Election Night Reporting Using R & Quarto

on YouTube ↗22m

" summary

Built a full election night pipeline using R and Quarto that pulls from the official vendor database, prebuilds and stores over 400 maps and tables, and deploys a responsive site via Netlify. Uses dbplyr for SQL, targets pipeline with dynamic branching to update only changed precincts, saves artifacts to S3, and renders reactable tables and leaflet maps. Update cadence drops from minutes to ~90 seconds, an election with nearly one million ballots runs without crashing, and a dynamic “close races” page uses an 8% threshold.

AI-written summary · about

" chapters (7)

1. 0:00Introduction
2. 1:02Why R and Quarto
3. 3:59Real-time data visualization
4. 9:12Building the pipeline
5. 13:17Replicating workflow
6. 16:30Optimizing pipeline
7. 20:46Success and demo

" related talks

• 10mVisualizing voting resultsHernan Martinez · AWE 2015
• 01h 42mRecreating, Understanding, and Visualizing FiveThirtyEight's...Matthew Brems, Joseph Nelson · PyData DC 2018
• 28mDelicious Democracy: snagging and mapping citizen dataKeith Moss · FOSS4G SotM Oceania 2019
• 15m"Red State, Blue State: ElectionMatthew Ericson · Where 2.0 2009
• 20mJournalism with RStudio, R, and the tidyverseLarry Fenn · rstudio::conf 2019
• 38mThere Are No Shortcuts in Organizing, but Technology Sure Does HelpVicki Crosson, Shane Moore · Strange Loop 2022
• 21mTracking COVID 19 with RubyOlivier Lacan · RubyConf 2020
• 01h 21mInteractive Data Visualization Tools in RAshton Drew · PyData Carolinas 2016