Visualising the US Liberal Quotient 1948-2015
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
legacy_app adavis Dec 24, 2017


A visual exploration of the Americans for Democratic Action (ADA) Voting Scores. The Shiny app is hosted at

Alternatively, you can fork it on GitHub (preferred) or download a local copy:

shiny::runGitHub(username = "ciflikli", repo = "adavis")

Running the above code as it is will remove the files once the app is closed. You can pass an additional destdir = argument to save the contents locally in a more persistent manner. Make sure you have the required libraries installed; the list can be found at the beginning of global.R script in the app folder.


  • Utilizes about 36,000 US legislator voting scores originally collected by the Americans for Democratic Action (ADA) foundation between 1947 and 2015
  • "[Annual voting records] served as the standard measure of political liberalism. Combining 20 key votes on a wide range of social and economic issues, both domestic and international, the Liberal Quotient (LQ) provides a basic overall picture of an elected official's political position"
  • Actual data compiled by Justin Briggs Blog post | .xlsx file
  • Uses both nominal and adjusted ADA scores Procedure
  • Data can be grouped by year, state, chamber, and party
  • Data visualizations using statebins, plotly, ggExtra and base R


State of the Union

  • Charts a single Year > Chamber > Party combination on a first-level political (administrative) division map displaying either nominal or adjusted LQ score (averaged by state):

Change from Last Year

  • Plots LQ score change from previous year (state-level). Hover info offers the precise amount in percentages:


  • Allows the user to select two states and compare their voting scores in a specified timeframe. Works under the C/P constraint; i.e. it compares representatives from the same party/chamber:


  • Enables looking up specific representatives (3,371 in total) found in the dataset; offers descriptives on Year, Congress, District and plots their LQ scores, mean, and plus/minus one standard deviation:


  • Users can change viridis colour palettes (viridis, plasma, inferno, and magma) and direction (whether higher values are light or dark) throughout the app:

Legacy App