In previous sections, we focused on how often students
access online content pages and online problems. It also matters how much time the students spend with each resource. It
is difficult to realistically assess time on task since online
access data only shows when new resources are being accessed, but there is no data on what happened in between:
when there are 2 h between successive transactions, the student may have spent 2 h pondering the problem or the student may have stepped out to have a coffee. To get some
handle on this, we looked at the time difference between two
transactions of the same student on the same online resource,
reflecting what may be called “access frequency.” Examples
for such transaction time differences would be the time between pulling up a problem and submitting an attempt, flipping to and from a resource, or the time between two subsequent attempts to solve a problem. The longer these times
get, the more meaningless they are, as the student could have
done an increasing amount of unrelated activities; on the
other hand, for times of a few minutes, the student may indeed have stayed on task. There was a total of about 900 000
subsequent transactions on the same resource for male and
1 250 000 subsequent transactions on the same resource for
female students.