Spain_maps.Rmd

---
title: "Mortality rate in Spain 2020"
authors: "Natalia Garcia, Ángela Gómez, Lucía Martín, Ana Solbas"
output:
  html_document:
    df_print: paged
---

### Libraries

```{r}
library(sf)
library(mapSpain) # https://doi.org/10.5281/zenodo.5366622, https://ropenspain.github.io/mapSpain/
library(tidyverse)
library(ggplot2)
```

## DATA

```{r}
mydata<-read.csv("Provincias_datos.csv")
mydata
```

```{r}
#Getting the total number of deaths by province differentiating between sex
totaldata_sex <- mydata[mydata$Mes == "Total",]
totaldata_sex <- totaldata_sex[totaldata_sex$Lugar != "Total",]
totaldata_sex

# getting the total number of deaths by province without sex
totaldata <- totaldata_sex[totaldata_sex$Sexo == "Total",] 
totaldata 

# Summary of totaldata 
total_sum <- totaldata[,c(1,4)]
total_sum
```

```{r}
# Getting province codes and adding them to the summary table
cpro <- esp_dict_region_code(total_sum$Lugar, origin = "text", destination = "cpro")
total_sum$cpro <- cpro
total_sum <- total_sum[order(total_sum$cpro),] # order by cpro
total_sum
```

```{r}

# Extracting population in 2019 from the MapSpain package
census <- mapSpain::pobmun19

codelist <- mapSpain::esp_codelist

census <-
  unique(merge(census, codelist[, c("cpro", "codauto")], all.x = TRUE))

census_prov <-
  aggregate(cbind(pob19, men, women) ~ cpro, data = census, sum)

# Calculating the percentage of women per province
census_prov$porc_women <- census_prov$women / census_prov$pob19
census_prov$porc_women_lab <-
  paste0(round(100 * census_prov$porc_women, 2), "%")

# Calculate death rate per 100.000 habitants
death_census <- merge(total_sum, census_prov)
death_census$porc_muerte <- death_census$Total / death_census$pob19
death_census$porc_muerte_lab <-
  paste0(round(1000 * death_census$porc_muerte, 2))
death_census # table with all usefull information


```

## Mortality rate in 2020 (per 1000 population)

```{r}

# Plotting spain map divided by provinces
esp_prov <- esp_get_prov_siane() 
esp_prov <- merge(esp_prov, death_census)
can_prov <- esp_get_can_provinces()
Can <- esp_get_can_box()

ggplot(esp_prov) +
  geom_sf(aes(fill = porc_muerte), # coloring provinces by death rate
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Reds", rev = TRUE), # defining gradient of colors
    n.breaks = 10,
    labels = function(x) {         # printing formatted labels
      sprintf("%1.0f",1000 * x)
    },
    guide = guide_legend(title = "Mortality rate")
  ) +
   labs(      # labels and tittles
    title = "Mortality rate in Spain (per 1000 population)",
    subtitle = "INE Data (2020)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./mortalityrateSpain.svg", width = 12, height = 6, device='svg', dpi=700)

```

## Total number of deaths per province in Spain in 2020

```{r}

# Plotting spain map divided by provinces
esp_prov <- esp_get_prov_siane()
esp_prov <- merge(esp_prov, death_census)
can_prov <- esp_get_can_provinces()
Can <- esp_get_can_box()

ggplot(esp_prov) +
  geom_sf(aes(fill = Total), # coloring provinces by death number
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Reds", rev = TRUE),  # defining gradient of colors
    n.breaks = 10,
    labels = function(x) {   # printing formatted labels
      sprintf("%1.0f", x)
    },
    guide = guide_legend(title = "Number of deaths")
  ) +
   labs(   # labels and tittles
    title = "Total Number of deaths in Spain",
    subtitle = "INE Data (2020)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./totaldeathSpain.svg", width = 12, height = 6, device='svg', dpi=700)

