Visualizing Results From College Scorecard Data

This project visualizes the data from the college report card, https://collegescorecard.ed.gov/data/, using R.

To create HTML page using RStudio

You can simply run scorecard_county_map.Rmd in RStudio using the Knit HTML button at the top of the screen.

To create HTML page using R and rmarkdown

To create the HTML file from the primary script file, open an R session in the ./scripts directory and use

rmarkdown::render('scorecard_county_map.Rmd')

Required data

Aside from the files hosted in the repository, a few other larger data sources must be in the ./data/ folder:

• MERGED2011_PP.csv from the College Scorecard Data
• gz_2010_us_050_00_500k.* cartographic boundary files from the U.S. Census Bureau. Download the 500k file from the page (or directly from this link) and unzip all files in the ./data directory

Required R packages

The following R packages and their dependencies are required:

libs <- c('dplyr',                 # make data wrangling easier
          'gstat',                 # kriging
          'geojsonio',             # to handle geojson data type
          'htmlwidgets',           # way to bind javascript libraries
          'knitr',                 # combine R code and markdown
          'leaflet',               # interactive mapping module
          'maptools',              # mapping projections
          'RColorBrewer',          # create color palettes
          'readr',                 # faster reading of large csv files
          'rgdal',                 # interact with gdal GIS framework
          'rgeos',                 # interact with geos GIS framework
          'tidyr')                 # make data wrangling easier

rmarkdown is also required to produce the final HTML file if you choose not to use RStudio.

Note that the geospatial packages (e.g., rgdal and rgeos) may require the installation of geospatial libraries on your local machine. For OS X, many can be installed using prepackaged binaries found here.