Plot 3 boxplots into one plot

Question

Having three dataframes like this:

Dataframe 1:

df1 <- structure(list(company = structure(c(3L, 5L, 1L, 2L, 4L, 3L, 
5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 
1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 
2L, 4L), .Label = c("amazon", "bsd", "google", "so", "yahoo"), class = "factor"), 
    period = structure(c(2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
    1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 
    2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
    1L), .Label = c("after", "before"), class = "factor"), val = c(0.262776250810038, 
    0.187917588433778, 0.697682733346741, 0.158756228911086, 
    0.378985944448169, 0.249033541149918, 0.157828875332395, 
    0.762575137985743, 0.148767625304462, 0.394985586914259, 
    0.268776116734822, 0.177604969721347, 0.694811289133204, 
    0.160510379656321, 0.389823691090702, 0.280675292172242, 
    0.181169135885232, 0.655493731983643, 0.177839601349691, 
    0.387633795892829, 0.257949543026971, 0.169661013161717, 
    0.665359433308753, 0.149795535295301, 0.384002592120846, 
    0.244474983799245, 0.162231011597506, 0.650253625617304, 
    0.147493910750598, 0.424582690889589, 0.291490692945409, 
    0.241190141002436, 0.622555920538089, 0.215134857321624, 
    0.383108757346205, 0.25750262563965, 0.230989251636835, 0.708699246944202, 
    0.193749860338316, 0.427264195213515)), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -40L))

Second:

df2 <- structure(list(company = structure(c(3L, 5L, 1L, 2L, 4L, 3L, 
5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 
1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 
2L, 4L), .Label = c("amazon", "bsd", "google", "so", "yahoo"), class = "factor"), 
    period = structure(c(2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
    1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 
    2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
    1L), .Label = c("after", "before"), class = "factor"), val = c(0.262776250810038, 
    0.187917588433778, 0.697682733346741, 0.158756228911086, 
    0.378985944448169, 0.249033541149918, 0.157828875332395, 
    0.762575137985743, 0.148767625304462, 0.394985586914259, 
    0.268776116734822, 0.177604969721347, 0.694811289133204, 
    0.160510379656321, 0.389823691090702, 0.280675292172242, 
    0.181169135885232, 0.655493731983643, 0.177839601349691, 
    0.387633795892829, 0.257949543026971, 0.169661013161717, 
    0.665359433308753, 0.149795535295301, 0.384002592120846, 
    0.244474983799245, 0.162231011597506, 0.650253625617304, 
    0.147493910750598, 0.424582690889589, 0.291490692945409, 
    0.241190141002436, 0.622555920538089, 0.215134857321624, 
    0.383108757346205, 0.25750262563965, 0.230989251636835, 0.708699246944202, 
    0.193749860338316, 0.427264195213515)), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -40L))

Third:

df3 <- structure(list(company = structure(c(3L, 5L, 1L, 2L, 4L, 3L, 
5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 
1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 2L, 4L, 3L, 5L, 1L, 
2L, 4L), .Label = c("amazon", "bsd", "google", "so", "yahoo"), class = "factor"), 
    period = structure(c(2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
    1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 
    2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
    1L), .Label = c("after", "before"), class = "factor"), val = c(0.262776250810038, 
    0.187917588433778, 0.697682733346741, 0.158756228911086, 
    0.378985944448169, 0.249033541149918, 0.157828875332395, 
    0.762575137985743, 0.148767625304462, 0.394985586914259, 
    0.268776116734822, 0.177604969721347, 0.694811289133204, 
    0.160510379656321, 0.389823691090702, 0.280675292172242, 
    0.181169135885232, 0.655493731983643, 0.177839601349691, 
    0.387633795892829, 0.257949543026971, 0.169661013161717, 
    0.665359433308753, 0.149795535295301, 0.384002592120846, 
    0.244474983799245, 0.162231011597506, 0.650253625617304, 
    0.147493910750598, 0.424582690889589, 0.291490692945409, 
    0.241190141002436, 0.622555920538089, 0.215134857321624, 
    0.383108757346205, 0.25750262563965, 0.230989251636835, 0.708699246944202, 
    0.193749860338316, 0.427264195213515)), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -40L))

We can create a boxplot for every data frame using this:

library(stringr)
library(ggplot2)

df1 = pivot_longer(df1, everything())
df2 = cbind(df1,str_split_fixed(df1$name,"_",2))
colnames(df2)[3:4]=c("Company","Time")

ggplot(df2, aes(x=Company, y=value, fill=Time)) + geom_boxplot() + coord_flip()

How is it possible to have the box plots of the three dataframes in the same plot next to other and have the same y-axis (only one time) and have title that "this it df1", "this it df2" and "this it df3"? x-axis will be 3 time.

Answer 1

How about with a facet_wrap?

First we'll combine all the data into one nice data.frame with bind_rows, setting a new column based on the data.frame name with the .id = argument.

Then we'll plot the combined data and use facet_wrap to split by the new DF factor.

library(dplyr)
library(ggplot2)
combined.df <- bind_rows(df1,df2,df3,.id = "DF") %>%
  mutate(DF = paste0("DataFrame_",DF))
ggplot(combined.df, aes(x=company, y=val)) +
  geom_boxplot(aes(fill = period)) + coord_flip() +
  facet_wrap(~ DF)  

Answer 2

You can use the package ggpubr to achieve it. Like this:

library(ggplot2)
library(ggpubr)

colnames(df1) <- c('Company','Time','Value')
colnames(df2) <- c('Company','Time','Value')
colnames(df3) <- c('Company','Time','Value')

graf1 <- ggplot(df1,aes(x=Company, y=Value,fill=Time)) +
  geom_boxplot() +
  ggtitle('Dataset 1') +
  guides(fill=FALSE)
graf2 <- ggplot(df2,aes(x=Company, y=Value,fill=Time)) +
  geom_boxplot() +
  ggtitle('Dataset 2') +
  theme(axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank()) +
  guides(fill=FALSE)

graf3 <- ggplot(df3,aes(x=Company, y=Value,fill=Time)) +
  geom_boxplot() +
  ggtitle('Dataset 3') +
  theme(axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

ggarrange(graf1,graf2,graf3,nrow = 1,ncol = 3)