The goal of reschola is to support the workflow for data analysis at Schola Empirica through a set of templates, themes, utilities, and documentation. The vignettes for the package document the Schola Empirica way of doing data analysis and provide guidance and tips on tools and good practice.

Installation

You can install the released version of reschola from Github with:

remotes::install_github(
"scholaempirica/reschola",
ref = remotes::github_release(),
build_vignettes = TRUE
)

r-universe is kindly building and hosting compiled binaries of the current development versions for Windows and MacOS, so you can proceed with a standard call with r-universe repository specified:

install.packages("reschola", repos = "https://scholaempirica.r-universe.dev")

This could be done with remotes as well, by ommiting the ref argument:

remotes::install_github("scholaempirica/reschola", build_vignettes = TRUE)

What is inside the box and how to use it

See the Getting started vignette (vignette('reschola', package = 'reschola')) for an overview of templates, utilities and guidance in this package that together provide the infrastructure and mental models for the Schola Empirica way of working with data and reporting.

Acknowledgments

All content in this package arose out of conversations with the guys at Schola Empirica and their desire to do data analysis well.

The tools and concepts contained in it are heavily inspired by many wonderful members of the R community; references to specific contributions are contained in individual vignettes.

Logo made using https://connect.thinkr.fr/hexmake/. The approach to project templates and some other components is heavily inspired by ratlas, including the reuse of components from hrbrthemes.