Alternate Attribute Report Chart which uses % from each course rather than raw number

[mattminnick]

At our meeting in June we identified an issue with where courses with large numbers of students (especially those with multiple measures of an attribute) dominate the attribute report chart, e.g., in the 2018-2019 year, Attribute 1 for some sample courses we're measuring looks like:
Course BE MA ME EE
MATH 1ZB3 1752 2526 1995 1320
ENGPHSY 2P04 2 5 10 27
ENGPHYS 3BA3 2 3 12 29

Currently the CEAB Attribute Report adds up the number in each category and uses this to generate the chart, which in this case means the summary chart is only talking about MATH 1ZB3.

Because we still need to measure the large common classes and the smaller department-specific classes in our GA report, one way to not have the large common classes dominate the data is to only compare the percentages from each course rather than the raw number. Therefore:
Request is to create an Alternate Attribute Report Chart which calculates percentages BE, MA, ME, & EE for each course then uses the average BE, MA, ME, & EE across all courses to generate the chart:

Since the table was not unanimous which method was better, perhaps this could be an alternate chart to the current one rather than a replacement.

[smiths]

Thank you @mattminnick for creating this issue. The screen shots of the specific data is very helpful for capturing what you are after. A short-term option you would have would be to separate out your students from those in other departments. Instructors might not want to bother, but they do have the data with the classlist, so they could just do the measures on the subset of students where the measures apply.

I've labelled this issue as an enhancement. It would be nice to have this option in the future, but for the current round of work on MEASURE, I think @arana2 should focus on the large report that includes all course reports. He'll be updating us on this during today's Graduate Attributes Committee meeting.

[mattminnick]

Wolfram just reminded the GA group that a few meetings ago we agreed that we would try to switch to percentage-based measures. Bumping this to remind about the request.