993, 10896, 12622, 16048, 2435551, 2124191, 111958, 3…
$ Total_general <dbl> 24889, 24889, 28670, 28670, 4559742, 4559742, 147044, 1…
$ code <dbl> 504, 504, 407, 407, 803, 803, 61, 61, 503, 503, 505, 50…
$ ratio_per_1000 <dbl> 562.2162, 437.7838, 440.2511, 559.7489, 534.1423, 465.8…
[1] 0```{r}
# 1. Data Preparation
# Join the geographic data with the values
carte_data <- banderies_nad83 %>%
left_join(ado_ratios, by = c("CODE_INS" = "code")) %>%
group_by(NOM) %>%
summarise(
valeur_totale = sum(valeur, na.rm = TRUE),
geometry = first(geometry)
) %>%
st_as_sf()
``````{r}
# 2. Visualization
ggplot() +
# Base map with total intensity
geom_sf(data = carte_data,
aes(fill = valeur_totale),
color = "white", size = 0.2) +
# Points by gender (overlay)
geom_sf(data = ado_ratios %>%
left_join(banderies_nad83, by = c("code" = "CODE_INS")) %>%
st_as_sf(),
aes(geometry = geometry,
color = Gender,
size = valeur),
alpha = 0.7) +
# Color scales
scale_fill_viridis_c(option = "D", name = "Total intensity") +
scale_color_manual(values = c("Male" = "#FF6B6B", "Female" = "#4ECDC4")) +
scale_size_continuous(range = c(2, 10), name = "Value by gender") +
# Theme
theme_minimal() +
labs(title = "Adolescents affected in DRC by province and gender",
caption = "Source: Your data")
```
```{r}
ggplot() +
geom_sf(data = banderies_nad83, fill = "grey90", color = "white") +
geom_sf(data = ado_ratios %>%
left_join(banderies_nad83, by = c("code" = "CODE_INS")) %>%
st_as_sf(),
aes(fill = valeur, geometry = geometry),
color = "white", size = 0.2) +
facet_wrap(~Gender) +
scale_fill_viridis_c(option = "D", name = "Absolute value") +
theme_void() +
labs(title = "Distribution by gender and province")
```
```{r}
ggplot() +
geom_sf(data = banderies_nad83, fill = "grey90", color = "white") +
geom_sf(data = ado_ratios %>%
left_join(banderies_nad83, by = c("code" = "CODE_INS")) %>%
st_as_sf(),
aes(fill = valeur, geometry = geometry),
color = "white", size = 0.2) +
# Province names in red
geom_sf_text(data = banderies_nad83,
aes(label = NOM, geometry = geometry),
color = "red", # Red color
size = 3,
check_overlap = TRUE) +
facet_wrap(~Gender) +
scale_fill_viridis_c(option = "D", name = "Absolute value") +
theme_void() +
labs(title = "Distribution by gender and province")
```
```{r}
carte_combi <- ado_ratios %>%
group_by(code) %>%
summarise(
prop_female = sum(valeur[Gender == "Female"])/sum(valeur),
valeur_tot = sum(valeur)
) %>%
left_join(banderies_nad83, by = c("code" = "CODE_INS")) %>%
st_as_sf()
ggplot() +
geom_sf(data = carte_combi,
aes(fill = valeur_tot, color = prop_female),
size = 0.5) +
scale_fill_viridis_c(option = "D", name = "Intensité totale") +
scale_color_gradient2(low = "#FF6B6B", mid = "white", high = "#4ECDC4",
midpoint = 0.5, name = "% Femmes") +
theme_minimal()
```
```{r}
ggplot() +
geom_sf(data = banderies_nad83, fill = "grey90", color = "white") +
geom_sf(
data = ado_ratios %>%
left_join(banderies_nad83, by = c("code" = "CODE_INS")) %>%
st_as_sf(),
aes(fill = valeur, geometry = geometry),
color = "white", size = 0.2
) +
# Facets
facet_wrap(~Gender, ncol = 2) +
# Single color scale
scale_fill_viridis_c(option = "D", name = "Valeur absolue") +
# Theme
theme_void() +
labs(title = "Répartition genrée des adolescents affectés")
```
```{r}
