Reproducible Data Analysis Workflow

I recommend the tutorial, “A Reproducible Data Analysis Workflow With R Markdown, Git, Make, and Docker” as a starting point for R-based data analyses [@peikert].

I also recommend using the integrated development environment (IDE) for R, RStudio.

Component Description
RMarkdown

Dynamic document generation

  • “literate programming paradigm”

  • Eliminate copy and paste errors

  • knitr (export to other formats such documents, presentations)

  • papaja (APA formatting of results in text)

  • stargazer (journal ready tables)
Git

Version control tracking

  • Commits = “snapshots”

  • Commit Message = Description of update

  • Repository = Collection of commits

  • Hash = unique identifier

Note

RStudio has a GUI for Git.
Make Dependency management
Docker

Containerization

  • Virtual software environment

  • Reproduce results independent of host operating system