BarplotsINEDeaths.Rmd

---
title: "BarplotsINEMortalitySex"
output: html_document
date: '2022-12-27'
---

```{r setup, include=FALSE}
#--------------------- Import libraries
library(tidyr)
library(ggplot2)
library(reshape2)
library(dplyr)
```

## Import data

*Data Source*:

```{r cars}
data<-read.csv("Provincias_datos.csv", sep = ",")
# Glimpse of the dataset 
head(data)
```

# Total Deaths Number Rate Visualizations

## 1. Month Deaths Bar Chart

## Data preprocessing

Taking total cases, and month data, dropping missing values and translating month labels

```{r dp1}
datatotalmonth = data[which(data["Lugar"]=="Total"),]
datatotalmonth = datatotalmonth[which(data["Mes"]!="Total"),]
datatotalmonth = datatotalmonth[which(data["Sexo"]=="Total"),]
datatotalmonth = datatotalmonth %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs1}
datatotalmonth = datatotalmonth[,c("Mes", "Total")]

levels_order <-  c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December")
datatotalmonth[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plotmonth = melt(datatotalmonth)
colnames(plotmonth) = c("Month", "Stratification", "Deaths")
```

## Plotting

```{r pressure, echo=FALSE}
ggplot(data = data.frame(plotmonth), mapping = aes(x=Month,y=Deaths))+
  scale_x_discrete(limits = levels_order)+
  geom_bar(stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = .3)+
  geom_path(aes(x = Month, y = Deaths), size = 1.5, color="red", group = 1, linewidth = .5)+
  ylab("Total Deaths Number")+
  ggtitle("Total Deaths Number per month in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 12),
        axis.text.y = element_text(size = 10),
        axis.title=element_text(size=14))
ggsave(filename = "figures/mortality_month.png", width = 15, height = 5, device='png', dpi=700)
```

## 2. Month Sex Deaths Bar Chart

## Data preprocessing

Taking total cases, sex data and month data, dropping missing values and translating month and sex labels

```{r dp2}
datatotalmonthsex = data[which(data["Lugar"]=="Total"),]
datatotalmonthsex = datatotalmonthsex[which(data["Mes"]!="Total"),]
datatotalmonthsex = datatotalmonthsex %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs2}
datatotalmonthsex = datatotalmonthsex[,c("Mes", "Sexo","Total")]

# This time for translation of labels it is necessary to replicate the sex for each of the three sex labels per  instance
datatotalmonthsex["Sexo"] <- rep_len(c("Total", "Male", "Female"),3)

# This time for translation of labels it is necessary to replicate the months for each of the three sex labels per instance
levels_order <-  c(rep_len("January",3), rep_len("February",3), rep_len("March",3), rep_len("April",3), 
                   rep_len("May",3), rep_len("June",3), rep_len("July",3), rep_len("August",3), 
                   rep_len("September",3), rep_len("October",3), rep_len("November",3), rep_len("December",3))
datatotalmonthsex[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plotmonthsex = melt(datatotalmonthsex)
colnames(plotmonthsex) = c("Month", "Sex", "Stratification", "Deaths")
df=data.frame(plotmonthsex)
```

## Plotting

```{r 2 pressure, echo=FALSE}
monthsex <- ggplot(data = df, mapping = aes(x=Month,y=Deaths))+
  scale_x_discrete(limits = levels_order)+
  geom_bar(data=df[df$Sex=="Total", ], stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = 1.2)+
  geom_point(data=df[df$Sex!="Total", ],aes(color=Sex, x = as.factor(Month), y = Deaths, group=Sex), size = 2.5)+
  geom_path(data=df[df$Sex!="Total", ], aes(color=Sex, x = as.factor(Month), y = Deaths, group=Sex), size = 1.5, linewidth = .8)+
  ylab("Total Deaths Number")+
  ggtitle("Total Deaths Number per month in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 12),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=12))
monthsex  + scale_fill_manual("legend", values = c("Female" = "#F96666", "Male" = "#7895B2"))

ggsave(filename = "figures/mortality_pmonth_sex.png", width = 15, height = 5, device='png', dpi=700)
```

