Lab: getting started with R

NOTE: To preview this file, click the “Render” button in RStudio.

Installing R and RStudio

(Skip this part if you’ve already installed R and RStudio.)

To install R (latest release: 4.4.0), visit https://www.r-project.org and choose your system. Click on the download R link in the middle of the page under “Getting Started.” Download and install the installer files (executable, pkg, etc) that correspond to your system.

Although you can use R without any integrated development environment (IDE), you will need to install RStudio, by far the most popular IDE for R, for this summer. Basically, it makes your life with R much easier and we will be using it throughout the program. To install RStudio, visit https://posit.co/download/rstudio-desktop and choose your system. The installer is preferred. If you have RStudio installed but not the latest version, just download the latest installer and install.

Typical workflow

Writing R scripts

You can type R commands directly into the Console (lower left pane), but this can become quite tedious and annoying when your work becomes more complex. Instead, you can code in R Scripts. An R Script is a file type which R recognizes as storing R commands and is saved as a .R file. R Scripts are useful as we can edit our code before sending it to be run in the console.

In RStudio, to open a new R Script: File > New File > R Script.

Using Quarto

An Quarto file is a dynamic document for writing reproducible reports and communicating results. It contains the reproducible source code along with the narration that a reader needs to understand your work.

There are three important elements to a Quarto file:

  • A YAML header at the top (surrounded by ---)
  • Chunks of R code surrounded by ```
  • Text mixed with simple text formatting like ## Heading and italics

(Note that this file itself is a Quarto document.)

If you are familiar with the LaTeX syntax, math mode works like a charm in almost the same way:

\[ f (x) = \frac{1}{\sqrt{2\pi}} \exp \left( - \frac{x^2}{2} \right) \]

A chunk of embedded R code is the following:

# R code here
print("Hello World")
[1] "Hello World"

All the lab documents will be Quarto files so you need to know how to render and convert them into a reader-friendly documents. We recommend to render as html file but if you have LaTeX installed, you can change the format to pdf.

For more details on Quarto, see the comprehensive manual online and the Quarto chapter of R for Data Science (2e). See also the guide on Markdown Basics for more on Markdown syntax. For code chunk options, see this guide.

Installing R packages

R performs a wide variety of functions, such as data manipulation, modeling, and visualization. The extensive code base beyond the built-in functions are managed by packages created from numerous statisticians and developers. The Comprehensive R Archive Network (CRAN) manages the open-source distribution and the quality control of the R packages.

To install an R package, using the function install.packages and put the package name in the parentheses and the quote. While this is preferred, for those using RStudio, you can also go to “Tools” then “Install Packages” and then input the package name.

install.packages("tidyverse")

Important: NEVER install new packages in a code block in a .qmd file. That is, the install.packages() function should NEVER be in your code chunks (unless they are commented out using #). The library() function, however, will be used throughout your code: The library() function loads packages only after they are installed.

If in any time you get a message says: “Do you want to install from sources the package which needs compilation?” Choose “No” will tend to bring less troubles. (Note: This happens when the bleeding-edge version package is available, but not yet compiled for each OS distribution. In many case, you can just proceed without the source compilation.)

Each package only needs to be installed once. Whenever you want to use functions defined in the package, you need to load the package with the command:

library(tidyverse)

Here is a list of packages that we may need (but not limited to) in the following lectures and/or labs. Make sure you can install all of them. If you fail to install any package, please update R and RStudio first and check the error message for any other packages that need to install first.

library(tidyverse)
library(devtools)
library(ranger)
library(glmnet)

Basic data type and operators

Data type: vector

The basic unit of R is a vector. A vector is a collection of values of the same type and the type could be:

  • numeric (double/integer number): digits with optional decimal point
v1 <- c(1, 5, 8.3, 0.02, 99999)
typeof(v1)
[1] "double"
  • character: a string (or word) in double or single quotes, “…” or ’…’.
v2 <- c("apple", "banana", "3 chairs", "dimension1", ">-<")
typeof(v2)
[1] "character"
  • logical: TRUE and FALSE
v3 <- c(TRUE, FALSE, FALSE)
typeof(v3)
[1] "logical"

Note: Oftentimes, factor is used to encode a character vector into unique numeric vector.

player_type <- c("Batter", "Batter", "Hitter", "Batter", "Hitter")
player_type <- factor(player_type)
str(player_type)
 Factor w/ 2 levels "Batter","Hitter": 1 1 2 1 2
typeof(player_type)
[1] "integer"

Data type: lists

Vector can store only single data type:

typeof(c(1, TRUE, "apple"))
[1] "character"

List is a vector of vectors which can store different data types of vectors:

roster <- list(
  name = c("Quang", "Akshay", "Nick", "Princess", "Yuchen", "JungHo", "Daven"),
  role = c("Instructor", "TA", "TA", "TA", "TA", "TA", "TA"),
  is_TA = c(FALSE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE)
)
str(roster)
List of 3
 $ name : chr [1:7] "Quang" "Akshay" "Nick" "Princess" ...
 $ role : chr [1:7] "Instructor" "TA" "TA" "TA" ...
 $ is_TA: logi [1:7] FALSE TRUE TRUE TRUE TRUE TRUE ...

