Preliminaries

Verify that you’ve completed these steps before moving on!

This short demo will get you started with working with visualizations in an RMarkdown file. In order for this demo to work, you need to have the tidyverse package installed. If you don’t remember if you’ve installed it, open your Console window and run:

library(tidyverse)

If you don’t get an error message, then you have the package installed. If you do get an error, run the following in your Console window to install it:

install.packages("tidyverse", repos = "https://cran.rstudio.com/")

It will take a few moments for the install to be completed. If it completes without error, move on to the next section. If you encounter errors and cannot resolve them, post on Slack for help.

Interactive RStudio

RStudio can be used to run code blocks interactively and display the results, allowing you to preview what your knitted file will look like. You have two easy methods available to you for running the code in a code block. The graphical way is to click the green arrow located on the far-right side of a code block. The keyboard shortcut is place your cursor inside the block and then press CTRL SHIFT ENTER all together. The image below illustrates:

Use one of these methods whenever you want to run the code within a block.

Setup chunks

When you are given an RMarkdown template file, it will frequently have a setup code block at the top of the page, much like the one below. These chunks will configure the knitting procedure, which controls how your output documents will look. Often, they will also load the libraries you will need to complete an exercise or assignment. If you reopen a file after restarting RStudio or switching projects, you should always run this block first.

Give it a try, run the block below.

# DO NOT ALTER THIS CHUNK
knitr::opts_chunk$set(echo = TRUE, eval = FALSE, fig.width = 5,
                      fig.asp = 0.618, out.width = "70%", dpi = 120,
                      fig.align = "center", cache = FALSE)
# Load required packages
suppressPackageStartupMessages(library(tidyverse, quietly = TRUE))

If you didn’t get an error, then excellent! If so, then tidyverse hasn’t been installed correctly, in which case post on Slack for help.

Demo: The mpg dataset

Viewing the dataset

When loading tidyverse, several practice datasets are automatically loaded, one of which is the mpg dataset. It’s good practice to start by looking at the dataset and getting familiar with the different columns and rows. You can do this within RStudio, try running the code block below:

mpg

You can also read more about the dataset by running:

?mpg

in the Console window.

Question

What is the data contained within the mpg dataset?

Answer

Make a scatterplot

It’s very easy to make a scatterplot in R using the ggplot2 library. This is the library you will read about in Chapter 3 of R for Data Science. Let’s use it to plot each car’s highway fuel efficiency (hwy) as a function of the engine size (displ). The code block below will make this plot, try running it!

ggplot(data = mpg) +
  geom_point(mapping = aes(x = displ, y = hwy))

Question

Using your intuition, what part of the above code block tells R to put the displ variable on the horizontal axis and the hwy variable on the vertical axis?

Answer

Make a slight change

Exercise

Take a look at the mpg table again (in the section Viewing the dataset). Choose a categorical variable that isn’t displ. Then, change the visualization code below so that your chosen variable replaces displ on the horizontal axis (the vertical should still be hwy).

Answer

ggplot(data = mpg) +
  geom_point(mapping = aes(x = displ, y = hwy))

Categories by color

The following code block extends the example with a new input, color = class. Try running it below:

ggplot(data = mpg) +
  geom_point(mapping = aes(x = displ, y = hwy, color = class))

Question

What does adding color = class inside the aes() code section do?

Answer

Other variations

Try running these other variations as well and observing their outputs.

First, we move color outside of aes() and set it equal to the string "blue":

ggplot(data = mpg) +
  geom_point(mapping = aes(x = displ, y = hwy), color = "blue")

Here we replace color with size:

ggplot(data = mpg) +
  geom_point(mapping = aes(x = displ, y = hwy, size = class))

Here we replace color with shape:

ggplot(data = mpg) +
  geom_point(mapping = aes(x = displ, y = hwy, shape = class))

Exercise

Create your own figure using the mpg dataset! Follow the patterns you see in the above examples in order to build a new visualization. It should be different from the examples you’ve done so far, but otherwise, it’s up to you.

Answer

End

When you are done, save the file and try knitting it to HTML to make sure that it works. If you get an error message and it won’t knit, then you will need to fix something in one of the code blocks. If the HTML rendering looks funny, first preview it in a real web-browser. On the lower-right of RStudio, click the Files tab, click visualization_mini-assignment.html, and select “View in Web Browser”. The image below illustrates:

If it still looks like something is wrong, check through your RMarkdown file again and make sure you are following the proper RMarkdown syntax rules. Consult the RMarkdown Cheatsheet and RMarkdown Reference for help.