Engineering study schedule: spaced problem practice that survives the semester
Published · Updated · by the Paced team
What the research says about problem practice
- Spacing the problems is what lasts. In two experiments with 216 college students learning one kind of mathematics problem, splitting ten problems across two sessions a week apart gave no benefit on a test one week later and an extremely large one four weeks later; doing nine problems instead of three in one sitting (overlearning) had no effect at either delay (Rohrer & Taylor, 2006).
- Retrieval beats restudy. A meta-analysis of the testing effect found that tests requiring recall gave larger benefits than tests of recognition, which supports effortful retrieval (Rowland, 2014). Re-solving a problem is recall; rereading its solution is recognition.
- The gap should grow with the deadline. Across 317 experiments, the gap that maximised retention grew with the time until the test (Cepeda et al., 2006). A midterm in three weeks and a final in fourteen call for different paces.
- Students misjudge this. With flashcards, spacing beat cramming for 90% of participants, yet 72% believed cramming had worked better (Kornell, 2009); a review of ten techniques rated distributed practice and practice testing high utility, rereading low (Dunlosky et al., 2013).
The semester as topics
- One topic per problem type and per course. “Nodal analysis”, “second-order ODEs with constant coefficients”, “mass balance on a reactor”: the thing the problem set asks. A chapter title is not a topic.
- The first study is the problem set. Working the week's problems with the notes open is the first study (1 h 30 to 3 h). The reviews are shorter: two or three problems of the same type, closed notes, then a comparison with the worked solution.
- Pace by exam date. Material for the final: growing gaps (Standard: D0, D+1, D+3, D+7, D+15, D+30…). Foundations you will reuse next year: fewer, longer gaps (Light: D0, D+2, D+7, D+21, D+60, D+150). Material for a midterm in three weeks: tighter gaps, then back to Standard once it is over.
- Mix the courses. A review session that moves from circuits to thermodynamics to statics spaces each one and forces you to pick the method, as the exam will.
- Labs and deadlines are fixed; reviews flex. Measure an ordinary week, keep one day off, and let the plan move reviews around the deadlines rather than the other way round.
What a semester costs in reviews
Take 5 courses, one new problem-type topic per course per week, 1 h 30 each for the problem set, on the Standard pace, with three hours on six days (18 h a week) once lectures and labs are removed. Our calculation, from the pace and the default review curve of Paced:
| Week | Problem sets | Reviews | Total | Fits? |
|---|---|---|---|---|
| 1 | 7 h 30 | 3 h 47 | 11 h 17 | yes |
| 2 | 7 h 30 | 5 h 50 | 13 h 20 | yes |
| 3 | 7 h 30 | 7 h 10 | 14 h 40 | yes |
| 4 | 7 h 30 | 7 h 10 | 14 h 40 | yes |
| 5 | 7 h 30 | 8 h 15 | 15 h 45 | yes |
| 6 | 7 h 30 | 8 h 15 | 15 h 45 | yes |
| 7 | 7 h 30 | 8 h 15 | 15 h 45 | yes |
| 8 | 7 h 30 | 8 h 15 | 15 h 45 | yes |
| 9 | 7 h 30 | 8 h 42 | 16 h 12 | yes |
| 10 | 7 h 30 | 9 h | 16 h 30 | yes |
| 11 | 7 h 30 | 9 h | 16 h 30 | yes |
| 12 | 7 h 30 | 9 h | 16 h 30 | yes |
| 13 | 7 h 30 | 9 h | 16 h 30 | yes |
| 14 | 7 h 30 | 9 h | 16 h 30 | yes |
Every week fits; the peak is week 10 at 16 h 30. The reviews of September land in October and November, on top of the new problem sets: that is the week-eight wall every engineering student knows. The usual fixes, in order: smaller topics, the Light pace for the foundations, and a budget measured rather than hoped for. Try your own numbers in the schedule calculator; the general method is in how to plan revision.
When the week goes wrong
A lab report or a project deadline will eat a week. Move the reviews, do not delete them: a late re-solve still counts, and scheduling of study, not late nights, was what went with higher grades among 324 undergraduates (Hartwig & Dunlosky, 2012). See how to catch up without starting over.
Paced plans the re-solves of every problem type on their days, inside the hours left by lectures, labs and deadlines, moves what does not fit, and warns you before the week it will not fit at all. Any course, any field.
Questions
Does spaced repetition work for maths and engineering, not just facts?
The experiment closest to engineering coursework says yes: college students who solved ten problems of one kind in two sessions a week apart did much better four weeks later than students who solved all ten in one sitting, while the one-week test showed no difference. Extra problems in one sitting (overlearning) did nothing. Spacing problem practice is what pays at the exam, weeks later.
How do I turn a course into topics?
One topic per problem type or per week of lectures: the kind of problem a problem set or a past exam asks, not the chapter title. A topic should be something you can first-study in one or two hours and re-solve from a blank page in a quarter of that.
What does a review look like for a problem set?
Re-solve two or three problems of that type from a blank page without the worked solution, then compare. Rereading the solution is the review that feels best and tests least. If you cannot start the problem, the topic needs another first study, not another review.
How do I fit reviews around labs and problem-set deadlines?
Count the deadlines as fixed, measure what is left per day on an ordinary week, and plan reviews only inside that. Reviews are short; the trap is adding new topics faster than the reviews they generate can fit. The table on this page shows when that happens.
Does Paced contain engineering content or solutions?
No. Paced is a planner: you bring your course notes, problem sets and past exams, and it schedules when to first-study and re-solve each topic inside the hours your timetable leaves. It does not teach and it does not promise a grade.
Sources
- Rohrer, D., & Taylor, K. (2006). The effects of overlearning and distributed practise on the retention of mathematics knowledge. Applied Cognitive Psychology, 20(9), 1209–1224. 216 students solving one kind of maths problem: splitting ten problems across two sessions a week apart did nothing at one week but helped a lot at four weeks; extra problems in one sitting (overlearning) did nothing.
- Rowland, C. A. (2014). The effect of testing versus restudy on retention: A meta-analytic review of the testing effect. Psychological Bulletin, 140(6), 1432–1463. Meta-analysis of the testing effect: recall tests gave larger benefits than recognition tests, which supports effortful retrieval.
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. Meta-analysis of 839 assessments in 317 experiments: the gap that maximises retention grows with how long you need to remember.
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. Review of 10 techniques: practice testing and distributed practice rated high utility; rereading and highlighting low.
- Kornell, N. (2009). Optimising learning using flashcards: Spacing is more effective than cramming. Applied Cognitive Psychology, 23(9), 1297–1317. Three experiments with flashcards: spacing beat massing for 90% of participants, yet 72% believed massing had worked better.
- Hartwig, M. K., & Dunlosky, J. (2012). Study strategies of college students: Are self-testing and scheduling related to achievement?. Psychonomic Bulletin & Review, 19(1), 126–134. Survey of 324 undergraduates: self-testing and scheduling were associated with GPA; low performers studied late at night more often and all students were driven by deadlines.