Introduction to Data Visualization in R

This workshop is designed to serve two related goals: 1) to teach you how to use flexible, powerful visualization tools in R to understand important patterns in your scientific data, and 2) to show you how to translate your scientific findings into clear and impactful figures for publications, presentations, and grant applications. I will strike a balance between the conceptual and practical aspects of data visualization. I will also provide detailed examples and code for major graphical idioms in social and behavioral sciences (e.g., time series plots) so that you can refer to these after the workshop is over.

Our workshop is designed for graduate students, post-doctoral scholars, faculty, and research scientists in the behavioral, social, and health sciences. Although we will focus on broad principles of graphical design, the primary purpose of the workshop is to prepare participants to create figures for understanding data and for scientific publications and presentations. Students are not expected to be R experts. We will walk through the mechanics of using tidyverse packages, including ggplot2, to prepare and visualize data. Students will also be provided with full R source code for all examples. We expect that participants will have a basic understanding of R or another functional programming language (e.g., Python). If you are not yet familiar with R, there are many free resources, including this overview prepared by Rick Gilmore

Chapter 1: Data Wrangling in R
1.1 The tidyverse paradigm for readable, transparent data processing
1.2 Preparing data for visualization in ggplot2
1.3 Recommendations for data management and reproducibility

Chapter 2: Conceptual foundations of data visualization
2.1 Principles of data visualization and graphical grammar
2.2 The nomenclature of visual displays
2.3 Maximizing readability and accuracy: Visual and graphical perception
2.4 Principles of graphic design, style, and use of color

Chapter 3: Understanding graphical idioms in social and behavioral sciences
3.1 Choosing the right idiom for the question
3.2 Refining the idiom to emphasize the key points
3.3 A guided tour of quantitative visualization
3.4 Learning from good and bad examples: deconstructing figures to understand what works

Chapter 4: Visualizing continuous data
4.1 An overview of ggplot2 syntax
4.2 Saving visuals as high-quality vector or raster graphic files
4.3 Mastering the foundational idioms: scatterplots, bar charts, line charts, and boxplots
4.4 Creating multi-layer figures: combining raw data with statistical summaries

Chapter 5: Visualizing categorical and count data
5.1 Tabular displays: how to make effective contingency tables
5.2 Graphical displays: normalized stacked bar charts and mosaic plots
5.3 Displaying categorical and continuous data: generalized pair plots, heatmaps, and small multiples

Chapter 6: High-impact visualization: preparation figures for presentation and publication
6.1 Translating a graph that makes sense to you into one that will communicate effectively to others
6.2 Perfecting the plot: margins, color palettes, font sizes, and captions
6.3 Using image editors to assemble and refine multi-panel plots
6.4 A start-to-finish example of creating a complex publication-quality figure
6.5 Recommendations for presenting figures in a talk

Software demonstrations are conducted using R and Inkscape, which are free, open-source software. For creating multi-panel figures, some demonstrations will also use Adobe Illustrator, which is available to many academic researchers at low or no cost. Participants will be asked to follow along and run R code that will be provided, so it is ideal to have a computer with R installed available during the class. Prior to the workshop, participants will be provided with instructions for setting up their local computers to have all necessary software for completing the workshop. Many of the lectures focus on principles of data visualization and graphical design, and these lectures do not depend on a particular software package.

Introduction to Data Visualization is a four-day workshop originally taught live via Zoom by Michael Hallquist. Daily lectures were held from 9:00 to 5:00 with morning, lunch, and afternoon breaks. Sessions consisted of comprehensive lectures, detailed presentation of real-data examples with live demonstrations in R, response to participant questions, and general discussion.

Self-paced participants will receive 12-month access to unlimited viewing of all original lecture recordings as well as indefinite retention of extensive PDFs of lecture and computer demonstration notes and all associated computer scripts and data files.