knitr::opts_chunk$set(echo = TRUE, cache = TRUE)

Packages

Loading required packages:

Setting global theme:

# theme_set(theme_classic())

Data

Loading the built in iris data:

data(iris)

Loading the student survey data:

student <- readxl::read_excel("Data/StudentSurveyData.xlsx")

Scatter Plot

ggplot(data = iris) + 
  geom_point(mapping = aes(x = Sepal.Length, y = Sepal.Width)) +
  theme_classic()

ggplot(data = iris) +
  aes(x = Sepal.Length, y = Sepal.Width) +
  geom_point() 

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_point(color = "tomato1")

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_point(size = 3)

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_point(shape = 16)

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width)) +
  geom_point(shape = "diamond filled")

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width, col = Species)) +
  geom_point()

ggplot(data = iris, aes(x = Sepal.Length, y = Sepal.Width, col = Species)) +
  geom_point() + 
  labs(x = "Sepal Length", y = "Sepal Width", col = "Species", 
       title = "Scatter Plot of Sepal Length vs Width")

Histogram

ggplot(iris, aes(x = Sepal.Length)) +
  geom_histogram(binwidth = 1, fill = "lightblue", col = "white")

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length), bins = 10, fill = "lightblue", col = "black") 

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length), bins = 10, fill = "lightblue", col = "white") 

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), bins = 10, col = "white")

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), bins = 10, col = "white", alpha = 0.6)

Extra

res <- sample(1:100, 10)
res
##  [1] 32 65 57 21 86 82 70 56 18  4
sum(res)
## [1] 491

The total of the two numbers are 491.

Reduce Gap between Plot and Axis

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  coord_cartesian(expand = FALSE)

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  scale_y_continuous(
    breaks = seq(-10, 30, by=5),
    expand = expansion(
      mult = c(-0.1, 1),      # expands upper portion of the plot by 20%
      add = c(10, 0)
      )        # increases gap at the bottom portion by 10 unit
    )

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  scale_y_continuous(expand = expansion(add = c(0, 5))) +
  scale_x_continuous(expand = expansion(add = c(0, 0)))

Facet

facet_wrap

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1)

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1, scales = "free")

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1, scales = "free_y")

facet_grid

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_grid(rows = vars(Species))

ggplot(data = student) +
  geom_histogram(aes(x = GPA, fill = Employment), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_grid(rows =  vars(Employment), cols = vars(Gender))

ggplot(data = student) +
  geom_histogram(aes(x = GPA, fill = Employment), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_grid(rows =  vars(Employment), cols = vars(Class))

student %>% 
  mutate(Class = factor(Class, 
                        levels = c("Freshman","Sophomore", "Junior","Senior"))) %>% 
  ggplot() +
  geom_histogram(aes(x = Spending, fill = Employment), bins = 10, col = "white", alpha = 0.5) +
  facet_grid(rows =  vars(Employment), cols = vars(Class))

Theme

Built in Themes

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1) +
  theme_classic()

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1) +
  theme_bw()

Themes from other packages

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1) + 
  theme_calc() 

p1 <- ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1) +
  labs(
    title = "Histogram of Sepal Length by Species",
    x = "Sepal Length",
    y = "Frequency",
    fill = "Species",
    subtitle = "Using Facet and Other Customizations",
    caption = "Data: Iris"
  )

# ggThemeAssistGadget(p1)
p1 + theme(
  axis.text = element_text(family = "Times",
    size = 17, angle = 45), 
  panel.background = element_rect(fill = "hotpink4"),
    plot.background = element_rect(fill = "antiquewhite3"))
## Warning in grid.Call(C_stringMetric, as.graphicsAnnot(x$label)): font family
## not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database
## Warning in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
## font family not found in Windows font database

Manually Changing Color

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), 
                 bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1) +
  scale_fill_manual(values = c("setosa" = "#6C1C80", "versicolor" = "#30A19C", "virginica" = "#123B96"))

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), bins = 10, col = "white", alpha = 1) +
  facet_wrap(vars(Species), ncol = 1) +
  scale_fill_brewer(palette = "Set2")

ggplot(data = iris) +
  geom_histogram(aes(x = Sepal.Length, fill = Species), bins = 10, col = "white", alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1) +
  scale_fill_hue(
    l = 80, c = 100,   # adjust luminosity and chroma
    h = c(90, 360)     # adjust range of hues
  )

