PopAging DataViz
Visualizing Population
Data in
A THREE-DAY WORKSHOP AT FLORIDA STATE UNIVERSITY
October 22nd—24th, 2024
Around 9 AM – 4:30 PM
Devoe Moore Conference Room (Bellamy Building)
Instructor:
Sakeef M. Karim
skarim@amherst.edu
About the Workshop
PopAging DataViz is a workshop on visualizing population data within ’s graphical ecosystem. During the workshop, we will be using the ggplot2
library as our workhorse. Over the course of three days, workshop participants will be exposed to many of the principles undergirding the Grammar of Graphics framework for visualizing data (see Wilkinson 2005) — a grammar that serves as the lifeblood of ggplot2
and its many extensions.
Then, the course will slowly build in complexity. Concretely, participants will be exposed to modules on (1) generating population pyramids and other plots that are germane to population research (e.g., Lexis diagrams); (2) leveraging interactivity and animations to bring data visualizations to life; and (3) visualizing statistical quantities of substantive interest (e.g., average marginal effects, adjusted predictions) to clarify model results.
The workshop will draw on a range of packages—including, but not limited to, tidycensus
, leaflet
, mapview
and cansim
—as well as a series of hands-on exercises designed to concretize high-level concepts and principles related to data visualization. All workshop materials are available (and will be maintained) on this course website and the companion repository.
Sponsors
This short course is being co-hosted by the Consortium on Analytics for Data-Driven Decision-Making (CAnD3) and our partners at Florida State University, including the Pepper Institute on Aging and Public Policy, the Center for Demography and Population Health, and the College of Social Sciences and Public Policy.
CAnD3 is hosted at McGill University and supported in part by funding from the Social Sciences and Humanities Research Council of Canada (SSHRC).
Inspirations
Cédric Scherer’s (2022) course on Graphic Design with ggplot2
, Andrew Heiss’ (2024) course on Data Visualization with R
, and the third edition of ggplot2: Elegant Graphics for Data Analysis
(Wickham, Navarro, and Pedersen 2023) all serve as inspirations for this three-day workshop.
Preliminaries
Programming Language
As noted, this short course will be centred around—or anchored to—the programming language for statistical computing and visualization. If you do not have on your machine, please download the open-source software by clicking here before following the relevant prompts.
Integrated Development Environment
Participants are encouraged to run their code using RStudio
,1 a powerful open-source IDE optimized for . If you do not have RStudio
on your machine, you can download the application by clicking here.
Packages
Before the workshop, please download or clone the companion repository. Thanks to the wonderful renv
package, simply running renv::restore()
before executing any lines of code in your source-code editor should ensure that all required packages are automatically available to you.
If you’ve never cloned a repository, please watch the video embedded below.2
API Keys
You’ll need to obtain valid API keys to use the tidycensus
and cancensus
libraries.
Fonts
Please download and install the IBM Plex Sans
and Inconsolata
typefaces via Google Fonts or by downloading the compressed file embedded below:
References
Footnotes
Participants should, of course, feel free to use other IDEs (e.g.,
Positron
), environments, or source-code editors to run the material featured in this short course.↩︎The video details how to use
RStudio
to easily clone a repository hosted on GitHub. There are, of course, other ways to cloneGit
repositories — inclusive of the command line.↩︎