A Directory for Transparent Data Analaysis Exercises for Fall 2017

Built on a skeleton directory to jumpstart an analysis project

• Look at the Git history of this file to see how it evolved
• I just changed the title name and added this paragraph here
• I added some project related files

Setup

1. Start a New Project in Rstudio
2. Use Version Control > Git > Repo Url https://github.com/statsccpr/proj_skel.git
3. Use a relevant domain-specific name for your local project (NOT proj_skel)
4. Choose local destination for your local project

• Note: uses /misc/.Rprofile which you can modify

Instructions

• place chickenscratch prototype scripts in /tests/proto/
• place good-standing scripts in /scripts/
• place raw data in /data/raw/
• place processed data in /data/proc/
• place output in /output/
• place writeup in /writeup/
• formal assertion-travisci tests in /tests/

Optional push to Github

• push your local project to new github repo (NOT proj_skel)

0. login github
1. https://github.com/new
2. See this

• git remote add github git@foo_new_name_from_step_1
• git push -u github master

3. See general reference on remote

Optional Collaboraters

1. Tell collaborators to Fork your github repo as their own local repo
2. They make changes and Push to their own global repo
3. Everyone can New Pull Request > Compare Across Forks
4. Pull head fork edits (yours) into base fork edits (theirs)