Density Plot

ggplot(data = iris) +
  geom_density(aes(x = Sepal.Length, fill = Species), alpha = 0.5)

ggplot(data = iris) +
  geom_density(aes(x = Sepal.Length, fill = Species), alpha = 0.5) +
  facet_wrap(vars(Species), ncol = 1)

Histogram + Density Plot

ggplot(data = iris, aes(x = Sepal.Length, fill = Species)) +
  geom_density(alpha = 0.5, color = "white") +
  geom_histogram(aes(y = after_stat(density)), alpha = 0.5, bins = 10) +
  facet_wrap(vars(Species), ncol = 1)

ggplot(iris, aes(Sepal.Length)) +
  geom_histogram(aes(y = after_stat(density)), 
                 color = "#000000", fill = "#0099F8") +
  geom_density(color = "#000000", fill = "#0EE100", alpha = 0.5) +
  geom_vline(aes(xintercept = mean(Sepal.Length)), 
             color = "#000000", size = 1, linetype = "dashed") +
  labs(
    title = "Distribution of Sepal Length",
    subtitle = "Made by ggplot2",
    caption = "Source: Iris Data",
    x = "Sepal Length",
    y = "Density"
  ) +
  theme_classic() +
  theme(
    plot.title = element_text(color = "blue", size = 16, face = "bold"),
    plot.subtitle = element_text(size = 10),
    plot.caption = element_text(face = "italic")
  ) +
  annotate("text", x = 5.9, y = 0.6, 
           label = paste0("Mode: ", round(DescTools::Mode(iris$Sepal.Length),1) ), hjust = 0)
## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

Bar plot / Column plot

geom_bar

ggplot(student) +
  geom_bar(aes(x = Computer)) 

ggplot(student) +
  geom_bar(aes(x = Computer)) +
  coord_flip()

ggplot(student) +
  geom_bar(aes(y = Computer))

geom_col

student %>% 
  count(Computer) %>% 
  ggplot() +
  geom_col(aes(x = Computer, y = n))

student %>% 
  count(Computer) %>% 
  ggplot() +
  geom_col(aes(x = Computer, y = n)) +
  coord_flip()

student %>% 
  count(Computer) %>% 
  ggplot() +
  geom_col(aes(y = Computer, x = n))

student %>% 
  count(Computer) %>% 
  ggplot() +
  geom_bar(aes(x = Computer, y = n), stat = "identity")

Arranging bars

geom_bar modification

ggplot(student) +
  geom_bar(aes(x = Computer)) +
  scale_x_discrete(limits = c("Laptop", "Desktop", "Tablet"))

ggplot(student) +
  geom_bar(aes(x = fct_infreq(Computer))) +
  labs(x = "Computer Usage Status")

ggplot(student) +
  geom_bar(aes(x = fct_infreq(Computer) %>% fct_rev())) +
  labs(x = "Computer Usage", y = "frequency", 
       title = "A simple bar plot")

ggplot(student) +
  geom_bar(aes(x = Computer), fill = c("black","bisque3","red")) +
  theme_bw()

ggplot(student) +
  geom_bar(aes(x = Computer, fill = Computer)) +
  scale_fill_manual(values = c("black","bisque3","red"))

geom_col modification

student %>% 
  count(Computer) %>% 
  ggplot() +
  geom_col(aes(x = Computer, y = n)) + 
  scale_x_discrete(limits = c("Laptop", "Desktop", "Tablet"))

student %>% 
  count(Computer) %>% 
  ggplot() +
  geom_col(aes(x = reorder(Computer, -n), y = n))

student %>% 
  count(Computer) %>% 
  ggplot(aes(x = reorder(Computer, -n), y = n, label=n)) +
  geom_col() +
  geom_text(vjust=-1, color="black", size=3.5) +
  theme_minimal() +
  ylim(0, 100)

student %>% 
  count(Computer) %>% 
  ggplot(aes(x = reorder(Computer, -n), y = n)) +
  geom_col() +
  geom_text(aes(label=n), vjust=-0.5, color="black", size=3) +
  theme_minimal()

