A bit of background. For real life projects you will probably create a repository for each project (or at least each investigator). We cannot have that many private repositories for this class. So, instead each project is a directory. Probably what you would have done anyways before you had these cool tools to drive you crazy.
We have created a high level folder structure for you when you cloned the class repository. Within each project folder create the following folders/directories:
Data Folder in your DATA directory in the Repositories folder.
Create a Code folder that will contain your “literately programmed” statistical software code files.
Create a Docs folder that will contain communication with the investigator, the project description and other resources you use to complete the project.
Create a Reports folder that will contains the graphs that you generate by the Code, the current version of report(s) for the investigator, your statistial appendix.
Note, these folders don’t need to have lots of versions of the file code and reports files. Let version control do its work.
You need a dataclean(.SAS/R) that cleans the data.
You need a descriptives(.SAS/R) that makes means, std, graphs, etc. This likely creates the crude Table 1 in your reports.
You need an analysis(.SAS/R) that does the actual modeling, testing. With version control you can get into the habit of deleting code once you know you won’t use it. This is instead of keeping it in your file but commenting it out. You can put a comment in the code (yes, literate programming) that says you did these things and chose to go a different way.