4 Data analysis
In a GIN workflow, it may also be a submodule,
mirror of a GitHub or GitLab repository.
Content
This folder will contain either code (in python, R or
matlab), or textual description of the analysis, or both.
Organization
The content may be organized following the different
experiments, or by different figures created or a mix of
these two. In addition we have 2 folder already present in
the template:
990\_code\_libraries
Here comes the
dependencies of your software, it might be the dependencies
themselves, or a link/mirror to the dependencies.
You may want to use a container technology to deal with your
code libraries.
991\_preregistration
In order to prove that
your experimental design is preventing harking and
p-hacking, you should describe how you will analyze the data
before you actually start collecting it.
Any document that do that should be saved or mirrored here.
Note that you should have a time stamped version
of this file, proving it was produced
before data acquisition.
One may use osf or
zenodo for this purpose.