#Upload the data
teachers<-read.csv("teachers_generated.csv")

#Inspect the data structure
str(teachers)
nrow(teachers)
ncol(teachers)
colnames(teachers)

#Change the column names
colnames(teachers)<-c("country", "salary", "occupation", "life_satisfy")
colnames(teachers)

#Check unique values
unique(teachers$occupation)
unique(teachers$country)

#indexing [row, column]
teachers[2,3]
teachers[2,"country"]
teachers[,2] #all rows, column 2
teachers[3,] #only the third row, all columns

#Check distributions
hist(teachers$salary[which(teachers$country=="France")])
hist(teachers$salary[which(teachers$country=="Germany")])
hist(teachers$salary[which(teachers$country=="Italy")])

#Calculate medians for each country 

median(teachers$salary[which(teachers$country=="France")])
median(teachers$salary[which(teachers$country=="Germany")])
median(teachers$salary[which(teachers$country=="Italy")])


#The same but using the function aggregate (var1 ~ var2, data= , FUN= )
aggregate(salary ~ country, data = teachers, FUN=median)
#you can use different ways to do the same; morre compact script

#IQR
aggregate(salary ~ country, data = teachers, FUN=IQR)

#Build a boxplot (y ~ x)
boxplot(teachers$salary ~ teachers$country)

#Giving titles
boxplot(teachers$salary ~ teachers$country, main="Salary difference among teachers",
        xlab="Countries", ylab="Salary [euro/month]")


#Build a scatter plot - plot (y ~ x)
plot(teachers$life_satisfy ~ teachers$salary)

#Change points color
?plot
plot(teachers$life_satisfy~teachers$salary, col="hotpink2",bg="hotpink2",pch=21)

#Add titles.
plot(teachers$life_satisfy~teachers$salary, col="hotpink2",bg="hotpink2",pch=21,
     main="Dependence of satisfaction from salary", xlab="Salary [euro/month]",
     ylab="Satisfaction")


#Changing dots colors depending on the country
table_col<-data.frame("country"=c("France", "Germany", "Italy"),
                      "color"=c("red", "blue", "green"))
color<-table_col$color[match(teachers$country, table_col$country)]

plot(teachers$life_satisfy~teachers$salary, col=color,bg=color,pch=21,
     main="Dependence of satisfaction from salary", xlab="Salary [euro/month]",
     ylab="Satisfaction")

#about ggplot
install.packages("ggplot2")
library("ggplot2")


#Help-> cheat sheets->Data Visualization with ggplot2
#Build the same scatter plot using ggplot
ggplot(data=teachers, aes(x=salary, y=life_satisfy))+
  geom_point()


#Adding the geometric layer "geom_point"
ggplot(data=teachers, aes(x=salary, y=life_satisfy))+
  geom_point(aes(color=country))

#Changing the dots color depending on the country+size of the dots depending on the salary size
ggplot(data=teachers, aes(x=salary, y=life_satisfy))+
  geom_point(aes(color=country, size=salary))


#Add titles with a layer "labs"
ggplot(data=teachers, aes(x=salary, y=life_satisfy))+
  geom_point(aes(color=country, size=salary))+
  labs(x = "Salary [euro/month]", 
       y = "Life satisfaction", 
       title ="Dependence of life satisfaction from salary") 


#Build a box plot

ggplot(data=teachers,aes(x=country, y=salary) )+
  geom_boxplot()


#Add colors and titles

ggplot(data=teachers,aes(x=country, y=salary) )+
  geom_boxplot(aes(color=country))+
  labs(x = "Country", 
       y = "Salary [euro/month]", 
       title ="Salary difference among teachers")


#Take the legend away
ggplot(data=teachers,aes(x=country, y=salary) )+
  geom_boxplot(aes(color=country), show.legend = FALSE)+
  labs(x = "Country", 
       y = "Salary [euro/month]", 
       title ="Salary difference among teachers")