Open plschmitz opened 6 months ago
As it turns out, among the 2634 institutions included in our IPEDS import, there are only 3 Private R1 institutions and 3 Private R2 institutions in EPSCoR jurisdictions, vs 36 Private R1 institutions and 37 Private R2 institutions in non-EPSCoR jurisdictions.
If we can implement the counts in each category as part of the filtergraph UI, users will be able to see at least part of the problem here.
Adding Lauren as this is a good intro to writing up analyses of the data. Question: where do we add this? Is it in the intro, quick start, a separate section in the help docs? Almost certainly a blog post (as well).
I think we could do the following:
And I agree with a blog post AND making this aspect clear in any training (a couple examples), reports (maybe exhaustive-ish?), etc.
Pushing to 2.2. Adding counts to the legend is a fair amount of work. We can do the blog post independent of the software release cycle.
Add a caveats on viewing data section to the guide.
Use the case of comparing private institutions by EPSCoR status. It appears to indicate that there is a huge difference between private institutions in EPSCoR and non-EPSCoR jurisdictions. If the user investigates by filtering the institutions by carnegie classification, it emerges that the dataset is skewed such that institutions from EPSCoR jurisdictions are mostly smaller (not R1 or R2) while the non-EPSCoR jurisdictions (like the dataset as a whole) is heavily weighted to R1 and R2 institutions. Said differently, most of the EPSCoR jurisdictions have large publics in the dataset, and few or no large privates. See attached graph