Boxplots can be used to get the minimum, median, maximum, and quartiles of the data being plotted. Boxplots also help to identify outliers:
b = boxplot(Salaries$salary~Salaries$rank, col=brewer.pal(8,"Set1")) names(b) # [1] "stats" "n" "conf" "out" "group" "names" b$stats # [,1] [,2] [,3] # [1,] 63100.0 62884.0 57800.0 # [2,] 74000.0 82350.0 105890.0 # [3,] 79800.0 95626.5 123321.5 # [4,] 88597.5 104331.5 145098.0 # [5,] 97032.0 126431.0 194800.0 # attr(,"class") # AsstProf # "integer"
The output is as follows:
The pros and cons of using boxplots are as follow:
Pros:
Boxplots can be used to get an understanding of the distribution, the outliers, and skewness of the data through an intuitive plot.
Cons:
- Can only be used with numerical variables
- Due to summarization, the original data is not displayed in the charts