student %>% 
  count(Computer, Class) %>% 
  ggplot(aes(x = reorder(Computer, -n), y = n)) +
  geom_col() +
  geom_text(aes(label=n), vjust=-0.5, color="black", size=3) +
  theme_minimal() +
  facet_wrap(vars(Class)) +
  ylim(0, 35) +
  labs(x = "Computer Usage")

student %>% 
  count(Computer, Class) %>% 
  ggplot(aes(x = reorder(Computer, -n), y = n)) +
  geom_col(fill = "cornflowerblue") +
  geom_text(aes(label=n), vjust=-0.5, color="black", size=3) +
  theme_light() +
  facet_wrap(vars(Class)) +
  ylim(0, 35) +
  labs(x = "Device usage", y = "Freq.", title = "Frequency of Device Usage by Class") + 
  theme(plot.title = element_text(hjust = 0.5),
        strip.text = element_text(colour = 'black'))

Values on the bars

ggplot(student, aes(y = Major)) +
  geom_bar() +
  geom_text(aes(x = after_stat(count - 1), label = after_stat(count)), 
            stat = "count", 
            size = 4, col = "white") +
  labs(x = "Freq.", y  = NULL)

# ggplot(student, aes(y = Major, fill = Computer)) +
#   geom_bar(position = "dodge") +
#   geom_text(aes(x = after_stat(count + 1), label = after_stat(count)), 
#             stat = "count", 
#             size = 3,
#             position = position_dodge(1)) +
#   labs(x = "Freq.", y  = NULL)

Stacked and Percentage Filled Bar Plot

ggplot(student) +
  geom_bar(aes(x = Class, fill = Employment))

ggplot(student) +
  geom_bar(aes(x = Class, fill = Employment), position = "stack") 

ggplot(student) +
  geom_bar(aes(x = Class, fill = Employment), position = "dodge") 

ggplot(student) +
  geom_bar(aes(x = Class, fill = Employment), position = "dodge") 

ggplot(student) +
  geom_bar(aes(x = Class, fill = Employment), position = "fill")

Arranging bars

student %>% 
  mutate(Class = factor(Class, levels = c("Freshman","Sophomore", "Junior","Senior"))) %>% 
  ggplot() +
  geom_bar(aes(x = Class, fill = Employment), position = "fill") 

ggplot(student) +
  geom_bar(aes(x = Class, fill = Employment), position = "fill") +
  scale_x_discrete(limits = c("Freshman","Sophomore", "Junior","Senior"))

Values on bars

ggplot(student, aes(x = Class, fill = Employment)) +
  geom_bar(position = "fill") +
  geom_text(aes(label = after_stat(count)), size = 3,
            stat = "count", position = position_fill(vjust = 0.5)) 

CGPfunctions::PlotXTabs2(
  data = student,
  y = Gender,
  x =  Computer, 
  results.subtitle = FALSE, 
  sample.size.label = TRUE, palette = "Set1",
  ggtheme = ggplot2::theme_bw()
) + labs(title = "Stacked Barplot of Device Usage by Gender")
## Warning in .recacheSubclasses(def@className, def, env): undefined subclass
## "ndiMatrix" of class "replValueSp"; definition not updated

Legend customization

Legend position

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  labs(x = "Sepal Length", y = "Sepal Width",
       title = "Scatter Plot of Sepal Length vs Width") +
  theme(legend.position = "right")

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  labs(x = "Sepal Length", y = "Sepal Width",
       title = "Scatter Plot of Sepal Length vs Width") +
  guides(color = guide_legend(position = "bottom"))

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  labs(x = "Sepal Length", y = "Sepal Width",
       title = "Scatter Plot of Sepal Length vs Width") +
  guides(
    color = guide_legend(
      title = "Species Name",
      position = "bottom",
      direction = "horizontal",
      title.position = "left",
      reverse = FALSE)
  )

Hide legend for a specific attribute

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  guides(color = "none")

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  guides(size = "none")

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  guides(color = "none", size = "none")

Hide all legends

ggplot(data = iris) +
  geom_point(aes(x = Sepal.Length, y = Sepal.Width, col = Species, size = Petal.Length)) + 
  theme(legend.position = "none")