## 3. Distribution of Deaths per month boxplot

## Data preprocessing

Taking total cases, sex data and month data, dropping missing values and translating month and sex labels

```{r dp3}
datatotal = data[which(data["Lugar"]!="Total"),]
datatotal = datatotal[which(data["Mes"]!="Total"),]
datatotal = datatotal[which(data["Sexo"]=="Total"),]
datatotal = datatotal %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs3}
datatotal = datatotal[,c("Lugar", "Mes", "Total")]

# Gets number of provinces in the dataset for translation of month values through repetition
numprov <- length(unique(datatotal["Lugar"])[,1])

# This time for translation of labels it is necessary to replicate the months for each of the three sex labels per instance
levels_order <-  rep_len(c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December"), numprov)
datatotal[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plottotalbp = melt(datatotal)
colnames(plottotalbp) = c("Province", "Month", "Stratification","Deaths")
df=data.frame(plottotalbp)
```

## Plotting

```{r 3 pressure, echo=FALSE}
monthsex <- ggplot(data = df, mapping = aes(x=Month,y=Deaths))+
  scale_x_discrete(limits = c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December"))+
  geom_boxplot()+
  ylab("Total Deaths Number")+
  ggtitle("Total Deaths Number per month in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 12),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=12))
monthsex
ggsave(filename = "figures/mortality_distribution_month.png", width = 15, height = 5, device='png', dpi=700)
```

## 4. Plotting by province with ordered medians

```{r sbs99}

 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(Province,
                              Deaths,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$Province,
                             levels = levels(group_ordered))


```

## Plotting

```{r 4 pressure, echo=FALSE}
monthsex <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=Deaths))+
  geom_boxplot()+
  xlab("Province")+
  ylab("Total Deaths Number")+
  ggtitle("Total Deaths Number per province in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=10))
monthsex
ggsave(filename = "figures/mortality_distribution_province.png", width = 20, height = 5, device='png', dpi=700)
```

```{r 69 pressure, echo=FALSE}
dataaccumulativeprov = datatotal %>% group_by(Lugar) %>%
  summarise(sum.Total = sum(Total))



# Necessary to reformat for ggplot
plotprov_acc = melt(dataaccumulativeprov)
colnames(plotprov_acc) = c("Province", "Strat:Agg_Funct", "Deaths")
df=data.frame(plotprov_acc)


 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(Province,
                              Deaths,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$Province,
                             levels = levels(group_ordered))


#---------------------- plot

monthccaa <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=Deaths))+
  geom_bar(stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = 0.6)+
  ylab("Total Deaths Number")+
  xlab("Province")+
    ggtitle("Total Deaths Number per province in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=12))
monthccaa
ggsave(filename = "figures/mortality_sum_province.png", width = 16, height = 7, device='png', dpi=700)
```

## 5. Grouping by Autonomic Community

```{r impccaa}

map_ccaa_prov <- read.csv("data/ccaa_provincia.csv")
colnames(map_ccaa_prov) <- c("CCAA","Lugar")
dataccaa <- merge(data,map_ccaa_prov,by="Lugar")

```

## Data preprocessing

Taking total cases, sex data and month data, dropping missing values and translating month and sex labels

```{r dp4}
dataccaa = dataccaa[which(data["Lugar"]!="Total"),]
dataccaa = dataccaa[which(data["Mes"]!="Total"),]
dataccaa = dataccaa[which(data["Sexo"]=="Total"),]
dataccaa = dataccaa %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs4}
dataccaa = dataccaa[,c("CCAA", "Mes", "Total")]

# Gets number of provinces in the dataset for translation of month values through repetition
numprov <- length(unique(datatotal["Lugar"])[,1])

# This time for translation of labels it is necessary to replicate the months for each of the three sex labels per instance
levels_order <-  rep_len(c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December"), numprov)
datatotal[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plotccaabp = melt(dataccaa)
colnames(plotccaabp) = c("CCAA", "Month","Stratification","Deaths")
df=data.frame(plotccaabp)
```

```{r sbs31}

 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(CCAA,
                              Deaths,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$CCAA,
                             levels = levels(group_ordered))


```

## Plotting

```{r 5 pressure, echo=FALSE}
monthccaa <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=Deaths))+
  geom_boxplot()+
  ylab("Total Deaths Number")+
  xlab("Autonomous Community")+
  ggtitle("Total Deaths Number per A.C. in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=10))
monthccaa
ggsave(filename = "figures/mortality_distribution_ccaa.png", width = 15, height = 5, device='png', dpi=700)
```

```{r 6 pressure, echo=FALSE}
dataaccumulativeccaa = dataccaa %>% group_by(CCAA) %>%
  summarise(sum.Total = sum(Total))

# save info to write to csv later
dataCCAAMort<-dataaccumulativeccaa

# Necessary to reformat for ggplot
plotccaa_acc = melt(dataaccumulativeccaa)
colnames(plotccaa_acc) = c("CCAA", "Strat:Agg_Funct", "Deaths")
df=data.frame(plotccaa_acc)


 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(CCAA,
                              Deaths,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$CCAA,
                             levels = levels(group_ordered))


#---------------------- plot

monthccaa <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=Deaths))+
  geom_bar(stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = 0.6)+
  ylab("Total Deaths Number")+
  xlab("Autonomous Community")+
    ggtitle("Total Deaths Number per A.C. in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=12))
monthccaa
ggsave(filename = "figures/mortality_sum_ccaa.png", width = 16, height = 7, device='png', dpi=700)
```