```

## Population density in Spain, 2019

```{r}

# Plotting spain map divided by provinces
esp_prov <- esp_get_prov_siane()
esp_prov <- merge(esp_prov, census_prov)
can_prov <- esp_get_can_provinces()
Can <- esp_get_can_box()

ggplot(esp_prov) +
  geom_sf(aes(fill = pob19), # coloring provinces by population
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Blues", rev = TRUE),  # defining gradient of colors
    n.breaks = 10,
    labels = function(x) {  # printing formatted labels
      sprintf("%1.0f", x)
    },
    guide = guide_legend(title = "Population")
  ) +
   labs(  # labels and tittles
    title = "Population density in Spain",
    subtitle = "INE Data (2019)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./populationSpain.svg", width = 12, height = 6, device='svg', dpi=700)

```

## "Mortality rate in Spain (per 1000 population) in January 2020

```{r}
# Getting total number of deaths by province in January
enero <- mydata[mydata$Mes == "Enero",]
enero <- enero[enero$Lugar != "Total",]
enero <- enero[enero$Sexo == "Total",]
enero <- enero[,c(1,4)]

# Getting province codes
cpro <- esp_dict_region_code(enero$Lugar, origin = "text", destination = "cpro")
enero$cpro <- cpro
enero <- enero[order(enero$cpro),] # order by cpro

# Calcculating death rates in January
rate_enero <- merge(enero, census_prov)
rate_enero$porc_muerte <- rate_enero$Total / rate_enero$pob19
rate_enero$porc_muerte_lab <-
  paste0(round(1000 * rate_enero$porc_muerte, 2), "%")
rate_enero
```

```{r}
# Plotting spain map divided by provinces
esp_prov <- esp_get_prov_siane() 
esp_prov <- merge(esp_prov, rate_enero)
can_prov <- esp_get_can_provinces()
Can <- esp_get_can_box()

ggplot(esp_prov) +
  geom_sf(aes(fill = porc_muerte),  # coloring provinces by death rate
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Greens", rev = TRUE), # defining gradient of colors
    n.breaks = 6,
    limits = c(0, 1.3/1000),
    labels = function(x) {   # printing formatted labels
      sprintf("%s",1000 * x)
    },
    guide = guide_legend(title = "Mortality rate")
  ) +
   labs(    # lables and titles
    title = "Mortality rate in Spain (per 1000 population) in January 2020",
    subtitle = "INE Data (2020)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./mortalityJanuary.svg", width = 12, height = 6, device='svg', dpi=700)

```

## Mortality rate in Spain (per 1000 population) in April 2020

```{r}
# Getting total number of deaths by province in April
abril <- mydata[mydata$Mes == "Abril",]
abril <- abril[abril$Lugar != "Total",]
abril <- abril[abril$Sexo == "Total",]
abril <- abril[,c(1,4)]

# Getting province codes
cpro <- esp_dict_region_code(abril$Lugar, origin = "text", destination = "cpro")
abril$cpro <- cpro
abril <- abril[order(abril$cpro),] # order by cpro

# Calculating death rates in April
rate_abril <- merge(abril, census_prov)
rate_abril$porc_muerte <- rate_abril$Total / rate_abril$pob19
rate_abril$porc_muerte_lab <-
  paste0(round(1000 * rate_abril$porc_muerte, 2), "%")
rate_abril

```

```{r}

# Plotting spain map divided by provinces
esp_prov <- esp_get_prov_siane() 
esp_prov <- merge(esp_prov, rate_abril)
can_prov <- esp_get_can_provinces()
Can <- esp_get_can_box()

ggplot(esp_prov) +
  geom_sf(aes(fill = porc_muerte),  # coloring provinces by death rate
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Greens", rev = TRUE), # defining gradient of colors
    n.breaks = 6,
    limits = c(0, 1.3/1000),
    labels = function(x) {   # printing formatted labels
      sprintf("%s",1000 * x)
    },
    guide = guide_legend(title = "Mortality rate")
  ) +
   labs(    # lables and titles
    title = "Mortality rate in Spain (per 1000 population) in April 2020",
    subtitle = "INE Data (2020)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./mortalityApril.svg", width = 12, height = 6, device='svg', dpi=700)

