An open API service indexing awesome lists of open source software.

https://github.com/btskinner/oa_online_broadband_rep

Replication files for Skinner (2019) Making the connection: Broadband access and online course enrollment at public open admissions institutions.
https://github.com/btskinner/oa_online_broadband_rep

broadband journal-article online-learning replication research-in-higher-education

Last synced: 10 days ago
JSON representation

Replication files for Skinner (2019) Making the connection: Broadband access and online course enrollment at public open admissions institutions.

Awesome Lists containing this project

README

          

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](https://link.springer.com/article/10.1007/s11162-018-9539-6). _Research
> in Higher Education_, 60(7), 960–999.

# Requirements

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

1. [R](https://cran.r-project.org)
2. [CmdStan](https://mc-stan.org/users/interfaces/cmdstan)
3. [Bash](https://www.gnu.org/software/bash/)
4. [git](https://git-scm.com)
5. [sqlite](https://www.sqlite.org/index.html)

To build all data, including broadband databases, you will need
[unar](https://theunarchiver.com/command-line) to unpack the large
compressed files. The easiest way to get it is with
[homebrew](https://brew.sh):

```bash
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

```bash
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:

```bash
# --- 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`

```bash
cd ./scripts/bash
./make_bb.sh ../../data/broadband/zip ../../data/broadband .
```

3. Download required R packages, found in `get_packages.R`
4. Clean IPEDS using `clean_ipeds.R`
5. Clean Broadband using `clean_bb.R`
6. Clean data using `clean_data.R`
7. Download and build
[CmdStan](https://mc-stan.org/users/interfaces/cmdstan)
8. Compile `*.stan` model scripts
9. Run analyses using `./scripts/bash/run_stan.sh`

```bash
# example call (assuming in ./scripts/bash directory)
./run_stan.sh ../stan/sl_normal ../../data/cleaned sl_normal full ../../output both ../r
```

10. Make figures using `make_figures.R`
11. 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.