eLearning Experts / Evidence

Completion rates and what they are measuring

Frequently the design, occasionally the learners.

Low completion is generally attributed to learner motivation, and comparing courses within one organisation usually undermines that reading: the same population completes some and abandons others at very different rates, which points at the courses.

The diagnostic data is usually available and rarely examined. Where in the course do people stop, and is it the same place for everybody. A consistent drop-off at one point is a design finding with a specific location, and it is normally something identifiable: a long segment, a technical obstacle, an activity requiring other people, or a sudden jump in difficulty.

It is also worth distinguishing between people who never started, who stopped early, and who stopped near the end. Those three have different causes and the aggregate rate conflates them entirely.

Where a course is genuinely optional and the material is reference-like, low completion may not be a problem at all. Somebody who took the two sections they needed has used the course correctly, and measuring them as a failure is a measurement error.

There is one further reason a consistent drop-off point matters, which is that it is usually cheap to fix. The cause is typically a single long segment, an obstacle in the platform or an unclear instruction, and addressing it can move completion by a substantial margin for an afternoon's work, which is a better return than any redesign.