You will see there are subdirectories for Raw and Clean Data, and for RScripts and DoFiles used to generate content. RScripts are always run from the RScripts directory, while DoFiles are run from the experiment directory by declaring a global variable. All scripts have the name of the figure or table they are generating.
File cleaning processes are generally done in R and take Raw data as an input and have DB data as an output. However, some slight variable recoding is done in STATA in some cases when running regressions.
We share all our work through github, and update the Overleaf project through Dropbox.