My colleague and I have been discussing this topic and it seems time to bring in the big brains.
At our university, we run the same courses every semester with new students (like any university). Of course, there might be some small changes each semester but we try to minimize those. We want to use the data we collect to train models, develop interventions, etc.
The question is, which is better. Should we reset the course each semester or should be backup and restore it to a new course? More importantly, why exactly should we do that?
I understand reset will remove the user data from the course. The logs will remain but all forum posts, grades, etc. will disappear. That said, I believe the grades will still be connected to the user but not to the course.
It seems to be that resetting a course is going to lose data and, being the data-hungry person that I am, I would think it would be best to keep all the data and simply run the subsequent courses as new courses. However, my colleague believes that reset would be cleaner (than restore), and would not impact the LA in any way. I can see his point but the technician in me can only see that at least some data is going to be lost in the reset process.
Note, the concerns at this point are only about LA. We are not concerned about the student experience of no longer being able to go back and review their old materials online. While that issue might factor into the final decision, here I really want to focus on the LA issues.
I would love to know all the details so we can make a decision on this.