Skip to main content

Introduction to Reproducible Data Science

Ch. 5 Anatomy of an R Project

5.1 Overview of Approach

R Projects (opened with .RProj files), are a helpful organizing structure for data science projects. By opening the .RProj file, all paths become relative, making it easier to reference files in your code (data/file.csv, vs, C:/Users/nelson/Documents/NestedFolder/Project/data/file.csv).

5.2 Creating an R Project

In RStudio, navigate to: File > New Project. Next create a project from a New Directory if you do not have a preexisting folder of code, and Existing Directory if you do.

5.3 Project Organization

Within the R Project folder, I typically organize the folders and code as follows:

  • data: folder containing all project data that is processed locally.
  • scripts: folder containing all scripts to run against your data.
  • output: folder to store any deliverables created in your scripts.
  • reference: any materials (PDFs, Documents, Presentations) to support your work.

For example, in a research project your folder may look something like this:

project_name
  project.RProj
  data
    demographics.csv
    rt.csv
  scripts
    load_install_libs.R
    load_data.R
    preprocess_data.R
    visualize_data.R
    model_data.R
    pipeline.R
  output
    data
      tidy_final_dataset.csv
    plots
      figure1.png
      figure1.pdf
      figure1.eps

Click here to download a sample pipeline