Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Feed of basic statistics in tidy data format #1038
This is a medium-term enhancement request, but @tracykteal and I have tried at times to answer specific questions using the wealth of data available in AMY. Short of querying the normalized web of tables, this can be challenging.
It would be nice to create some DB views that are made available using tidy data(PDF) best practices. I'm imagining this as a few views pertaining to instructor, workshop, partner, learner segments of the data. I think this can likely be done in a way that no personally identifying information is made available if we heavily aggregate. Ideally I'd like the feed to be open in order to improve transparency and involve the community in conducting analysis of it.
Example datasets I'd like to see for starters:
Then we can write analysis scripts that keep our relevant slides and pages that draw from AMY data up-to-date. This minimum set above would be so helpful to answer routine questions I get from prospective partners.
These plots are already available in AMY in "reports" menu (i.e. here; the plots are not displayed due to a bug, but this issue will be fixed in the next release) and raw data is available as REST endpoint, although both plots and raw data are not open to public.
This is easy to add (~10 lines of code).
These are a bit more tough to add, but I think these can be implemented as new REST endpoints, so that you can download it as csv/json/yaml. @pbanaszkiewicz any comments about implementation? @jduckles do you want somehow to visualize these datasets? If yes, what kind of plots do you want?