Reordering levels of legend

ggplot(data = student) +
  geom_bar(aes(y = Computer, fill = Gender), position = "fill") +
  scale_fill_discrete(breaks = c("Male","Female")) +
  theme(legend.position = "bottom")

Axis Customization

# install.packages(scales)
ggplot(student, aes(x = Class, fill = Employment)) +
  geom_bar(position = "fill") +
  labs(y = "Propotion") +
  scale_y_continuous(labels = scales::label_percent())

student %>% 
  mutate(Class = factor(Class, 
                        levels = c("Freshman","Sophomore", "Junior","Senior"))) %>% 
  group_by(Class, Gender) %>% 
  summarize(AvgSpending = mean(Spending)) %>%
  ungroup() %>% 
  ggplot() +
  geom_col(aes(fill = Class, y = AvgSpending, x = Gender), position = "dodge") +
  theme(legend.position = "bottom") +
  scale_y_continuous(labels = scales::label_currency(prefix = "BDT "))
## `summarise()` has grouped output by 'Class'. You can override using the
## `.groups` argument.

Box Plot

ggplot(student, aes(x = Class, y = Spending)) +
  geom_boxplot() +
  geom_jitter()

iris %>% 
  ggplot(aes(x = Species, y = Sepal.Length)) +
  geom_boxplot() +
  geom_jitter()

Code for ggpubr:

student %>% 
  mutate(Class = factor(Class, 
                        levels = c("Freshman","Sophomore", "Junior","Senior"))) %>% 
  ggboxplot(x = "Class", y = "Spending",
            color = "Class", palette =c("#00AFBB", "#E7B800", "#FC4E07","black"),
            add = "jitter", shape = "Class") +
  theme(legend.position = "none") +
  geom_pwc(method = "t_test") +
  stat_compare_means(method = "anova", label.y = 50)

p2 <- ggboxplot(data = iris, x = "Species", y = "Sepal.Length",
          color = "Species", add = "jitter", palette = "npg") +
  theme(legend.position = "none") +
  geom_pwc(method = "t_test")

p2

ggadjust_pvalue(
  p2, p.adjust.method = "bonferroni",
  label = "{p.adj.format}{p.adj.signif}"
)

p3 <- student %>% 
  mutate(Class = factor(Class, 
                        levels = c("Freshman","Sophomore", "Junior","Senior"))) %>% 
  ggboxplot(x = "Class", y = "Spending",
            color = "Class", palette =c("#00AFBB", "#E7B800", "#FC4E07","black"),
            add = "jitter", shape = "Class") +
  theme(legend.position = "none") +
  geom_pwc(method = "t_test") +
  stat_compare_means(method = "anova", label.y = 50)

p3

ggadjust_pvalue(
  p3, p.adjust.method = "bonferroni",
  label = "{p.adj.format}{p.adj.signif}", hide.ns = TRUE
)

esquisse add-ins

# install.packages("esquisse")

ggplot(iris) +
  aes(x = Sepal.Length, fill = Species) +
  geom_histogram(bins = 30L) +
  scale_fill_manual(
    values = c(setosa = "#6C1C80",versicolor = "#30A19C",virginica = "#123B96")
    ) +
  labs(x = "X label", y = "Y label", title = "Title",
    subtitle = "Subtitle", caption = "Caption", fill = "Fill label") +
  theme_bw() +
  theme(
    legend.position = "bottom",
    plot.title = element_text(face = "bold.italic"),
    plot.subtitle = element_text(face = "italic"),
    axis.title.y = element_text(face = "italic"),
    axis.title.x = element_text(face = "italic")
  ) +
  facet_wrap(vars(Species), ncol = 1)

Combining multiple plots

patchwork

library(patchwork)
(p3 | p2) /
  p1

See more: https://patchwork.data-imaginist.com/articles/patchwork.html

ggarrange

library(gridExtra)
ggarrange(p1, 
          ggarrange(p2, p3, ncol = 2),
          ncol = 1, 
          nrow = 2)

Integration of Plotly in ggplot2

p1 %>% plotly::ggplotly()
p4 <- ggplot(student, aes(x = Class, y = Spending)) +
  geom_boxplot() +
  geom_jitter()
plotly::ggplotly(p4)