• Overview of R
• Input and evaluation
• Atomic classes of objects
• Data types
• Help documents
• Working directory
• Reading and looking at data
• Extracting subsets
• Vectorized operations
• Control structures
• Writing functions

• R is a dialect of S.
• S is a language developed by John Chambers at Bell Labs in 1976.
• Version 3 of S was rewritten in C language in 1988.
• R was created by Ross Ihaka and Robert Gentleman at the University of Auckland, New Zealand in 1991.
• The R Core Group was formed in 1997.
• R version 1.0.0 was released in 2000.
• R version 3.4.1 was released on June 30, 2017.

• One is granted:
  1. The freedom to run the program for any purpose.
  2. The freedom to study how the program works.
  3. The freedom to share copies of the program.
  4. The freedom to improve or modify the program, and release the improvements to the public.
• “We call this free software because the user is free.” (The Free Software Foundation)

• Use <- as the assignment operator.
• = is also eligible but I recommend <-. Here is my reason for that.

• # starts comments. Comments won’t be executed.

  x <- 5
  name <- "Yu-Zhen"  # keying my name in
  # Nothing happens.

• Incomplete expressions

x <- 5    # inputting the value of x
x         # auto-printing

[1] 5

print(x)  # explicit printing

[1] 5

• Character: "Yu-Zhen", letters, "Hello world!"
• Numeric: pi, 123, Inf
• Integer: 10L, 123L
• Complex: 7 + 3i
• Logical: TRUE or FALSE
• Use the function class to find out which class the object is.

• NA (Not Available) and NaN (Not a Number)
• An NaN is an NA, but the converse is false.
• is.na and is.nan are used to test if objects are NA and NaN, respectively.

• Use the function c (or vector) to create a vector.
• An integer sequence from m to n can be created by m:n.
• Vectors accept only the same class of elements.

• Coercion occurs when different classes are input. (logical < integer < numeric < complex < character)

  c(TRUE, 1 + 3i, "Yay!")

  [1] "TRUE" "1+3i" "Yay!"

• Explicit coercion can be performed with as.* functions.

• Use list to create a list.

• Lists are similar with vectors but accept different classes of elements.

  my_list <- list(1:3, letters, TRUE)
  my_list2 <- list(first = 1:3, second = letters, third = TRUE)
  # What's the difference between my_list and my_list2?

Create a list like this:

$one
[1] "a" "b" "c" "d"

$two
[1] 76 77 78 79 80

$three
[1]  TRUE FALSE  TRUE

• Matrices are like vectors with two dimensions (row and column).

• Use matrix to create a matrix (column-wise by default).

  matrix(1:6, 2, 3)

       [,1] [,2] [,3]
  [1,]    1    3    5
  [2,]    2    4    6

• Bind columns or rows by cbind or rbind respectively.

  rbind(1:4, 5:8)

       [,1] [,2] [,3] [,4]
  [1,]    1    2    3    4
  [2,]    5    6    7    8

• factor creates factors from vectors.

• The “levels” argument can be specified to assign base level.

  factor(c("high", "high", "low", "mid", "low"),
         levels = c("low", "mid", "high"))

  [1] high high low  mid  low 
  Levels: low mid high

• Further reading about factor levels
• Factors are very useful for analyzing categorical variables.

• Pay extra attention when coercing factors into numbers.

• data.frame can create one data frame.
• Data frames look like matrices but allow different classes of elements across columns.
• cbind and rbind are also effective for data frames.

Try to make the following data frame. Save the object as "practice.frame".
Hint: It's like creating a list but with the same length of vectors.

  num   age RT MMSEpass
1   1 young 12     TRUE
2   2   old 50    FALSE
3   3 young 15     TRUE

• ? or help gets the help document of the function.
  • Some arguments have default values.
  • Arguments can be passed by matching name or matching position.
• ?? or help.search finds functions with the key word.
• Getting more help
  • Google
  • R_Language on PTT
  • Stack Overflow

• Files will be saved to and retrieved from the working directory if no path is specified.
• getwd retrieves current working directory path.
• setwd sets another specific path as working directory.

Let’s create a folder and set it as the working directory for this tutorial! Download the following two files and put them in that directory.
Please avoid Chinese characters and spaces in your path.

• r-tutorial-demo.csv
• r-tutorial-practice.csv

• read.table or read.csv inputs a table into R as a data frame.
• The important arguments include:
  • file: the file path or file.choose
  • header: whether the first row should be regarded as names
  • sep: specified separator for the table (e.g. “,” for .csv files)

  demo.table <- read.csv("r-tutorial-demo.csv")

• write.table or write.csv outputs data in a spreadsheet.

• load inputs previously saved workspace into R.
• Workspace contains objects we’ve assigned.

