Why Hours of Study Often Produce Little Progress
A common pattern in maths looks like this: the lesson made sense, the notes look neat, the night before the test felt productive, and the paper still goes badly. Students in this position are rarely lazy. Many are studying in ways that feel good but build little. Others have fallen behind in a subject where each topic rests on the last, so a gap from three weeks ago quietly breaks today’s problems.
What readers usually want is simple: a method that works in the time they have, and a way to tell whether it is working. The research below answers both, and it points away from the habits most students rely on.
The Illusion of Knowing
Definition. An illusion of competence is the feeling of understanding that comes from seeing something, not from being able to produce it. Reading a worked solution feels clear because every step is supplied. The test then asks for the same steps from a blank page.
The largest review of study techniques, by Dunlosky and colleagues (2013) in Psychological Science in the Public Interest, rated ten common techniques. Practice testing and distributed practice earned the highest overall utility rating. Five techniques were rated low, including summarisation, highlighting and underlining, and rereading. In other words, the most popular habits sit near the bottom of the evidence.
Rule 1: Attempt Before You Look
Retrieval practice means pulling an answer out of memory instead of putting it back in. For maths, that means attempting a problem cold, even if the attempt fails, and only then checking the solution. The effort of retrieving is what strengthens the memory. This is known as the testing effect.
Here is the evidence in numbers. In a classic experiment by Roediger and Karpicke (2006), students read a passage and then either studied it four times or studied once and took three recall tests. Five minutes later, the group that studied four times recalled 83% of the ideas, against 71% for the testing group. One week later the order flipped: 40% for the rereaders and 61% for the testers.
Reading the Numbers Correctly
The gap in that experiment is a small lesson in percentages. At one week, 61% against 40% is a difference of 21 percentage points. Relative to the rereaders, it is a 52.5% improvement (21 ÷ 40). Stated the second way, the result sounds larger, and both statements are true. The guide on percentage change versus percentage points explains why the wording matters.
| Group (4 study rounds) | After 5 minutes | After 1 week |
|---|---|---|
| Study, study, study, study | 83% | 40% |
| Study, test, test, test | 71% | 61% |
| Change for the testing group | −12 points | +21 points |
The short-term loss explains why rereading feels better. It is better for the next few minutes. The advantage reverses once time passes, which is the situation every exam creates.
Common Techniques, Side by Side
| Technique | Evidence rating (Dunlosky 2013) | How it looks in maths |
|---|---|---|
| Practice testing | Highest utility | Attempt problems cold, then check |
| Distributed practice | Highest utility | Short sessions across several days |
| Rereading | Low utility | Reading the textbook chapter again |
| Highlighting and underlining | Low utility | Colouring formulas in notes |
| Summarisation | Low utility | Rewriting the chapter in shorter form |
The review rated other techniques too, but these five cover most of what students actually do.
Rule 2: Space the Practice
Distributed practice splits the same study time across several days. The memory fades a little between sessions, and recalling it again makes it stronger, a pattern linked to the spacing effect. Three 30-minute sessions across a week beat one 90-minute session on the night before, even though the hours are equal.
A simple plan. Practise a new topic on the day it is taught, again after one or two days, then after about a week, then shortly before the test. A student who has ten topics can keep a one-page list and tick each topic as it returns. The aim is not volume; it is that no topic goes untouched for more than a week or so.
Rule 3: Mix Problem Types
Most textbooks group problems by section: ten questions on one method, then ten on the next. Interleaving shuffles them. In a study by Rohrer and Taylor (2007), students who practised mixed problems did vastly better on a test one week later than students who practised in blocks. The reason is that mixed practice forces a decision the blocked version removes: which method does this problem need?
That decision is the real exam skill. A set of derivative problems, for example, may need the product, quotient or chain rule, and the first job is to choose. The guide to choosing a differentiation rule shows how, and the derivative calculator can confirm each result.
Rule 4: Study Worked Examples Actively
A worked example is a solved problem with every step shown. It is a strong way to learn a new method, provided the reading is active. Cover the solution, attempt the first step, then uncover and compare. Next, change one number and solve the variant. For word problems, the math word problem solver shows the set-up, which is often the hard part.
Keep a short error log: a page listing each mistake, its cause (concept, set-up, arithmetic or reading) and the fix. After two weeks, patterns appear. A student who sees “sign error” nine times knows exactly what to check, and the checks in how to check a math answer target those slips directly.
