Skip to content
Harvard-MIT HST Summer 2016 Undergraduate Research Program: A R/Shiny app that plots choropleth maps of the USA (by county) colored in density by median income, fraction of black/white demographics and total population.
R JavaScript CSS HTML Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Data
Sample Images
ShinyApp
README.md
censusPlot.R
cleanData.py

README.md

CWAS_Census

A R/Shiny app that plots choropleth maps of the USA (by county) colored in density by median income, fraction of black/white demographics, total population, blue/red voting % and % below poverty. Uses choroplethR package in R. Raw files were downloaded from the 2000 and 2010 Census and selected columns were extracted into /Data/Raw_All_2000.xlsx and /Data/Raw_All_2010.xlsx. Column titles were changed before exporting to /Data/ACS_2000.csv and /Data/ACS_2010.csv. cleanData.py removes entries in the form xx000 (cumulate data for each state/unincorporated territory) and null values that arise for counties with population of 0 (30113, 51560, 51780). This data was staged in a local MySQL database. Note: Household income includes individuals who are single (the data we used). This is different from family income.

censusPlot.R demonstrates how choropleth plots may be generated. The code in censusPlot.R was copied over to the server-side script in the Shiny App. To run the Shiny App, click "Run App" after opening /ShinyApp/server.R in RStudio.

App functionality:

  • Plots choropleth maps of USA by parameter of user's choice using census data
  • Displays year-by-year comparison for a single variable against all counties, with either census data or data of user's choice
  • Displays ICD9 rates for user uploaded table Below are plots of the USA colored by median income, black population, white population and total population by county.

Make sure to run devtools::install_github('rstudio/DT') (for the buttons in table download) Also to rebuild choroplethr with the custom function, do this in the directory (after modifying usa.R): http://kbroman.org/pkg_primer/pages/build.html

http://bootstraptaste.com/demo/Butterfly/ library(devtools)

build()

install()

Notes: In the 2012 election data Alaska is completely missing and Kalawao county in Hawaii even though there's state level data. County level data just wasn't made available after the 2012 election. Data was pulled from https://www.theguardian.com/news/datablog/2012/nov/07/us-2012-election-county-results-download. State level population estimates for 2012 were pulled from http://www.governing.com/gov-data/state-census-population-migration-births-deaths-estimates.html to generate regional level voting percentages for 2012.

You can’t perform that action at this time.