R uses a specific type of list, data frame, containing the same number of rows with unique row names.

str(iris)
'data.frame':   150 obs. of  5 variables:
 $ Sepal.Length: num  5.1 4.9 4.7 4.6 5 5.4 4.6 5 4.4 4.9 ...
 $ Sepal.Width : num  3.5 3 3.2 3.1 3.6 3.9 3.4 3.4 2.9 3.1 ...
 $ Petal.Length: num  1.4 1.4 1.3 1.5 1.4 1.7 1.4 1.5 1.4 1.5 ...
 $ Petal.Width : num  0.2 0.2 0.2 0.2 0.2 0.4 0.3 0.2 0.2 0.1 ...
 $ Species     : Factor w/ 3 levels "setosa","versicolor",..: 1 1 1 1 1 1 1 1 1 1 ...
typeof(iris)
[1] "list"

Operators

We can perform element-wise actions on vectors through the operators:

  • arithmetic: +, -, *, /, ^ (for integer division, %/% is quotient, %% is remainder)
v1 <- c(1,2,3)
v2 <- c(4,5,6)

v1 + v2
[1] 5 7 9
v1 * v2
[1]  4 10 18
v2 %% v1
[1] 0 1 0
  • relation: >, >=, < ,<=, ==, !=
5 > 4
[1] TRUE
5 <= 4
[1] FALSE
33 == 22
[1] FALSE
33 != 22
[1] TRUE
  • logic: ! (not), & (and), | (or)
(5 > 6) | (2 < 3)
[1] TRUE
(5 > 6) & (2 < 3)
[1] FALSE
!(5 > 6) & (2 < 3)
[1] TRUE
  • sequence: i:j (: operator, i and j are any two arbitrary numbers)
1:5
[1] 1 2 3 4 5
5:1
[1] 5 4 3 2 1
-1:-5
[1] -1 -2 -3 -4 -5
-1:5
[1] -1  0  1  2  3  4  5

Loading .csv files

Most of the data provided to you are in .csv format. In the code chunk below, we use the read_csv() function (from the readr package, part of the tidyverse) to load a dataset that is saved in a folder located in the SURE GitHub repository. In quotations, insert the file path where the dataset is located, which in this case is online. However, typically you’ll save .csv files locally first and put them in an organized folder to access later.

library(tidyverse)
heart_disease <- read_csv("https://raw.githubusercontent.com/36-SURE/36-SURE.github.io/main/data/heart_disease.csv")
head(heart_disease)

Looking for help

If you have any R problem, the best step is to use the help() function (or equivalently the ?). For example,

help(str)
help(lm)

Or you can use the command ?…

?str
?lm

Double question marks can lead to a more general search.

??predict

You should ALWAYS consult the R help documentation first before attempting to google around (or ask ChatGPT) for a solution.

Exercises

  1. Create four vectors, v1 and v2 are numeric vectors, v3 is a character vector and v4 is a logic vector. Make sure the length of v1 and v2 are the same. (Hint: a way to check the length is to use the function length())
# R code here
  1. Preform add, minus, product and division on v1 and v2.
# R code here
  1. Create four statements with both relation and logic operators, that 2 of them return TRUE and 2 of them return FALSE.
# R code here
  1. Create 2 sequences with length 20, one in an increasing order and the other in a decreasing order.
# R code here
  1. The following gapminder dataset contains health and income outcomes for 184 countries from 1960 to 2016 from the , accessed via the dslabs package. How many of the rows in the dataset are from the Caribbean (coded as Caribbean in region)? How about Eastern Europe? Can you summarize the counts for all regions? (Hint: table())
library(dslabs)
data(gapminder)

# R code here

Text formatting in Quarto

There are a lot of ways to format text in a Quarto document, e.g., italics and bold (just scan through this .qmd file to see how this was done). See this guide for more tips/tricks. In particular, check out the Markdown Basics and other guides under Authoring. See also this guide on R code chunk options.

As you’ll see throughout this summer (and especially with your project), well-formatted .html files can be a great way to showcase data science results to the public online. (Check out the project showcase from 2023 and 2022.)

Customizing RStudio

RStudio theme

RStudio can be customized with different themes. To explore built-in themes,

  • Navigate to the menu bar at the top of your screen
  • Choose Tools > Global Options > Appearance
  • Change your RStudio theme under Editor theme

(FYI, Quang uses the Tomorrow Night Bright theme.)

Note that within the Appearance tab, there are also options for changing your Editor font, Editor font size, etc.

RStudio panes

Within RStudio, there are several panes (e.g., Console, Help, Environment, History, Plots, etc.). To customize, go to Tools > Global Options > Pane Layout, and arrange the panes as you see fit.

Feel free to explore other options within the Tools > Global Option menu.