A quick note on the early MOOC years
Massive open online courses, usually called MOOCs, became widely known in the early 2010s. Some earlier experiments existed, but 2012 was often described in the press as the “year of the MOOC”, when several large providers launched and enrollments in single courses ran into the tens of thousands. The headline promise was reach: a good lecturer could now teach a global audience at little extra cost per student.
This article is about what happened to course design after that first wave. It does not rank or review any particular provider, and it avoids precise figures, because many early statistics were reported in different ways and are easy to misread.
What the first MOOCs revealed
The early courses tended to copy the familiar university format: recorded lectures cut into short videos, weekly deadlines, quizzes and sometimes a forum. They were open to join, and many people did. Observers then noticed a pattern that came to be widely discussed: a large number of people enrolled, but only a small share finished. Reported completion rates were often low, though the figures vary with how “enrolled” and “completed” are counted.
That pattern was interpreted in different ways. One view was that low completion was a failure of design. Another was that many enrolled people never intended to finish: some wanted to browse, some to sample a single lecture, some to check out the format. Both views contain some truth, and the sensible conclusion is that completion alone is a blunt measure of whether a course served its audience.
Why design moved toward completion
Course designers took several lessons from the early period, most of them about the experience of learners without a teacher present.
- Learners need feedback. Large courses with little personal contact meant students often did not know whether they were understanding the material.
- Fixed schedules did not suit everyone. Rigid weekly deadlines caused many people to fall behind and quietly leave.
- Passive watching is a weak way to learn. Research on how people learn suggests that active practice, such as answering questions from memory, tends to help more than simply watching, although the details depend on the subject and the learner.
- Community mattered. Forums and study groups helped some people persist, but they needed some moderation and structure to work.
In response, many courses added more frequent practice questions, flexible deadlines, self-paced options and clearer signals of progress. Learning analytics gave designers a way to see where learners stopped, and to redesign those points.
What “mastery” means here
The idea of mastery-based learning is older than MOOCs. It rests on the view that most learners can reach a given standard if they get enough time and the right support, and so progress should depend on showing understanding of a unit rather than on the calendar. In its typical form:
- The course is divided into small units with clear learning objectives.
- The learner studies a unit and then checks their understanding with an assessment.
- If the result shows the unit is not yet understood, the learner reviews and tries again.
- Only after reaching a set standard do they move on.
Mastery-based progression is closely related to competency-based education, though the two terms are not always used identically. Mastery models often emphasize repeated attempts and targeted feedback, while competency models often emphasize recognized skill outcomes, sometimes tied to a credential.
How online delivery helps and where it struggles
Digital tools make mastery models easier to run at scale. A system can offer a bank of practice questions, give instant feedback, record which units are complete, and let each learner repeat as needed. Combining this with spaced repetition and retrieval practice, which are well-supported ideas in learning science, gives designers a reasonable foundation.
The difficulties are real too.
- Assessment design is hard. A quiz can show recall without showing understanding. Writing questions that reveal real mastery takes care.
- Automated feedback has limits. It works best for well-defined answers and less well for open-ended work.
- Pace can drift. With no deadlines, some learners stall. Light structure, such as suggested weekly targets, often helps.
- Standards must be set. Deciding what counts as “mastered” is a judgment, and different settings choose different thresholds.
- Recognition varies. A learner who masters a unit may not receive any formal credit for it.
What the evidence does and does not say
Research on mastery learning in conventional classrooms has generally been positive, but the results vary with the subject, the age group and how closely the model was followed. Evidence about open online courses specifically is more mixed, partly because the learners differ so much from each other and because studies define success differently. It is safest to say that mastery-based design is a plausible and widely used response to the problems the early MOOCs exposed, not a proven fix.
Reading the history sensibly
The move from MOOCs to mastery is better seen as a gradual refinement than a clean break. Large open courses still exist, and many borrowed ideas from mastery models, such as shorter units and frequent practice. Meanwhile, mastery thinking has been taken up in schools and training programs that have little to do with MOOCs. For a learner, the practical takeaway is to look for courses that state their learning goals clearly, give regular feedback and let you repeat work without penalty, and to be realistic that finishing a course still depends on time, motivation and circumstances.