This repository contains R template for analysing data from experiments and surveys and justification to follow certain conventions and structures. This document is available as a rendered html at https://tuomaseerola.github.io/R_template/.
You can check out the slides about reproducible research or see R_template in action which uses a (too) complex behavioural dataset to clarify the steps of the process.
More extensive discussion and examples will be available in the repository related to book titled Music and Science – Guide to Empirical Music Research.
If you do have access to computer able to run the software, you can also start the analysis in RStudio and R as an independent process in a browser by using Binder, 
The folder structure is described in detail in the report, but it the main purposes of the process are clearly separated into separate folders:
/dataData in read-only format (preferably CSV or TSV format)./mungeAll operations to pre-process, recode, or trim data./scrAll actual R scripts used in the analysis./figuresOutputs from the scripts/docsOutputs from the reports
In this repository, contents.R will be needed to reproduce the analysis in R. It contains an example of how the different stages, structures and processes in the analysis can be executed in a coherent order and manner (i.e., loading, transforming, screening the data, and then visualising, applying statistical analyses, creating figures, and tables). This is designed as a small tutorial for reproducible research from the perspective of our needs in music and science.
report.Rmd will create the report that incorporates comments and the actual analyses and produces either html or pdf file (report.html, report.pdf) in the docs folder.
runtime.txt and install.R are auxiliary files for Binder and not actually needed in the analysis.
Proceed to either a fuller explanation of the process at https://tuomaseerola.github.io/R_template/ or check the quarto slides about R_template in action to explore the steps of the analysis process.
This repository and the documents cannot be considered as a quick R tutorial or a statistics tutorial but simply a way to explain to PG students and collaborators of how clear analyses schemes can be created, followed, and shared. This has helped us to be productive, minimise errors and speak the same language, even though sharing is usually done internally (Dropbox or OneDrive) and only rarely through Github or OSF. For further tutorials, please seek more information about reproducibility in science and RStudio online learning pages.
While I here advocate the use of R, Python and jupyter notebooks would be equally good choice for transparent and reproducible analysis workflows.
Tuomas Eerola, Durham University, UK