Replication files for Skinner (2019) Making the connection: Broadband access and online course enrollment at public open admissions institutions.
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
figures
scripts
tables
.gitignore
README.md
makefile

README.md

This repository contains the replication files for

Skinner, B.T. (2019). Making the connection: Broadband access and online course enrollment at public open admissions institutions. Research in Higher Education, 1-40.

Requirements

To run the core analyses, you will need the following programs on your machine (tested on MacOS):

  1. R
  2. CmdStan
  3. Bash
  4. git
  5. sqlite

To build all data, including broadband databases, you will need unar to unpack the large compressed files. The easiest way to get it is with homebrew:

brew install unar

You can also use Homebrew to update Bash, git, and sqlite if you want, though that shouldn't be necessary a relatively modern machine.

To replicate analysis and tables, clone or download the repository, and choose one of the following methods:

Makefile

If your machine has make, simply run the makefile from the terminal

make

This makefile will

  1. download and install all required R packages
  2. download all required data files
  3. build/clean data
  4. run analyses
  5. knit tables/figures

OPTIONS

At the top of the makefile, adjust the settings to best your machine. The defaults should work for most users:

# --- settings -----------------------------------

# set to https or ssh (probably https for most)
git_type := https

# cores (used when building cmdstan)
cores := 4

Piecemeal

You can you can also run the scripts one by one. You will need to:

  1. Get the data files listed in ./data/README.md
  2. Build Broadband database using scripts in ./scripts/bash
cd ./scripts/bash
./make_bb.sh ../../data/broadband/zip ../../data/broadband .
  1. Download required R packages, found in get_packages.R
  2. Clean IPEDS using clean_ipeds.R
  3. Clean Broadband using clean_bb.R
  4. Clean data using clean_data.R
  5. Download and build CmdStan
  6. Compile *.stan model scripts
  7. Run analyses using ./scripts/bash/run_stan.sh
# example call (assuming in ./scripts/bash directory)
./run_stan.sh ../stan/sl_normal ../../data/cleaned sl_normal full ../../output both ../r
  1. Make figures using make_figures.R
  2. Knit tables and figures document using table_figures.Rnw

NB Most R scripts assume command line arguments. If you are not using makefile or calling from the command line, you will need to replace args object with a vector of the proper paths indicated in the files.