How to draw nested categorical boxplots in R using ggplot2? | Salaries | StatswithR | Arnab Hazra

Here we explain how to generate a presentation/publication-quality nested categorical boxplots in R/R-studio using ggplot2. The codes for the steps explained in the video are as follows. Copy and paste them into R, run them one-by-one and try to understand what each argument is doing.
#datascience #datavisualization #visualization #ggplot2 #tidyverse #nestedboxplot #categoricalboxplot #boxplot #nested #categorical #Salaries #Salariesdataset #rstudio #rcoding
library(carData)
data("Salaries", package = "carData")
library(writexl)
write_xlsx(Salaries[ , c(1, 2, 6)], path = "Salaries.xlsx")
library(readxl)
data = read_excel("Salaries.xlsx")
head(data)
str(data)
library(ggplot2)
p = ggplot(data = data, aes(x=rank, y=salary / 1e3, fill=discipline)) + geom_boxplot()
ggsave(p, filename = "ggplot_nestbox1.pdf", height = 8, width = 8)
ranks = c("AsstProf", "AssocProf", "Prof")
data$rank = factor(data$rank, levels = ranks)
data$discipline = factor(data$discipline, levels = c("A", "B"))
data$salary = data$salary / 1e3
str(data)
p = ggplot(data = data, aes(x=rank, y=salary, fill=discipline)) + geom_boxplot()
ggsave(p, filename = "ggplot_nestbox2.pdf", height = 8, width = 8)
levels(data$rank) = c("Assistant Prof.", "Associate Prof.", "Full Prof.")
levels(data$discipline) = c("Theoretical", "Applied")
p = ggplot(data = data, aes(x=rank, y=salary, fill=discipline)) + geom_boxplot()
ggsave(p, filename = "ggplot_nestbox3.pdf", height = 8, width = 8)
p = ggplot(data = data, aes(x=rank, y=salary, fill=discipline)) +
stat_boxplot(geom = "errorbar") + geom_boxplot()
ggsave(p, filename = "ggplot_nestbox4.pdf", height = 8, width = 8)
p = ggplot(data = data, aes(x=rank, y=salary, fill=discipline)) +
stat_boxplot(geom = "errorbar") + geom_boxplot() +
ggtitle("Salary comparison") + xlab(NULL) + ylab("Nine-month salary (in thousands USD)")
ggsave(p, filename = "ggplot_nestbox5.pdf", height = 8, width = 8)
p0 = ggplot(data = data, aes(x=rank, y=salary, fill=discipline)) +
stat_boxplot(geom = "errorbar") + geom_boxplot() +
ggtitle("Salary comparison") + xlab(NULL) + ylab("Nine-month salary (in thousands USD)") +
theme(axis.text=element_text(size=18),
axis.title=element_text(size=18),
plot.title = element_text(size=20, hjust = 0.5))
ggsave(p0, filename = "ggplot_nestbox6.pdf", height = 8, width = 8)
p = p0 + theme(legend.text=element_text(size=18),
legend.title = element_text(size=18, hjust = 0.5),
legend.key.height = unit(2,"cm"),
legend.key.width = unit(2,"cm"),
legend.position = c(0.2, 0.8)) +
guides(fill=guide_legend(title="Discipline"))
ggsave(p, filename = "ggplot_nestbox7.pdf", height = 8, width = 8)
cols = rep(c("#E69F00", "#56B4E9"), length(levels(data$rank)))
p = p0 + theme(legend.text=element_text(size=18),
legend.title = element_text(size=18, hjust = 0.5),
legend.key.height = unit(2,"cm"),
legend.key.width = unit(2,"cm"),
legend.position = c(0.2, 0.8)) +
guides(fill=guide_legend(title="Discipline")) +
scale_fill_manual(values=cols)
ggsave(p, filename = "ggplot_nestbox8.pdf", height = 8, width = 8)

[@user-zm4td4wn8i]

This was so explicit and helpful. Thank you very much!!!

[@valentinouedraogo5298]

Interesting tutorial. Thanks a lot

[@rosebaulch2666]

Omg this video is so good and clear and easy to follow!!

[@statswithr602]

Thank you, Rose!

[@MrRahulukey]

Thank you so much.

[@letsseethenature2423]

This is a nice video to understanding boxplot making. I need to know how to change box and whisker's line width.

[@aaronm9491]

thank you so much

[@salmaasghar1674]

answer to my every querry solution to my all problems thanks

[@vincenzo4259]

Thanks

[@antonynjogu4721]

Thank you for this tutorial.

[@statswithr602]

Thank you for your comment Antony 😊

[@antonynjogu4721]

@@statswithr602 is it possible to add a regression line (abline). Kindly, help.

[@truongphu7407]

Thanks for your great tutorial, can you tell us how to reorder the "discipline" after all mentioned steps have been done"?

[@adeizamomoh2939]

Hi! Great and easy to follow lesson. I would like to use standard deviation as error bars, how can I do this?

[@vbhidalgo2508]

Hi. Instead of two plot per rank. How can tou you do that in one discipline. And plot it using facet?

[@eaaepro]

Hello sir, what kind of color code is that in the graphics?

[@letterinoscala8670]

It's very clear. How can you visualize more than 2 categorical factor?

[@karenbrodersen3847]

Thank you so much for your tutorial, I can finally make it work properly!! Can I ask you, how can I add more values to my y-axis? In your y-axis it ends at '200' and in my own it ends at '1.2', but my highest value is around '1.4', how would I add this value at the top of my y-axis? Thank you again very much, you saved my day! :)

[@statswithr602]

Thank you for your comment Karen, you can add a layer ylim(l,u), where l is the lowest value you want on the y axis and u is the highest value you want on the y axis... Good luck

[@karenbrodersen3847]

@@statswithr602 Thank you, that worked! :))