Getting Started with R: Installing Everything You Need

Step-by-step guide to installing R, RStudio, Git, Quarto, and registering for GitHub and Netlify so you have a complete working environment for data science.

Getting Started with R Step 1 of 7 Installing R Packages → View full track →

Video lecture

1 Introduction

During a typical data science course you will use the following software and services:

• R
• R-Studio
• Git
• Github
• Netlify

You will need to install the software and register for these services on your own. This tutorial provides you with all the information you need.

Please also note that there is a separate tutorial on how to install the packages (a.k.a. R extensions) that we are going to use. It is best to continue with Installing R Packages directly after you have completed this one.

2 An overview of the apps and services

R is the programming language we are using. Installing R basically gives your computer the ability to understand commands you will issue to it using the language R. R-Studio is an application that facilitates the development of R code. It is a so called Integrated Development Environment (IDE). You may think of it as a fancy editor, which not only allows you to write programs in R using a more elaborated text editor, but simultaneously lets you preview the graphs you are designing, or the reports you are writing. Because R-Studio facilitates the use of R, it is important to install R first, and then install R-Studio second.

Git is a so called version control system (VCS). It allows you to keep track of different versions of your scripts without saving them under different names. In other words, if you use Git, you can dispense of saving Script-v1.R, Script-v2.R and Script-v3.R, but instead only keep Script.R and save the different versions in the background, together with some comments about what you have changed. Moreover, it facilitates the joint work on the same document, something that is tremendously helpful for group works. While you may not learn how to use Git in detail, it is important to install it since many user-written features of R are distributed in a way that relies on Git.

Github can be thought of as a server for Git. R is an open source programming language with a very active community. Many people are constantly developing new features for R. For instance, if a new prediction algorithm gets developed, some people will soon write a so called R-package that implements this algorithm in R. Then you as an R user can download this package and use the new algorithm. This way, R is unlikely to ever become outdated. Since most developers use Git when developing the packages, they often distribute the programs in a way that aligns well with the use of Git. And the most prominent way is Github. Thus, you can view the source code for almost all R packages you will be using via Github. Github is free, and all you need is to register!

The final service we will be using is Netlify. It is an easy way to publish reports and host small websites. We will use it such that you can create reproducible reports and present them in a visually appealing way. Given the ease of use, it is a great tool if you want to polish your results up visually and send them to friends or colleagues who might not even use R themselves.

3 Installation guidelines

The information in this tutorial is provided for Mac OS, Windows, and Linux. Given the variety of different Linux distributions the comments on the latter are rather short. There are, however, plenty explanations to be found on the web for many different Linux distributions.

After you have completed the steps outlined below you are almost done: all what remains is to install the R packages we require over the course of the seminar. To this end, please consult the Installing R Packages tutorial.

3.1 Install R

The installation of R is very similar across operating systems (OS). The easiest way is to visit the R Homepage and to download the most recent version for your OS. In case you are using Mac OS and want to use Homebrew, its best to use this formula.

Important for Mac users: There are different versions of R for Intel chips, and Apple chips (M1, M2, etc.). It is very important that you install the correct version. If you are not sure whether your Mac contains a chip from Intel or Apple, click on the Apple symbol in the upper left of the screen, then click on About this Mac and you can see which processor your Mac is using in the new window. If you have an Apple chip, always install R for the so called arm64 architecture. Intel chip users must use the x86_64 architecture instead.

Only Windows: Install RTools

If you are using Windows, it is necessary to install RTools, which is required if you want to use packages written by others that are not officially released. To do so, simply visit the following website, download the installer, and install the software:

When asked during the installation process, do not select the box for Add rtools to system PATH, but do select the box for Save version information to registry.

Only Mac: Command Line Developer Tools

The Command Line Developer Tools could be thought of as the Mac pendant to RTools. These allow you to build R packages from source (meaning, basically, you can use packages that are in early stages of distribution, or packages that are not released on the official R servers).

The easiest way to install them is to open the App Terminal, and then to type

xcode-select --install

and press Enter. Then a pop up window will open and allow you to install the software.

3.2 Update R

In case R is already installed on your computer you should make sure that your version is more or less up to date. For recent courses you should use at least R version R 4.4.2. The version you are currently using is shown as soon as you start R.

Please note: if you installed R anew in the previous step, you do not need to update it. The information on updating R is mainly relevant for people who have installed R already some time ago.

MacOS users

For MacOS users, the easiest route to update R is to just re-install the most current version from the R Homepage. Keep in mind that in this case you might need to re-install all previously installed packages. If you have a lot of packages installed that you want to keep, the following steps facilitate the re-installation process. First, save a list with all the packages you installed yourself. To this end type the following into the R console:

package_overview <- installed.packages()
package_names <- as.vector(
  package_overview[is.na(package_overview[,"Priority"]), 1])
save(package_names, file="r_packages.rda")

After re-installing R, you then need to load the file you previously saved and identify the missing packages. You can use the following code to do so if you are in the working directory in which you saved the file "r_packages.rda":

load("r_packages.rda")
packages_new <- installed.packages()
packages_new_ <- as.vector(packages_new[is.na(packages_new[,"Priority"]), 1])
missing_packages <- setdiff(package_names, packages_new_)
install.packages(missing_packages)
update.packages()

Windows users

Windows users have a slightly more convenient route available to them: the installr package. It does not require you to re-install your packages. Just type the following code into your R console:

install.packages("installr")
library(installr)
updateR(TRUE)

For more information see the package website.

Linux users

Linux users simply install R via their package manager. A quick search on Google should provide you with the information that are relevant for your particular Linux distribution. Updating is usually straightforward as well: just run the respective command from your package manager.

3.3 R-Studio

Installing R-Studio is easy. The only thing you should keep in mind is that you should install R first, and R-Studio second. So, after installing R go to the R-Studio download page and download the RStudio Desktop version for your OS according to the installation instructions provided.

If you are on Mac and you are using Homebrew you may use this formula.

If you want to update R-Studio, you just install it again. Please note that the minimal version for recent courses should be RStudio 2023.12.1+402, which is from late January 2024. You can check your version by clicking on RStudio in the upper left part of your screen when R-Studio is open. Then click on About RStudio.

3.4 Git

Installing Git is straightforward, but the right approach depends on your OS.

MacOS

On MacOS you should install Git as part of the Command Line Developer Tools, which themselves are part of XCode (see above). Its easiest to run the following command from your Terminal:

git --version

If you get an output such as git version 2.34.1 you already have what you need. If not, you will be asked to install the respective software packages (see above).

Windows

On Windows you download Git for Windows from the official Webpage, which also provides you with all the relevant instructions.

Linux

On Linux use your package manager. In most cases the name of the relevant package is git-all, so on Ubuntu, for instance, you would install Git via sudo apt install git-all.

3.5 Install Quarto

Quarto allows you to write text and R code within one document. This is very useful in many instances, and allows you to create a wide variety of nicely looking and practically appealing outputs, including apps, websites, statistical reports, and much more. To install Quarto just follow the instructions from this webpage.

4 Register for the necessary services

4.1 Github

This is easy. Just visit https://github.com/ and sign up using your email account.

4.2 Netlify

This is easy as well. Visit https://www.netlify.com/ and click on Sign up in the upper right of the webpage. You can now either create a Netlify account by clicking on Email and register a new email address, or you can link Netlify to one of the other accounts you might already have.

5 Next steps

Once you have finished installing all the relevant software described in this tutorial, continue with Installing R Packages to get all the R packages you will need.