Week 2 Tutorial:
R Basics

POP77001 Computer Programming for Social Scientists

R and development environments

  • There is some choice of integrated development environments (IDEs) for R (StatET, ESS, R Commander)
  • However, over the last decade RStudio became the de factor standard IDE for working in R
  • You can also find R extensions for your favourite text editor (Atom, Sublime Text, Visual Studio Code, Vim)
  • For the purposes of consistency with Python part of the module, we will be using Jupyter with R.

Running R in Jupyter

  • In order to be able to run R kernel in Jupyter, you need to install package IRkernel:
    • Open R (in the terminal) or RStudio:
    • Run install.packages("IRkernel") to install the package
    • Wait until the package is installed
    • Run IRkernel::installspec() to initialize R kernel for Jupyter
    • Now you should be able to launch or edit a notebook with R kernel

Tip: When starting working with R in Jupyter run options(jupyter.rich_display = FALSE) command to switch off pretty printing and get the output (albeit less neat) consistent with output in RStudio

IRkernel

Package IRkernel is required to run R in Jupyter Notebook.

JupyterLab

JupyterLab Demonstration

Code Distribution

More often than not you want to record how analysis was performed.

There are 3 principal ways of distributing R code:

  • R script (.R file)
  • R Markdown/Quarto (.Rmd/.qmd file)
  • Jupyter Notebook (.ipynb file)

R Script

  • The most straightforward way to keep track of your R code.
  • Instead of writing your R commands in the interactive console,
  • You put them in a script and run then together or one at a time.
  • Can contain a mix of valid R commands and comments (lines starting with #).
  • Easy to edit and integrate into larger projects.

Markdown formatting basics

  • Use _ or * for emphasis (single - italic, double - bold, triple - bold and italic)
    • *one* becomes one, __two__ - two and ***three*** - three
  • Headers or decreasing levels follow #, ##, ###, #### and so on
  • (Unordered) Lists follow marker -, + or *
    • Start at the left-most position for top-level
    • Indent four space and use another marker for nesting like here
  • (Numbered) Lists use 1. (counter is auto-incremented)
  • Links have syntax of [some text here](url_here)
  • Images similarly: ![alt text](url or path to image)

Markdown vs R Markdown

  • Markdown:
    • Easy-to-read and easy-to-write plain text format;
    • Separates content from its appearance (rendition);
    • Widely used across industry sectors and academic fields;
    • .md file extension.
  • R Markdown (Quarto):
    • Allows combining of R commands with regular text;
    • Compiles into PDF/DOC/HTML and other formats;
    • Can be converted into slide deck or even website!
    • .Rmd file extension (.qmd for Quarto).

Extra

Ch 27: R Markdown in Wickham & Grolemund 2017

R Markdown

R Markdown

Rendering

### Title

Some text in *italic* and **bold**

Simple list:

- A
- B

Ordered list:

1. A
1. B

Example, where $Y_i = 5 + X_i + \epsilon$

```{r}
x_i <- 3
epsilon <- rnorm(1)
y_i <- 5 + x_i + epsilon
y_i
```

Title

Some text in italic and bold

Simple list:

  • A
  • B

Ordered list:

  1. A
  2. B

Example, where \(Y_i = 5 + X_i + \epsilon\)

x_i <- 3
epsilon <- rnorm(1)
y_i <- 5 + x_i + epsilon
y_i
[1] 6.924369

Naming conventions

  • Even while allowed in R, do not use . in variable names (it works as an object attribute in Python)
  • Do not name give objects the names of existing functions and variables (e.g. c, T, list, mean)
  • Use UPPER_CASE_WITH_UNDERSCORE for named constants (e.g. variables that remain fixed and unmodified)
  • Use lower_case_with_underscores for function and variable names

Extra

The tidyverse style guide

Code layout

  • Limit all lines to a maximum of 79 characters.
  • Break up longer lines
my_long_vector <- c(
  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, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41,
  42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60
)
    
long_function_name <- function(a = "a long argument", 
                               b = "another argument",
                               c = "another long argument") {
  # As usual code is indented by two spaces.
}

Reserved words

There are ~14 reserved words in R that cannot be used as names assigned to objects.

break NA
else NaN
FALSE next
for NULL
function repeat
if TRUE
Inf while

Extra

R reserved words

Exercise 1: Vector subsetting

  • Load built-in R object letters (lower-case letters of the Roman alphabet)
  • Calculate its length
  • Generate a vector of integers that starts from 1 and has the same length as letters
  • Assign to each integer corresponding lower-case letter as its name
  • Use these names to subset all vowels
  • Now, repeat the subsetting, but using indices rather than names

Tip: You can use function which() for determining the indices of vowels

letters
 [1] "a" "b" "c" "d" "e" "f" "g" "h" "i" "j" "k" "l" "m" "n" "o" "p" "q" "r" "s"
[20] "t" "u" "v" "w" "x" "y" "z"

Tabulation & Crosstabulation

  • R function table() provides an easy way of summarizing categorical variables
  • Note that variables represented as character vectors are implicitly converted to factors
# Top 10 most populous settlements on the island of Ireland
# https://en.wikipedia.org/wiki/List_of_settlements_on_the_island_of_Ireland_by_population
top_10_settlements <- c(
    "Dublin", "Belfast", "Cork", "Limerick", "Derry",
    "Galway", "Newtownabbey", "Bangor", "Waterford", "Lisburn"
)
# Corresponding provinces
provinces <- c(
    "Leinster", "Ulster", "Munster", "Munster", "Ulster",
    "Connacht", "Ulster", "Ulster", "Munster", "Ulster"
)

Tabulation & Crosstabulation

# Given that each town appears only once, cross-tabulation might not be the most informative
table(top_10_settlements, provinces)
                  provinces
top_10_settlements Connacht Leinster Munster Ulster
      Bangor              0        0       0      1
      Belfast             0        0       0      1
      Cork                0        0       1      0
      Derry               0        0       0      1
      Dublin              0        1       0      0
      Galway              1        0       0      0
      Limerick            0        0       1      0
      Lisburn             0        0       0      1
      Newtownabbey        0        0       0      1
      Waterford           0        0       1      0
# Instead, we can just get tabulate the `provinces` vector
# and check the value counts for each province
table(provinces)
provinces
Connacht Leinster  Munster   Ulster 
       1        1        3        5

Exercise 2: Working with Factors

  • As you note the output of table(provinces) is sorted alphabetically
  • Change this to reflect the actual counts
  • First, let’s store the result of tabulation for later re-use
  • Start from exploring the structure of this object with str()
  • What are the 2 main parts of this object? How are they stored?
  • Extract the relevant parts from the stored object
  • Save them as a named vector with provinces as names and counts as values
  • Use sort() function to sort the vector in a decreasing order (from largest to smallest)
  • Convert the original provinces vector into a factor with the levels ordered accordingly
  • Re-run table(provinces)
tab <- table(provinces)

Week 2 Exercise (unassessed)

  • Save a letters object under a different name
  • Convert saved object into a matrix of 13 rows and 2 columns
  • Subset letter ‘f’ using indices
  • Concatenate 3 copies of letters object together in a single character vector
  • Convert it into a 3-dimensional array, where each dimension appears as a matrix above
  • Subset all letters ‘f’ across all 3 dimensions