Wrangling geom_bar

Whereas geom_col() expects you to pass it a y-axis mapping column, geom_bar() doesn't take a y-axis call (at least by default).

Instead, geom_bar() takes your x-axis mapping and counts every single observation (or row of the passed data frame) for each class and then draws bars of corresponding heights

These two code chunks will give you the same plot:

# geom_col()
data %>%
  groupby(xAxisCol) %>%
  summarize(value = n()) %>%
  ggplot(aes(x = xAxisCol, y = value) + 
  geom_col()

# geom_bar()
data %>%
  ggplot(aes(x = xAxisCol)) +
  geom_bar()

Let's use geom_bar() to make a look at observations with a large number of cases by region our WHO data.

Este exercício faz parte do curso

Visualization Best Practices in R

Ver curso

Instruções do exercício

filter() the who_disease data into only observations (rows) with greater than 1000 cases.
Map the x-axis to the region column.
Add geom_bar() to the plot object to draw the bars.

Exercício interativo prático

Experimente este exercício completando este código de exemplo.

who_disease %>%
	# filter data to observations of greater than 1,000 cases
	___ %>%
	# map the x-axis to the region column
	ggplot() +
	# add a geom_bar call
	___

Editar e executar o código