3 Manipulating Columns

Select columns	select(), pull()
Arrange columns	relocate(), select()
Add columns	mutate(), transmute(), bind_cols()
Rename columns	rename()

Please copy the following code and paste them into a script in the RStudio. We will walk through them with the visual explanations from the dplyr cheat sheet.

Input

# Sample code from the dplyr cheat sheet
## 3.1 Select columns
pullmt=pull(mtcars, wt)
selectmt=select(mtcars, mpg, wt)

## 3.2 Arrange columns
relocatemt=relocate(mtcars, mpg, cyl, .after = last_col())

## 3.3 Add columns
mutmt=mutate(mtcars, gpm = 1 / mpg)
tmt=transmute(mtcars, gpm = 1 / mpg)

### A showcase for bind_cols
x <- data.frame(
  A = c('a', 'b', 'c'), 
  B = c('t', 'u', 'v'),
  C = c(1, 2, 3))
x
y <- data.frame(
  E = c('a', 'b', 'd'), 
  G = c('t', 'u', 'w'), #skip F because F = FALSE, so it's best not to use that letter.
  H = c(3, 2, 1))
y
# You have to make sure the binding is meaningful by yourself. 
# This is not the same as joins
bind_cols(x, y) 

## 3.4 Rename columns
rename(cars, distance = dist)

Visual walkthrough

Select columns: select() and pull()

What to notice: these functions keep only the columns you ask for, which narrows your data to relevant variables.

Arrange columns: relocate()

What to notice: this changes column order only; rows and values stay the same.

Add columns: mutate() and transmute()

What to notice: mutate() adds new columns while keeping existing ones, but transmute() keeps only new columns.

Rename columns: rename()

What to notice: only the column names change; the data inside each column does not.

Combine columns from multiple data frames: bind_cols()

What to notice: bind_cols() joins data by row position, so use it only when rows correspond correctly.

Practice 2

iris is a data frame with 150 cases (rows) and 5 variables (columns) named Sepal.Length, Sepal.Width, Petal.Length, Petal.Width, and Species. Make a new data frame which contains only Species and the ratio of Petal.Width and Petal.Length.

click here for solution

iris$Species=str_replace(iris$Species, 
            "setosa", 
            "bristle-pointed iris")

 
 

This page is meant to introduce functions that help manipulate columns.