Skip to content

flow-description.md

Latest commit

 

History

History
50 lines (27 loc) · 1.87 KB

flow-description.md

File metadata and controls

50 lines (27 loc) · 1.87 KB

Reproducibility Instructions

Environment

In order to start analysing data you will need to install R and RStudio on your computer. Then download this repository as a zip archive, unzip, and run the map.Rproj file to initiate RStudio session.

SIDE NOTE: All the changes you make will be local to your machine. Optionally, you can install github client, if you would like to version control your developments and share them with the master copy. If the last sentenced confused you, feel free to ignore it.

Open RStidop session using .Rproj file in the root of the repository.

Data

This repository does not contain any actual data. You must provide your own data files.

  1. Log in (create the account if haven't already)
  2. Proceed to Data Download
  3. Download Core Fat Files shown on the image below. Save in ./data-unshared/raw/RAND-YYYY-MM-DD
  4. Download the RAND HRS Data file
  5. Unzip the files into the same folder

Scripts

0. Ellis Island

./manipulation/0-ellis-island.R

Imports the raw .SAV files (SPSS format) and save unaltered data in .RDS format. Extracts and saves meta data for each file (one per selected wave). Adjust the path_folder definition as needed.

1. Scale Assembly

./manipulation/1-scale-assembly.R.

Using the extracted meta data This will use dead meta data files to subset and rename relevant variables.

2. Laundry Room

./manipulation/2-laundry-room.R

Further cleaning of the data.

Analysis

For a quick start, open ./reports/look-at-data.R and save a copy under a new name in a separate folder in ./sandbox/.