The Instructure Product Team is working on implementing a better solution to collect your feedback. Read Shiren's blog for more information.
If you have a course in which students are in multiple sections, (such as all students being in the same LEC section but being in different LAB sections) and you try creating groups in a group set with the settings "Split number of students per group" and "Require group members to be in the same section", the resultant groups will not be the proper size (with almost all tests producing differing group sizes at or below 50% the desired group size) AND the groups are made up of students who do not have matching sections (one matching, but not all).
For our real-world use case, a group of 4 with limited visibility is producing the following:
What would be expected here is:
The expected groups have membership numbers as close to the desired amount (4) as possible and all members are in the exact same sections.
Extend the logic of your group creator to check membership in multiple sections and verify that members are in the same sections. Currently, the logic just looks to see if there is one matching section. This is important, as we have many courses with LEC and LAB sections together and also have all of our undergrad and grad combined courses set to have an ALL section that includes all students, regardless of grad status. It is impossible for instructors in these courses to automatically generate groups. Also, the groups should be as close to the right size as possible.
I know the documentation says, "Note: If you require group members to be the same section and split the number of students per group, groups may contain varying numbers of students.", but that is meant to point out that you may have multiple groups with fewer than the desired group membership levels in order to maintain section restriction.
instructor,ta
Community helpTo interact with Panda Bot, our automated chatbot, you need to sign up or log in:
Sign inTo interact with Panda Bot, our automated chatbot, you need to sign up or log in:
Sign in