How to combine multiple ggplot geom_col() into one graph? - r

Question

Question
I am trying to combine three ggplot geom_bar into one geom_bar plot utilising dodge so I can visually compare data across two categorical and one numeric variables. What am I doing wrong?
Individual graphs work
Each graph works on it's own (with formatting issues) and I've been following answers on SO like How to overlay two geom_bar? but I'm not understanding what's needed to be done.
ONE <- ggplot(Ireland, aes(TargetGroup, FirstDosePC))+geom_bar(stat = 'identity',width = 0.8, fill = "green") +
facet_grid(.~Vaccine) +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))+
labs(title="1st Dose Ireland by Group & Vaccine Type",
caption = "(ECDC, 2021)",
x="Target Groups over 18",
y="First Dose Administered")
TWO <- ggplot(Italy, aes(TargetGroup, FirstDosePC))+ geom_bar(stat = 'identity',width = 0.8, fill = "blue") +
facet_grid(.~Vaccine) +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))+
labs(title="1st Dose Italy by Group & Vaccine Type",
caption = "(ECDC, 2021)",
x="Target Groups over 18",
y="First Dose Administered")
THREE <- ggplot(Latvia, aes(TargetGroup, FirstDosePC))+geom_bar(stat = 'identity',width = 0.8, fill = "red") +
facet_grid(.~Vaccine) +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))+
labs(title="1st Dose Latvia by Group & Vaccine Type",
caption = "(ECDC, 2021)",
x="Target Groups over 18",
y="First Dose Administered")
An example of failed code
My coding attempts look close to this but it seems to fail - I don't understand why. I am hoping to learn how to add three graphs together with labels and to use dodge
OneTwo <- ONE + geom_bar(FDPercent=Italy, aes(TargetGroup, FirstDose))+ geom_bar(stat = 'identity',width = 0.8, fill = "blue") +
facet_grid(.~Vaccine) +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))+
labs(title="1st Dose Italy by Group & Vaccine Type",
caption = "(ECDC, 2021)",
x="Target Groups over 18",
y="First Dose Administered")
My individual graphs look like this
The graph type I'm aiming for
and what I am aiming for is something like this but breaking it out by vaccine type to stretch my learning, etc (source https://towardsdatascience.com/track-covid-19-data-yourself-with-r-eb3e641cd4b3)
My raw data comes from
data <- read.csv("https://opendata.ecdc.europa.eu/covid19/vaccine_tracker/csv/data.csv", na.strings = "", fileEncoding = "UTF-8-BOM")
and is manipulated to test out R functions that has left me with a dataframe called FDPercent with a numeric column called FirstDosePC (Percentage of 1st Dose per country population) that is linked to Country with 30 EU countries (ISO 3166-1-alpha-2 categorical data) and 10 TargetGroup types (categorical) in the data frame.
> dput(head(FDPercent,3))
structure(list(Country = c("AT", "AT", "AT"), NumberDosesReceived = c(0L,
0L, 61425L), NumberDosesExported = c(0L, 0L, 0L), FirstDose = c(0L,
0L, 87L), FirstDoseRefused = c(NA_integer_, NA_integer_, NA_integer_
), SecondDose = c(0L, 0L, 0L), UnknownDose = c(0L, 0L, 0L), TargetGroup = c("Age18_24",
"Age18_24", "Age18_24"), Vaccine = c("UNK", "AZ", "COM"), Population = c(8901064L,
8901064L, 8901064L), Date = structure(c(18624, 18624, 18624), class = "Date"),
FirstDosePC = c("0.0000", "0.0000", "0.0010")), row.names = 21:23, class = "data.frame")
> str(FDPercent)
'data.frame': 116532 obs. of 12 variables:
$ Country : chr "AT" "AT" "AT" "AT" ...
$ NumberDosesReceived: int 0 0 61425 0 0 0 61425 0 0 0 ...
$ NumberDosesExported: int 0 0 0 0 0 0 0 0 0 0 ...
$ FirstDose : int 0 0 87 0 0 0 1299 0 0 0 ...
$ FirstDoseRefused : int NA NA NA NA NA NA NA NA NA NA ...
$ SecondDose : int 0 0 0 0 0 0 0 0 0 0 ...
$ UnknownDose : int 0 0 0 0 0 0 0 0 0 0 ...
$ TargetGroup : chr "Age18_24" "Age18_24" "Age18_24" "Age18_24" ...
$ Vaccine : chr "UNK" "AZ" "COM" "MOD" ...
$ Population : int 8901064 8901064 8901064 8901064 8901064 8901064 8901064 8901064 8901064 8901064 ...
$ Date : Date, format: "2020-12-28" "2020-12-28" "2020-12-28" "2020-12-28" ...
$ FirstDosePC : chr "0.0000" "0.0000" "0.0010" "0.0000" ...

Answer 1

With help from #kat in the comments - changed from geom_bar() to geom_col() and dropped the third variable
Ireland <- subset(FDPercent, Country == "IE") #contructed a subset by country
Italy <- subset(FDPercent, Country == "IT")
Latvia <- subset(FDPercent, Country == "LV")
ONE1 <- ggplot(Italy, aes(Date, as.numeric(FirstDosePC))) +
geom_col(fill = "red", alpha = 1, width = 7) + theme_minimal(base_size = 8) +
xlab(NULL) + ylab(NULL) + scale_x_date(date_labels = "%Y/%m/%d") #reduced the theme formating
OneTwo <- ONE1 + geom_col(data=Ireland, aes(Date,
as.numeric(FirstDosePC)),
fill="Green", alpha = 1,width = 5)
OneTwoThree <- OneTwo + geom_col(data=Latvia, aes(Date,
as.numeric(FirstDosePC)),
fill="black", alpha = 1, width = 2)
OneTwoThree + labs(title="Ireland, Italy, & Latvia - First Dose comparision", #added labels to the data
subtitle="Using First Dose Delivered per day as a percentage of population",
caption = "(ECDC, 2021)",
x="Date Administered",
y="% Population treated")