carte_bivariee <- ado_ratios %>%
group_by(code) %>%
summarise(
prop_female = sum(valeur[Gender == "Female"]) / sum(valeur),
valeur_tot = sum(valeur)
) %>%
left_join(banderies_nad83, by = c("code" = "CODE_INS")) %>%
st_as_sf()
ggplot() +
geom_sf(
data = carte_bivariee,
aes(fill = prop_female, alpha = valeur_tot),
color = "white", size = 0.3
) +
scale_fill_gradient2(
low = "#E41A1C", mid = "#F7F7F7", high = "#377EB8",
midpoint = 0.5, name = "Percentage of Females"
) +
scale_alpha_continuous(range = c(0.3, 0.9), name = "Total Intensity") +
theme_void() +
labs(title = "Intensity and Gender Distribution by Province")
```
```{r}
# 1. 📥 Load and clean the adolescent health data
ado <- read_excel("C:/Users/Alain/Downloads/ADO&JEUNES 2023.xlsx") %>%
rename(
Province = Indicateur,
Female = `(Tous)`,
Male = `...3`,
Total_general = `...4`
) %>%
filter(!is.na(Province)) %>%
mutate(
Female = as.numeric(Female),
Male = as.numeric(Male),
Total_general = as.numeric(Total_general)
) %>%
drop_na(Female, Male, Total_general)
# 2. 🌍 Add simulated coordinates for provinces (e.g., approximate centroids)
set.seed(123)
ado$lon <- runif(nrow(ado), min = 21, max = 31)
ado$lat <- runif(nrow(ado), min = -13, max = 5)
# 3. 🎯 Select a 20% random sample of provinces
set.seed(456)
ado$u <- runif(nrow(ado))
ado_small <- ado %>% filter(u <= 0.2)
# 4. 🔄 Convert to sf (projected geographic coordinate system)
ado_small.sf <- st_as_sf(ado_small, coords = c("lon", "lat"), crs = 4326)
ado_small.sf <- st_transform(ado_small.sf, crs = 3857)
# 5. 🗺️ Define the spatial window (based on observed coordinates)
W <- as.owin(st_bbox(ado_small.sf))
rsc <- 1000 # rescaling factor (1 unit = 1 km)
# 6. 🔘 Create spatial patterns
# Unmarked pattern
pattern.um <- ppp(
x = st_coordinates(ado_small.sf)[,1],
y = st_coordinates(ado_small.sf)[,2],
window = W
)
pattern.um <- rescale(pattern.um, rsc, "km")
plot(pattern.um, main = "Affected adolescents (unmarked pattern)")
# Marked pattern: pivot_longer to split by gender
ado_long <- ado_small %>%
pivot_longer(cols = c(Female, Male), names_to = "Gender", values_to = "value")
ado_long.sf <- st_as_sf(ado_long, coords = c("lon", "lat"), crs = 4326) %>%
st_transform(crs = 3857)
pattern.m <- ppp(
x = st_coordinates(ado_long.sf)[,1],
y = st_coordinates(ado_long.sf)[,2],
window = W,
marks = as.factor(ado_long.sf$Gender)
)
pattern.m <- rescale(pattern.m, rsc, "km")
plot(pattern.m, main = "Affected adolescents by gender (marked pattern)")
# 7. 📏 Distance analysis
between_dist <- pairdist(pattern.um)
between_dist_df <- as.data.frame(between_dist)
between_dist_df$min_d <- apply(between_dist_df, 1, function(x) sort(x)[2])
first_neib <- nndist(pattern.um)
# 8. 🗺️ Display map of DRC with case intensity
rdc <- ne_states(country = "Democratic Republic of the Congo", returnclass = "sf")
rdc_83 <- st_transform(rdc, crs = 3857)
# Join total cases to map
carte_rdc <- left_join(rdc_83, ado, by = c("name" = "Province"))
# 9. 🖼️ Choropleth visualization
ggplot() +
geom_sf(data = carte_rdc, aes(fill = Total_general), color = "white") +
scale_fill_viridis_c(name = "Reported cases") +
labs(
title = "Distribution of affected adolescents in DRC",
subtitle = "Total number of cases per province",
caption = "Source: ADO & JEUNES 2023"
) +
theme_minimal()
```
