R Shiny - Groups are not represented as specified in plot - r

Question

I've grouped some Data about the titanic the following way:
priceMutate <- mutate(titanic, PriceGroup = ifelse (Fare < 51,
'0 - 50',
ifelse(Fare >=51 & Fare < 101,
'51-100',
ifelse(Fare >= 101 & Fare < 151,
'101-150',
ifelse(Fare >= 151 & Fare < 201,
'151-200',
ifelse(Fare >= 201 & Fare < 251,
'201-250',
ifelse(Fare >= 251 & Fare < 301,
'251-300',
ifelse(Fare >= 301 & Fare < 351,
'301-350',
ifelse(Fare >= 351 & Fare < 401,
'351-400',
ifelse(Fare >= 401 & Fare < 451,
'401-450',
ifelse(Fare > 450,
'451+','?')))))))))))
"Fare" is the price payed for a ticket for the titanic. I've chosen steps of 50$.
Now here is my problem:
I've made a plot that shows the chance of survival regarding the price of the tickets:
output$ex15 <- renderPlot({
ggplot(priceMutate,
aes(x = PriceGroup,
fill = Status)) +
geom_bar(position = "fill")+
ggtitle("Überlebenschancen nach Preis des Tickets (gruppiert)")+
theme(plot.title = element_text(size = 16))+
scale_fill_manual(values = c("grey24", "snow"))+
labs(y= "Anzahl")
})
However this plot mixes up the groups I made and does not show the "?" for the not-available data!
Can anyone see a problem/mistake that I've made?
Here is a link to my dataset: https://drive.google.com/file/d/1xsIfkv1464etX23O0J9y35CviK0mKYQl/view?usp=sharing
Thank you a lot :)

Answer 1

As already mentioned by #YBS in the comments at least for your example data there is no observation which will be assigned a "?" as all values are in the range 0 to 512 and there are no missings.
Concerning your second issue, as you recoded the Fare column as a character your PriceGroups will be ordered alphabetically by default. And alphabetically a string starting with a 4 like 451+ comes before a string starting with a 5 like 51-100. If you want the categories to be ordered you have to convert to a factor with the levels set according to your desired order. This for example could be achieved via the cut function which makes it easy to recode a numeric to intervals and which will automatically convert to a factor. If you do that often I also would suggest to have a look at the santoku package which makes it even easier set nice labels.
Finally, instead of using your data I created a minimal reproducible example by using some fake random example data to mimic your real data:
library(shiny)
library(tidyverse)
# Create fake example data
set.seed(123)
titanic <- data.frame(
PassengerId = 1:100,
Survived = sample(0:1, 100, replace = TRUE),
Fare = runif(100, 0, 512)
)
# Set breaks and labels
breaks <- c(0, seq(51, 451, 50), Inf)
labels <- paste(breaks[-length(breaks)], breaks[-1], sep = "-")
labels[length(labels)] <- "451+"
priceMutate <- titanic %>%
mutate(PriceGroup = cut(Fare, breaks = breaks, labels = labels, right = FALSE),
Status = recode(Survived, "0" = "Dead", "1" = "Survived"))
ui <- fluidPage(
plotOutput("ex15")
)
server <- function(input, output, session) {
output$ex15 <- renderPlot({
ggplot(priceMutate,
aes(x = PriceGroup,
fill = Status)) +
geom_bar(position = "fill")+
ggtitle("Überlebenschancen nach Preis des Tickets (gruppiert)")+
theme(plot.title = element_text(size = 16))+
scale_fill_manual(values = c("grey24", "snow"))+
labs(y= "Anzahl")
})
}
shinyApp(ui, server)
#>
#> Listening on http://127.0.0.1:8734