Anatomy of a 45-Minute Session
A focused session needs a shape, or it drifts into reading. One workable split: five minutes writing from memory what the last session covered, twenty minutes on new or shaky material with worked examples, fifteen minutes on a mixed set from different topics, and five minutes updating the error log. Two such sessions in a day cost an hour and a half and cover both new learning and review. A phone in another room helps more than any app.
Three Ways to Practise Retrieval in Maths
Blank-page derivation. Close the book and rebuild a result from nothing, such as the quadratic formula. The guide on the quadratic formula derived gives the five steps to check against afterwards.
Two-sided flashcards. Put a problem type on the front, for example “solve a system with no obvious substitution”, and the method on the back. This trains recognition, the skill interleaving builds.
Teach it back. Explain a solved problem aloud as if to someone who missed the lesson. Any step that cannot be explained marks a gap. Explaining takes about two minutes per problem and shows quickly what rereading would hide.
A Realistic Week Before a Test
Seven days out, list every topic and mark the ones that feel shaky. On days six to four, attempt mixed problems from past work, five or six at a time, and log each error. On day three, redo only the logged errors. On day two, take a timed mixed set with no notes. The day before, do a short review of formulas from memory and stop early. Sleep matters more than one extra hour of cramming, because memory consolidates overnight.
Calculators belong in this plan as checkers, not crutches. Solve first, then compare with the scientific calculator, or confirm fraction work in the fraction calculator.
How to Tell the Method Is Working
Progress in maths is easy to feel and hard to measure, so a simple number helps. After each practice set, record the share of problems solved correctly on the first cold attempt, with no notes and no hints. Suppose week one gives 4 out of 10 and week three gives 6 out of 10. That is a rise of 20 percentage points, or a 50% relative improvement, and either figure shows that the routine is paying off.
Three other signs are worth watching. The error log starts repeating fewer entries. Choosing a method before starting becomes quicker. And a changed version of a familiar problem no longer feels like a new problem. If none of these moves after two weeks, the likeliest cause is that the sessions have drifted back into reading, and the fix is to return to cold attempts.
If You Are Already Behind
Falling behind is fixable, but the repair has to start at the gap, not the current chapter. Find the earliest topic that feels shaky by attempting a few older problems cold. Rebuild that topic with worked examples and retrieval practice, then move forward one topic at a time. Fifteen focused minutes a day on the missing foundation usually helps more than a weekend of re-reading the current chapter.
A practical order for catching up is arithmetic and fractions first, then algebra basics such as solving linear equations, then the current topic. Each stage takes a few days of short sessions, and the solve for x calculator works well for checking a hand solution once the attempt is done. Asking a teacher one specific question about the exact step that fails is also far more productive than asking for a general re-explanation.
What Is Overrated, Underrated and Often Wrong
Overrated: highlighting, neat colour-coded notes and watching explanation videos at speed. They feel like progress and produce little recall. Underrated: the error log and the first attempt. Both are free, and both turn mistakes into information. Often wrong: the belief that studying should feel smooth. Effortful practice feels harder and works better. When a session feels easy, it is worth asking whether the page, not the student, is doing the thinking.
Quick Answers
Why can I understand math in class but fail the test?
Following a solution and producing one are different skills. Practise by attempting problems cold, then comparing with the solution.
How do I study math the night before a test?
Attempt mixed problems from earlier work, redo logged errors, sleep, and skip new topics. Retrieval beats rereading in the short window too.
How many problems should I do per day?
Quality outweighs quantity. Five to ten attempted problems with a full check usually teach more than thirty copied from solutions.
Is rereading notes ever useful?
Briefly, to refresh a topic before testing yourself. It is a poor main strategy.
Related Mathvelo Tools
Further Reading and Sources
- Dunlosky et al. (2013), Psychological Science in the Public Interest — Summary from the Association for Psychological Science
- Roediger and Karpicke (2006), Psychological Science, 17(3), 249–255 — Test-enhanced learning
- Rohrer and Taylor (2007) — The shuffling of mathematics problems improves learning
- Spaced repetition — Background on the spacing effect
A note on these guides. Figures from studies are quoted from the cited papers or summaries. The weekly plan is a suggestion, not a guarantee. Opinions are labelled as opinions. This page is educational and not professional advice.