# Death Rate Per 1,000 inhabitants visualization

In reference to the population of provinces in the end of 2019

```{r}
death_census <- read.csv("Death_Census_Perc.csv")

# Merging on province to get populations of men, women and total populations in 2019
datapob19 <- death_census[, c('Lugar','men','women','pob19')]
dataDR <- merge(data,datapob19,by="Lugar")

```

Calculating Death Rate per month and province

```{r}
# Calculating Death Rate DR per population group
dataDR$DR <- NULL

# Creating variables indexing rows by stratification population (Total,men,women)
BOOLTOTAL <- which(dataDR['Sexo']=='Total')
dataDR[BOOLTOTAL,'DR'] <- ( dataDR[BOOLTOTAL,'Total'] / dataDR[BOOLTOTAL,'pob19'] ) * 1000

BOOLMEN <- which(dataDR['Sexo']=='Hombres')
dataDR[BOOLMEN,'DR'] <- ( dataDR[BOOLMEN,'Total'] / dataDR[BOOLMEN, 'men'] ) * 1000

BOOLWOMEN <- which(dataDR['Sexo']=='Mujeres')
dataDR[BOOLWOMEN,'DR'] <- ( dataDR[BOOLWOMEN,'Total'] / dataDR[BOOLWOMEN,'women']) * 1000
```

## 1. Month Death Rate Bar chart

## Data preprocessing

```{r}

datatotalmonth = dataDR[which(data["Lugar"]=="Total"),]
datatotalmonth = datatotalmonth[which(data["Mes"]!="Total"),]
datatotalmonth = datatotalmonth[which(data["Sexo"]=="Total"),]
datatotalmonth = datatotalmonth %>% drop_na()
datatotalmonth = datatotalmonth[,c("Mes", "DR")]

levels_order <-  c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December")
datatotalmonth[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plotmonth = melt(datatotalmonth)
colnames(plotmonth) = c("Month", "Stratification", "DeathRate")


```

## Plotting

```{r 10 pressure, echo=FALSE}
ggplot(data = data.frame(plotmonth), mapping = aes(x=Month,y=DeathRate))+
  scale_x_discrete(limits = levels_order)+
  geom_bar(stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = .3)+
  geom_path(aes(x = Month, y = DeathRate), size = 1.5, color="red", group = 1, linewidth = .5)+
  ylab("Death Rate (per 1,000 inhabitants)")+
  ggtitle("Death Rate per month in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 12),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=9))
ggsave(filename = "figures/deathR_month.png", width = 15, height = 5, device='png', dpi=700)
```

## 2. Month Sex Deaths Bar Chart

## Data preprocessing

Taking total cases, sex data and month data, dropping missing values and translating month and sex labels

```{r dp12}
datatotalmonthsex = dataDR[which(data["Lugar"]=="Total"),]
datatotalmonthsex = datatotalmonthsex[which(data["Mes"]!="Total"),]
datatotalmonthsex = datatotalmonthsex %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs12}
datatotalmonthsex = datatotalmonthsex[,c("Mes", "Sexo","DR")]

# This time for translation of labels it is necessary to replicate the sex for each of the three sex labels per  instance
datatotalmonthsex["Sexo"] <- rep_len(c("Total", "Male", "Female"),3)

# This time for translation of labels it is necessary to replicate the months for each of the three sex labels per instance
levels_order <-  c(rep_len("January",3), rep_len("February",3), rep_len("March",3), rep_len("April",3), 
                   rep_len("May",3), rep_len("June",3), rep_len("July",3), rep_len("August",3), 
                   rep_len("September",3), rep_len("October",3), rep_len("November",3), rep_len("December",3))
datatotalmonthsex[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plotmonthsex = melt(datatotalmonthsex)
colnames(plotmonthsex) = c("Month", "Sex", "Stratification", "DeathRate")
df=data.frame(plotmonthsex)
```

