introtor.lecture.9.the tidyverse

install.packages( ‘packageName’ )
Packages window is next to the Help window.
library( ‘packageName’ ): to load the package in. There may be some functions that have the same name with the original functions. So when you have loaded in some packages and you call a function with this kind of name, the function is in the package.
search on the internet for the documentations of these packages.

• Filter rows
• Sort the rows
• Select columns
• Add new columns
• Summarize data(use tapply): pay attention to the ‘na.rm’ parameter.

• Filter rows
  

  1 2	# Select rows that Species is equal to "virginica" filter( iris, Species == "virginica" )

• Sort the rows
  

  1 2 3

  # Sort iris with Sepal.Length decreasing # In default, it is in an increasing order arrange( iris, desc( Sepal.Length ) )

• Select columns
  

  1 2 3

  # Select one or more columns select( iris, Sepal.Length, Sepal.Width, Species)

• Add new columns
  

  1 2 3 4

  # Add a column at the end mutate( iris, Sepal.Product = Sepal.Width * Sepal.Length ) # By the way, "mutate" means change, but it's normally in a bad way.

• Summarize data
  

  1 2 3 4 5 6 7 8

  # This will return a data frame to us # with one column: mean(Sepal.Length) summarize( iris, mean(Sepal.Length) ) # This will return a data frame # with two columns: Species and Sepal.Length iris.grouped <- group_by( iris, Species ) summarize( iris.grouped, mean(Sepal.Length) )

The boxplot is another way to visualize a one-dimensional distribution of data, but it’s more abstract than the strip chart or the histogram. Instead, a boxplot only displays a few aspects of the data:
First, the center of the boxplot is a box with a line in the middle of the box. This line represents the median of the data, which is the value such that 50% of the data is bellow the value and 50% of the data is above the value.

1
2
3
4
5

boxplot( rivers,
         main = "Box plot of rivers data",
         xlab = "Length (miles)",
         horizontal = TRUE,
         col = "cyan3")

There’s a much more interesting way to use boxplots, however. Rather than just look at one group, we can also show all of the subgroup-specific boxplots in one graph.( Think about the stripcharts we drew before.)

1
2
3
4
5

boxplot( iris$Sepal.Length ~ iris$Species,
         main = "Stratified Stripchart for Sepal Length",
         ylab = "Sepal length (cm)",
         horizontal = TRUE,
         col = c( "cyan1", "dodgerblue1", "skyblue" ) )

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

# We set R to make graphs of 3 rows and 1 column.
par( mfrow = c(3,1) )

hist( iris$Sepal.Length[ iris$Species == "setosa" ],
      main = "Histogram of Setosa Sepal Length",
      ylab = "Sepal Length (cm)",
      breaks = 10,
      col = "cyan1" )

# Notice that the graphs don't have to be the same type.
stripchart( iris$Sepal.Length[ iris$Species == "setosa" ],
            main = "a. Stripchart of Setosa Sepal Length",
            ylim = c(0, 2),
            xlim = c(4, 8),
            method = "jitter",
            jitter = 0.5,
            pch = 24,  # this give us a hollow shape
            cex = 1.4,
            bg = "salmon2" 
            # bg means background color which will fill in the hollow shape.
            )

hist( iris$Sepal.Length[ iris$Species == "virginica" ],
      main = "Histogram of Virginica Sepal Length",
      ylab = "Sepal Length (cm)",
      breaks = 10,
      col = "cyan3" )

# At the end, we must set R to draw graphs in the normal way( 1 row 1 column)
par( mfrow = c(1,1) )

Todays Tips