Understanding user attended and activity times in Enrollment data

[nardell]

We are looking at a simple warning system (alerting advisors about students who may be slipping behind) based on user attendance and activity in their course-shells. This report is attended for the summer sessions classes going on, where time is short and the deadlines to drop classes are coming up soon. The report needs to search through all summer session course shells that are published and find cases where (in the first week of classes) there has been little or no activity from a student. From the Enrollments API there seems to be 3 different values in the the returned Enrollment object that on the surface could be useful for this purpose:

// The last activity time of the user for the enrollment, in ISO8601 format.  
"last_activity_at": "2012-04-18T23:08:51Z",  
// The last attended date of the user for the enrollment in a course, in ISO8601  // format.  
"last_attended_at": "2012-04-18T23:08:51Z",  
// The total activity time of the user for the enrollment, in seconds.  
"total_activity_time": 260,

From looking at discussions about last_attended_at, I see that this is a value logged by Teachers, and may not be useful for the purposes of this report (not sure if faculty reliably log this data). On the other hand total_activity_time and total_activity_time are better choices but not ideal. The people I present this report to will need to have this understanding. I wand to ask the community if there are any better options for finding cases where students are who enrolled in published Canvas courses are not participating, and may need a reminder. We want to make sure students to not make costly mistakes (both financially and academically).

Thanks

Michael Nardell