## Plotting

```{r 23 pressure, echo=FALSE}
monthsex <- ggplot(data = df, mapping = aes(x=Month,y=DeathRate))+
  scale_x_discrete(limits = levels_order)+
  geom_bar(data=df[df$Sex=="Total", ], stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = 1.2)+
  geom_point(data=df[df$Sex!="Total", ],aes(color=Sex, x = as.factor(Month), y = DeathRate, group=Sex), size = 2.5)+
  geom_path(data=df[df$Sex!="Total", ], aes(color=Sex, x = as.factor(Month), y = DeathRate, group=Sex), size = 1.5, linewidth = .8)+
  ylab("Death Rate (per 1,000 inhabitants)")+
  ggtitle("Death Rate per month in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 12),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=9))
monthsex  + scale_fill_manual("legend", values = c("Female" = "#F96666", "Male" = "#7895B2"))

ggsave(filename = "figures/deathR_pmonth_sex.png", width = 15, height = 5, device='png', dpi=700)
```

## 3. Distribution of Deaths per month boxplot

## Data preprocessing

Taking total cases, sex data and month data, dropping missing values and translating month and sex labels

```{r dp15}
datatotal = dataDR[which(data["Lugar"]!="Total"),]
datatotal = datatotal[which(data["Mes"]!="Total"),]
datatotal = datatotal[which(data["Sexo"]=="Total"),]
datatotal = datatotal %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs13}
datatotal = datatotal[,c("Lugar", "Mes", "DR")]

# Gets number of provinces in the dataset for translation of month values through repetition
numprov <- length(unique(datatotal["Lugar"])[,1])

# This time for translation of labels it is necessary to replicate the months for each of the three sex labels per instance
levels_order <-  rep_len(c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December"), numprov)
datatotal[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plottotalbp = melt(datatotal)
colnames(plottotalbp) = c("Province", "Month", "Stratification","DeathRate")
df=data.frame(plottotalbp)
```

## Plotting

```{r 13 pressure, echo=FALSE}
monthsex <- ggplot(data = df, mapping = aes(x=Month,y=DeathRate))+
  scale_x_discrete(limits = c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December"))+
  geom_boxplot()+
  ylab("Death Rate (per 1,000 inhabitants)")+
  ggtitle("Death Rate per month in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 12),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=9))
monthsex
ggsave(filename = "figures/deathR_distribution_month.png", width = 15, height = 5, device='png', dpi=700)
```

## 4. Plotting by province with ordered medians

```{r sbs39}

 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(Province,
                              DeathRate,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$Province,
                             levels = levels(group_ordered))


```

## Plotting

```{r 14 pressure, echo=FALSE}
monthsex <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=DeathRate))+
  geom_boxplot()+
  ylab("Death Rate (per 1,000 inhabitants)")+
  ggtitle("Death Rate per province in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=9))
monthsex
ggsave(filename = "figures/deathR_distribution_province.png", width = 20, height = 5, device='png', dpi=700)
```

```{r 59 pressure, echo=FALSE}
datatotal <- data[which(data["Sexo"]=="Total"),]

# Group by Province + sum deaths
dataaccumulativeprov = datatotal %>% group_by(Lugar) %>%
  summarise(sum.Total = sum(Total))

datapobprov = dataDR %>% group_by(Lugar)

dataDRplot <- merge(datapobprov,dataaccumulativeprov,by="Lugar")
dataDRplot$DR <- ( dataDRplot$sum.Total / dataDRplot$pob19 ) * 1000
dataDRplot <- dataDRplot[,c('Lugar','DR')]



# Necessary to reformat for ggplot
plotprov_acc = melt(dataDRplot)
colnames(plotprov_acc) = c("Province", "Strat:Agg_Funct", "DeathRate")
df=data.frame(plotprov_acc)


 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(Province,
                              DeathRate,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$Province,
                             levels = levels(group_ordered))
```

```{r 49 pressure, echo=FALSE}

#---------------------- plot

monthccaa <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=DeathRate))+
  geom_bar(stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = 0.6)+
  ylab("Death Rate (per 1,000 inhabitants)")+
  xlab("Province")+
    ggtitle("Death Rate per province in 2020")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=12))
monthccaa
ggsave(filename = "figures/deathR_sum_province.png", width = 16, height = 7, device='png', dpi=700)
```

