mytetz › Study guides › Why a person stops an online course
Why a person stops an online course
Most people who register for an open online course never finish it. The clearest measurement covers 2013 to 2018 across 261 HarvardX and MITx courses. In the 2017–18 year, 3.13% of participants completed a course. A participant is a learner who opened the courseware, and 52% of registrants never opened it.
Key points
- The denominator decides the number. Read it before you read the percentage.
- In the same year, 46.02% of learners on a paid verified track completed a course. Commitment changes the result more than content does.
- Among learners who stated an intent to complete, 15.42% did so. Intent alone is not enough.
- The data stop in May 2018. No later figure of this quality exists in a named source.
What does the completion data count?
Reich and Ruipérez-Valiente published "The MOOC pivot" in Science in January 2019. They analysed 565 course iterations of 261 HarvardX and MITx courses, with 12.67 million registrations from 5.63 million learners, between June 2013 and May 2018.
| Academic year | Completion rate |
|---|---|
| 2013–14 | 4.95% |
| 2014–15 | 5.85% |
| 2015–16 | 4.11% |
| 2016–17 | 3.98% |
| 2017–18 | 3.13% |
Three denominators appear in this field, and they give very different numbers. Confusion between them produces most of the bad statistics on the web.
| Group | Who is in it | Completion |
|---|---|---|
| Registrants | Everybody who signed up. 52% never opened the courseware. | Lower than 3.13% |
| Participants | Learners who opened the courseware at least once. | 3.13% |
| Paid verified track | Learners who paid for a certificate. | 46.02% |
What changes the completion rate?
The gap between 3.13% and 46.02% is the most useful fact on this page. The courses are the same. The videos are the same. The difference is the commitment of the learner.
Intent alone does not close the gap. Among learners who stated an intent to complete, 15.42% completed a course. So a plan in your head is worth about five times a casual registration, and about one third of a payment.
- The registration trap
- A free registration costs nothing, so it commits nothing. The low headline completion rate measures that cost, and not the quality of the course.
Why should you distrust the newer figures?
Many pages state a completion rate for 2026. Common examples are "15 to 20% for a free course", "60 to 65% for a paid programme" and "72% for corporate training". We followed each one. Every trail ended at a content site with no named primary source.
So this page gives a figure from 2017–18 with its year attached, instead of a current figure with no source. If you find a newer measurement in a peer-reviewed paper or in a platform's own filing, that number is better than this one. Check the denominator first.
Why does a person stop?
The paper measures the pattern, and not the cause. The four reasons below come from the structure of a self-paced course, and from the study research in the other guides on this site. Treat them as an explanation to test, and not as a measured result.
- No fixed time. A self-paced course has no lecture on Tuesday. A task with no time competes badly with a task that has one.
- No test until the end. A learner who never retrieves feels progress from video alone. The feeling drops as soon as a real question appears.
- A unit that is too big. A three-hour module gives no place to stop, so a learner stops in the middle and does not return.
- No cost to stop. Nobody notices. Nothing is lost. The paid track works partly because it removes this.
Six steps that help you finish
These steps apply the findings from the other guides. No study tests this list as a whole. Each step is cheap, and you can drop one that does not suit you.
- Set a finish date before you start. Write it in a calendar. A course with no end date has no middle.
- Book two fixed slots each week. Two short slots beat one long slot, and the gap between them helps your memory.
- Cut each module into pieces of 20 minutes. Stop at a piece boundary, and never in the middle.
- Close the video and say what it taught. One sentence is enough. This is the retrieval that a course often omits.
- Review last week's gaps before you start new material. Five minutes is enough.
- Add a real cost to a miss. Pay for the certificate, or tell one person your finish date.
Sources
- Reich, J., & Ruipérez-Valiente, J. A. (2019). The MOOC pivot. Science, 363(6423), 130–131. Figures from the authors' own supplementary data. doi:10.1126/science.aav7958
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks. Psychological Bulletin, 132(3), 354–380. doi:10.1037/0033-2909.132.3.354
- Roediger, H. L., III, & Karpicke, J. D. (2006). Test-Enhanced Learning. Psychological Science, 17(3), 249–255. doi:10.1111/j.1467-9280.2006.01693.x
Common questions
Is a 3% completion rate current?
No. It covers the 2017–18 academic year on two university platforms. Treat it as the best measured figure, and not as today's figure.
Does a low completion rate mean that online courses fail?
No. Many learners open a course for one specific part and leave when they have it. The rate measures the behaviour of free registrants, and not the value of the material.
Does a payment really help me finish?
The data show a strong association: 46.02% of paid learners completed a course, against 3.13% of participants. It does not prove that the payment causes the result. A learner who pays is also a learner who already decided.
What do I do when I have already stopped?
Do not restart at module one. Open the last piece that you finished, retrieve it from memory, then continue. A restart from the beginning is the second reason that people stop twice.
Where do you go next?
- How students study now — the survey data, and what it does not measure.
- When to review what you read — the schedule for step 5.
Start with a topic
A topic is smaller than a course, and you can finish one today. Pick one, read a short step, and go deeper where you want to.
Open the catalogue