```

## Mortality in Spain per Autonomous Community

```{r}
ccaa_mort <-read.csv("CCAA_DR_Mortality_Population.csv")

codauto <- esp_dict_region_code(ccaa_mort$CCAA, origin = "text", destination = "codauto")
ccaa_mort$codauto <- codauto
ccaa_mort<- ccaa_mort[order(ccaa_mort$codauto),] # order by cpro
ccaa_mort
```

### Mortality rate per Autonomous Comunity

```{r}
CCAA <- esp_get_ccaa()
CCAA <- merge(CCAA, ccaa_mort)
Can <- esp_get_can_box()
can_prov <- esp_get_can_provinces()

ggplot(CCAA) +
  geom_sf(aes(fill =DR),  # coloring provinces by death rate
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Reds", rev = TRUE), # defining gradient of colors
    n.breaks = 10,
    labels = function(x) {   # printing formatted labels
      sprintf("%1.0f", x)
    },
    guide = guide_legend(title = "Mortality rate")
  ) +
   labs(    # lables and titles
    title = "Mortality rate in Spain (per 1000 population) in 2020 per A.C.",
    subtitle = "INE Data (2020)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./mortalityrateSpainccaa.svg", width = 12, height = 6, device='svg', dpi=700)

```

### Total number of deaths per Autonomous Community

```{r}
CCAA <- esp_get_ccaa()
CCAA <- merge(CCAA, ccaa_mort)
Can <- esp_get_can_box()
can_prov <- esp_get_can_provinces()

ggplot(CCAA) +
  geom_sf(aes(fill = sum.Total),  # coloring provinces by death rate
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10, "Reds", rev = TRUE), # defining gradient of colors
    n.breaks = 10,
    labels = function(x) {   # printing formatted labels
      sprintf("%1.0f", x)
    },
    guide = guide_legend(title = "Number of deaths")
  ) +
   labs(    # lables and titles
    title = "Total Number of deaths in Spain per A.C.",
    subtitle = "INE Data (2020)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./totaldeathSpainccaa.svg", width = 12, height = 6, device='svg', dpi=700)

```

## Average age in 2019 rate per Province

```{r}
prov_age <-read.csv("AvgAgeProvince2019.csv")

colnames(prov_age) <- c("prov.shortname.es", "Age")
```

Plotting

```{r}

# Plotting spain map divided by provinces
esp_prov <- esp_get_prov_siane() 
esp_prov <- merge(esp_prov, prov_age, by="prov.shortname.es")
can_prov <- esp_get_can_provinces()
Can <- esp_get_can_box()

ggplot(esp_prov) +
  geom_sf(aes(fill = Age),  # coloring provinces by death rate
    color = "grey70",
    linewidth = .3, 
  ) +
  geom_sf(data = Can, color = "grey70") +
  scale_fill_gradientn(
    colors = hcl.colors(10,"BrwnY", rev = TRUE), # defining gradient of colors
    n.breaks = 6,
    limits = c(35, 51),
    labels = function(x) {   # printing formatted labels
      sprintf("%s years", x)
    },
    guide = guide_legend(title = "Averate Age")
  ) +
   labs(    # lables and titles
    title = "Averate Age per province in Spain at the end of 2019",
    subtitle = "INE Data (2019)"
  ) +
  geom_sf(data = can_prov) +
  geom_sf(data = Can) +
  theme_void() +
  theme(legend.position = c(0.1, 0.6))
ggsave(file="./ageprovince.svg", width = 12, height = 6, device='svg', dpi=700)

```