Class 00: Installing R

In class, you will be using the lab computers to work with the R programming language. I strongly suggest that you also setup R on your own machine for working with the homework and data reports. R is an open source programming language, meaning that you can install it for free on nearly any operating system. The note here guide you through install the core R system, the RStudio IDE, and relevant R packages.

Download R

The first step is to download the R programming language. To do this go to https://cran.r-project.org/ and select your platform:

For macOS, just download R-3.4.1 (or whatever is the most recent):

For Windows, first select base

And then Download R 3.4.1

One you have the .pkg (macOS) or .exe (Windows) file, install this on your computer according to the default settings.

RStudio

The files we just downloaded are the core R language files doing all the hard work of processing data. Next, we’ll install a helpful GUI frontend that make calling R easier.

Go to https://www.rstudio.com/. Click on Products => RStudio.

Scroll down to the DOWNLOAD RSTUDIO DESKTOP button and click on it.

Scroll down again to the Installers for Supported Platforms. The Windows link gives you an exe:

And the macOS link gives a dmg:

Now, install R or RStudio as you would any other program. It should link automatically to the version of R you just installed.

Installing R packages

The final step is to download all of the R packages that we will need for the semester. It is generally easier to do these all at once rather than as we go along.

Go ahead and launch RStudio. You should see a window that looks like this, as we saw on the class computers:

To install the packages required for class, run the following lines of code in the console. There may be a warning about one or two packages not being available. Note that this may take 5-10 minutes to finish; on slower connections or older computers, it may take even longer. If you run into any problems, please let me know!

pkgs <- c("dplyr", "ggplot2", "stringr", "ggmap", "ggrepel",
          "devtools", "viridis", "plotly", "jsonlite", "lazyeval",
          "knitr", "readr", "forcats")

install.packages(pkgs,
                 repos = "https://cloud.r-project.org",
                 type = "binary",
                 dependencies = TRUE,
                 quiet = TRUE)

Now, you will also need install the smodels package. I have written this package for this course, so you need to install it in a slightly different way.

devtools::install_github("statsmaths/smodels")

Finally, make the following changes in the RStudio IDE. These are very important, so don’t skip this step:

  • In the RStudio menu click on Tools > Global Options...
  • Look for the first drop-down: Save the workspace to .RData on exit
  • Change this first drop-down to Never
  • All of the check-boxes above this dropdown should be unchecked.

And that should be it! You now have the same system running on your machine that we have in class. If you run into any issues, please let me know. It is likely that I will need to see your computer to help, so bring your laptop to class or office hours and I will help get you setup.