StudyDuo
Most people who ask "why can't I finish online courses?" are running into the format, not a personal flaw. A self-paced course removes the deadlines, classmates and class times that carry people through school. In Harvard's and MIT's open courses on edX in 2017-18, about 3% of participants finished, and fewer than half of those who paid did.
What the online course completion rate really is
The most complete public numbers come from the two universities that founded edX. Justin Reich and José Ruipérez-Valiente analyzed every MITx and HarvardX course run from 2012 to May 2018: 565 runs of 261 courses, with 12.67 million registrations from 5.63 million learners. Of everyone who registered, 52% never opened the course materials, and dropout stayed high through the first two weeks.
How many finished depends on who you count:
| Group | 2013-14 | 2017-18 |
|---|---|---|
| Everyone who opened a course | ||
| Said in a survey they meant to finish | ||
| Paid for the verified certificate |
Source: Reich & Ruipérez-Valiente, 2019, Science, supplementary Table S3
Three things stand out. Even people who planned to finish mostly didn't: about 15% did. Paying learners finished far more often, but they were also the most committed to begin with, so the gap doesn't prove the money did the work. And the authors wrote that six years of investment in course development and learning research hadn't meaningfully improved any of these numbers. For everyone who opened a course, the best year was 2014-15, at 5.91%. A shortfall that steady across millions of people points at the design of the courses, not at you.
Why can't I finish online courses? 5 reasons with evidence
| Reason | What the research found | What to change |
|---|---|---|
| Nothing sets the pace | Managing time was one of four main reasons for dropping out, in 34 interviews | A fixed weekly time, yours or a cohort's |
| Nobody notices when you stop | Better interaction between people went with higher grades, in 23 courses | Classmates, a partner or a live room |
| One plan at the start | Planning prompts for 269,169 students lifted activity for a week or two, not completion | A plan you remake every week |
| Distraction eats the study time | Limits on distracting sites, set in advance, made finishing 40% more likely | Decide the limits before the session |
| "Done" was never defined | How learners saw a course shaped how they engaged with it | Choose the modules you need at the start |
1. Nothing sets the pace
In a classroom, the timetable decides when you study. A self-paced course hands you that decision every day, and every day you can put it off. Thommy Eriksson, Tom Adawi and Christian Stöhr interviewed 34 learners from two MOOCs, from people who finished to people who quit. They found four main factors behind dropping out. Two were about the course: what learners thought of its content and of its design. The other two were their social situation and their ability to find and manage time, and the paper's title points at the last one: "Time is the bottleneck."
2. Nobody notices when you stop
Skip a class and someone sees the empty seat; skip a module and nothing happens. Shanna Smith Jaggars and Di Xu rated the design of 23 online courses at two community colleges and compared it with students' grades. The quality of interaction between people in a course was positively and significantly related to how students did. That's a correlation from two colleges, but it matches Reich and Ruipérez-Valiente's reading of the wider research: advisors, tutors and peer groups are the most important supports for online students.
3. A plan made on day one fades
Most advice stops at "make a plan", and the largest test of that advice is sobering. In an earlier test in three courses, a short plan-making exercise at the start had raised completion from 14% to 18% among committed English-speaking students. René Kizilcec, Justin Reich and colleagues then ran the same kind of nudge, and several others, with 269,169 students in 247 Harvard, MIT and Stanford courses over 2.5 years.
The plans raised activity in the first week or two, then the effect faded, and final completion didn't change. A prompt to line up people who would check in on your progress kept students more active for longer, but it didn't significantly raise completion either. Writing about how a course connects to your values helped one group only: students in less-developed countries, in courses where they lagged behind. That result explains much of the frustration behind "why can't I finish online courses?": a plan made once can't carry a course that runs for weeks.
4. Distraction eats the study time
Richard Patterson tested three tools with 657 students in a MOOC. The first let students commit in advance to time limits on distracting websites. The second alerted them when they were on such sites, and the third blocked those sites while the course page was open. Only advance commitment worked: those students spent 24% more time on the course, earned grades 0.29 standard deviations higher and were 40% more likely to finish. Students with alerts or blocking did no better than the control group. What mattered was when the limit was chosen: ahead of time, with a clear head, rather than in the moment a distraction appeared.
5. "Done" was never defined
Some unfinished courses were never meant to be finished. Eriksson's team found that how learners understood a MOOC strongly shaped how they engaged with its contents. If you enrolled to learn one skill, stopping once you have it is a result. Say you took a spreadsheet course to learn pivot tables for work. Once that module is done and you've used them on a real file, the course has done its job. The trouble is drifting away without deciding, which leaves the course hanging over you as something you failed to do.
How to actually finish an online course
- Take one course at a time. Unenroll from, or archive, the others until this one is done. Every open course competes for the same weekly hours.
- Pick a run with a pace. If the course is offered with a start date and weekly deadlines, take that run. It also puts other learners on the same module in the same week, which makes the forum worth reading. If it's self-paced, write a date next to each module before you start.
- Start within two days of enrolling. The edX data shows where courses are lost: before the first page and in the first two weeks. Finish the first module while the reason you signed up is still fresh, then book the second.
- Remake the plan every week. Book the sessions as a repeating calendar event, then every Sunday write which module comes next and exactly when you'll do it. In Kizilcec's data, a single plan's effect lasted a week or two; a plan you remake every Sunday never gets that old.
- Set limits before you sit down. Choose your phone's screen-time limits for study hours ahead of time, the way Patterson's students did, rather than fighting each notification as it arrives.
- Put people around the course. Use the course forum, find a partner taking the same course with these places to look, join an online study group, or work in a live room. Give a group one rule: the same hour every week, and everyone says which module they finished since the last meeting. On StudyDuo, Study Hour is a daily hour that everyone begins together, with no breaks inside it. In the app, "Work on this" puts a task such as "Module 4, Python course" under your name in Focus Room.
- Decide in advance when you'd stop. Write a rule, such as: fewer than two modules in three weeks means switching to a run with deadlines or dropping the course. A rule like that settles the question before it turns into "why can't I finish online courses?" again. If the course stopped being useful, dropping it is a decision, not a failure.
If the course is mostly recorded video, these fixes for watching lectures cover the viewing itself. If what's missing is someone to report to, see what studies of accountability partners found.
Some programs build the company in: see a Coursera study group, an Open University study group and a WGU study group.
FAQ
Does the course itself make a difference?
A big one. Across the courses in Kizilcec and colleagues' field experiment, completion ranged from 1% to 65% among the students studied.
Do most online learners come back for a second course?
Not in the HarvardX and MITx data. Of the 1.1 million learners who started in 2015-16, only 12% took another of their courses the following year.
Are online college classes harder to pass than in-person ones?
Often, yes. Reich and Ruipérez-Valiente note that by most indications students do worse in online courses than in residential ones. The gap is widest for first-generation students, students from low-income families and under-represented minorities.