## 5. Grouping by Autonomic Community

```{r impccaa2}

map_ccaa_prov <- read.csv("data/ccaa_provincia.csv")
colnames(map_ccaa_prov) <- c("CCAA","Lugar")
dataccaa <- merge(dataDR,map_ccaa_prov,by="Lugar")

```

## Data preprocessing

Taking total cases, sex data and month data, dropping missing values and translating month and sex labels

```{r dp14}
dataccaa = dataccaa[which(dataccaa["Lugar"]!="Total"),]
dataccaa = dataccaa[which(dataccaa["Mes"]!="Total"),]
dataccaa = dataccaa[which(dataccaa["Sexo"]=="Total"),]
dataccaa = dataccaa %>% drop_na()
```

Subsetting and reformatting final dataframe

```{r sbs14}
dataccaa = dataccaa[,c("CCAA", "Mes", "DR")]

# Gets number of provinces in the dataset for translation of month values through repetition
numprov <- length(unique(datatotal["Lugar"])[,1])

# This time for translation of labels it is necessary to replicate the months for each of the three sex labels per instance
levels_order <-  rep_len(c("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December"), numprov)
datatotal[,"Mes"] <- levels_order

# Necessary to reformat for ggplot
plotccaabp = melt(dataccaa)
colnames(plotccaabp) = c("CCAA", "Month","Stratification","DeathRate")
df=data.frame(plotccaabp)
```

```{r sbs38}

 # Order boxes by median
group_ordered <- with(df,                      
                      reorder(CCAA,
                              DeathRate,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$CCAA,
                             levels = levels(group_ordered))


```

## Plotting

```{r 16 pressure, echo=FALSE}
monthccaa <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=DeathRate))+
  geom_boxplot()+
  ggtitle("Death Rate per A.C. in 2020")+
  ylab("Death Rate (per 1,000 inhabitants)")+
  xlab("Autonomous Community")+
  theme(plot.title = element_text(size=16))+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=9))
monthccaa
ggsave(filename = "figures/deathR_distribution_ccaa.png", width = 15, height = 5, device='png', dpi=700)
```

```{r sbs88}

datatotal <- data[which(data["Sexo"]=="Total"),]
dataccaa <- merge(datatotal,map_ccaa_prov,by="Lugar")

# Group by CCAA + sum deaths
dataaccumulativeccaa = dataccaa %>% group_by(CCAA) %>%
  summarise(sum.Total = sum(Total))

# Group by CCAA + sum pop19
datapobccaa <- merge(map_ccaa_prov,datapob19,by="Lugar")
datapobccaa= datapobccaa %>% group_by(CCAA) %>%
  summarise(sum.pob19 = sum(pob19))

dataDR <- merge(datapobccaa,dataaccumulativeccaa,by="CCAA")
dataDR$DR <- ( dataDR$sum.Total / dataDR$sum.pob19 ) * 1000
dataDR <- dataDR[,c('CCAA','DR')]


# Necessary to reformat for ggplot
plotccaa_acc = melt(dataDR)
colnames(plotccaa_acc) = c("CCAA", "Strat:Agg_Funct","DeathRate")
df=data.frame(plotccaa_acc)


# save info to write to csv later
dataCCAADR<-dataDR


```

```{r 17 pressure, echo=FALSE}
# Order boxes by median
group_ordered <- with(df,                      
                      reorder(CCAA,
                              DeathRate,
                              median))

# Create data with reordered group levels
datatotal_ordered <- df                             
datatotal_ordered$group <- factor(datatotal_ordered$CCAA,
                             levels = levels(group_ordered))

#---------------------- Sex-Month plot

monthccaa <- ggplot(data = datatotal_ordered, mapping = aes(x=group,y=DeathRate))+
  geom_bar(stat = "identity", position = 'dodge', color = "#7895B2", fill = "white", width = 0.6)+
    ggtitle("Death Rate per A.C. in 2020")+
  ylab("Death Rate (per 1,000 inhabitants)")+
  xlab("Autonomous Community")+
  theme(plot.title = element_text(size=16))+
  ylim(0,15)+
  theme(axis.text.x = element_text(angle=45, hjust=1, size = 11),
        axis.text.y = element_text(size = 12),
        axis.title=element_text(size=12))
monthccaa
ggsave(filename = "figures/deathR_sum_ccaa.png", width = 16, height = 7, device='png', dpi=700)
```

```{r sbs77}

# saved acc data groupby CCAA
CCAAdata <- merge(dataCCAADR, dataCCAAMort, by='CCAA')
write.csv(CCAAdata, "CCAA_DR_Mortality_Population.csv")

```