• save.image saves the whole current workspace as a .RData file.

  save.image("tutorial.RData")  # Save the whole workspace
  q()                           # Quit the R session
  # After turning on a new R session and setting the working directory
  load("tutorial.RData")

• source inputs a written .R file into R.
• dump can be used to save .R files.

demo.table

     age    sex height
1  young female  173.1
2  young   male  167.5
3  young   male  159.2
4  young   male  148.1
5    old female  165.9
6  young female  157.7
7  young   male  163.6
8  young   male  164.4
9    old female  159.7
10   old female  160.3

head(demo.table)

    age    sex height
1 young female  173.1
2 young   male  167.5
3 young   male  159.2
4 young   male  148.1
5   old female  165.9
6 young female  157.7

# Try tail(demo.table) by yourself

summary(demo.table)

    age        sex        height     
 old  :3   female:5   Min.   :148.1  
 young:7   male  :5   1st Qu.:159.3  
                      Median :161.9  
                      Mean   :161.9  
                      3rd Qu.:165.5  
                      Max.   :173.1  

str(demo.table)

'data.frame':   10 obs. of  3 variables:
 $ age   : Factor w/ 2 levels "old","young": 2 2 2 2 1 2 2 2 1 1
 $ sex   : Factor w/ 2 levels "female","male": 1 2 2 2 1 1 2 2 1 1
 $ height: num  173 168 159 148 166 ...

Take a look at the data frame you created earlier.

'data.frame':   3 obs. of  4 variables:
 $ num     : int  1 2 3
 $ age     : Factor w/ 2 levels "young","old": 1 2 1
 $ RT      : num  12 50 15
 $ MMSEpass: logi  TRUE FALSE TRUE

• [] extracts elements by names, indexes, or logical values.
  • It can be used to select more than one element.
  • It usually returns an object as the same class as the original.
• [[]] extracts elements from lists and data frames.
  • It can extract only one element.
  • Returned objects are not necessarily the same class as the original.
• $ extracts elements by name.

• By default, subsetting a single row or a single column from a matrix will return a vector, unless the argument "drop" is FALSE.

  demo.table$height

   [1] 173.1 167.5 159.2 148.1 165.9 157.7 163.6 164.4 159.7 160.3

  # What if I only need the 2nd, 4th, and 5th elements of the heights?

x <- 1:3
y <- 4:6
# x + y = ?

a <- 1:4
b <- 1:2
# a - b = ?

m <- matrix(6, 2, 2)
n <- matrix(1:4, 2, 2)
# m * n = ?
# m %*% n = ?

1. Read the .csv file, "r-tutorial-practice.csv", into your R session.
2. Look at the input data frame. Watch its components, structure, etc.
3. Revise the units of height and weight. (Height: from cm to m; Weight: from g to kg)
4. Calculate the BMIs of all the 11 participants and assign them as a new column called BMI of your data frame.
5. Report the BMIs of those who have MMSE passed.
6. Which participants have BMI greater than 25?
7. Export the revised data frame as a new csv file called "my-success.csv" to your working directory.

• == (equal to)
• >= (greater than or equal to)
• <= (less than or equal to)
• > (greater than)
• < (less than)
• != (not equal to)
• ! (not)
• & (and)

• | (or)

  1:3 != 2

  [1]  TRUE FALSE  TRUE

if (<condition 1>) {
  # do something
  # the only necessary part of if structures
}
else if (<condition 2>) {
  # do something different
}
else {
  # do something else
}

• A for loop takes an iteration variable, and assigns it successive values from a sequence or a vector.

  x <- c(90, 87, 60, 45, 50)
  for (i in 1:5) {
    if (x[i] >= 60) {
      print("Pass")
    }
    else {
      print("Fail")
    }
  }

  [1] "Pass"
  [1] "Pass"
  [1] "Pass"
  [1] "Fail"
  [1] "Fail"

• while loops begin with testing a condition. If it is true, then execute the body. After the execution, the condition will be tested again.

• while loops can possibly cause infinite loops if not written properly.

  i <- 1
  while (i < 4) {
    i <- i + 1
    print(i)
  }

  [1] 2
  [1] 3
  [1] 4

1. Write a loop to revise the unreasonable waist (> 35 inches) in your practiced data frame from centimeters to inches (1 inch = 2.5 cm).
2. The participant with missing waist information has 27 waist. Update your data frame accordingly. DO NOT do practice.table$waist[10] <- 27.
3. Export and overwrite "my-success.csv".

add <- function(x, y) {
  x + y
}
add(2, 3)

[1] 5

Try to create a function, which can subset and print out elements greater than a specified number from a numeric vector.

greaters <- function(x, n) {
  # Your codes here
}
# For example, greaters(1:5, 3) should return 4 and 5.