Personally, I would caution against reading too much into the total activity time.  While it is useful to know if the user has spent at least some time in a course vs never accessed the course, we have seen the data get significantly skewed and become unreliable for more in-depth analysis. 

For example, we have seen some users in a course with extremely high amounts of total activity that seemed unrealistic.  As far as we can tell from some limited testing, it appears that the total activity time may continue to increase if they are logged in to the course with an active session, even if they aren't actively on that tab in their browser.  Thus, by leaving it open overnight this can significantly increase the total activity time even though the user isn't actively engaging in the course.