- Intro to
ggplot2 - Visualizing distributions
- Changing the appearance of plots
- Graphing distributions across groups
- Boxplots and violin plots
- Scatter plots
- Line plots
- Introduction to
dplyr - Heatmaps
- Alluvial diagrams
- Primer on
tidyr - Bar plots
- Advanced bar plots and
lubridate
- Coefficient plots
- Predictive probabilities
- Maps using
geom_polygon() - Introduction to
sf - Choropleth maps
- Stamen map server
- Plotting density on a map
You should have both R and RStudio installed on your machine.
In this workshop, we will be using R together with the integrated development environment (IDE) RStudio. In addition to offering a "cleaner" programming development than the basic R editor, RStudio offers a large number of added functionalities for integrating code into documents, built-in tools and web-development.
There are no formal prerequisites for this workshop. However, I am assuming that participants have a basic understanding of R programming, in particular:
- Setting a working directory,
- Installing and loading packages,
- Reading and writing data,
- Basic data formats (scalar, vector, data frame),
- Basic variable types (numeric, character, factor, logical),
- Basic vector and data frame operations, such as subsetting, transforming variables, merging, reshaping, etc.
If you are unfamiliar with R or would like to brush up on your skills, take a look at my intro to data management workshop. The first two sessions go over basic R functionality and programming principles. The latter four sessions introduce data management operations using packages from the tidyverse suite. I also recommend taking a look at R for Data Science website and/or book for a great resource on learning R and data management.
The key to learning R is: Google! This workshop will give you an overview over data visualiztion in R, but to become truly proficient you will have to actively use it yourself, trouble shoot, ask questions, and google! The R mailing list and other help pages such as StackOverflow offer a rich archive of questions and answers by the R community. For example, if you google "recode variable in r" you will find a variety of useful websites explaining how to do this on the first page of the search results. Also, don't be surprised if you find a variety of different ways to execute the same task.
RStudio has a useful help menu. In addition, you can get information on any function or integrated data set in R through the console, for example:
??geom_tile
The teaching material is inspired by a course on Statistical Computing and Data Visualization by Abbass Sharif.
Packages used: dplyr, fivethirtyeight, gapminder, ggalluvial, ggmap, ggplot2, hflights, lubridate, maps, pscl, raster, readr, rmapshaper, sf, tidyr, tmaptools, viridis.
Additional data sources used:
- Armed Conflict Location & Event Data Project (ACLED), https://www.acleddata.com
- United States Environmental Protection Agency Air Data, https://www.epa.gov/outdoor-air-quality-data
Creator and instructor: Therese Anders (tanders@usc.edu)
This project is licensed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license.
Feel free to use/adapt the teaching materials, but do not use them commercially/sell them, and share them under the same license.