# Retrieval practice: low-stakes quizzing that teaches

**How do I use low-stakes quizzes for retrieval practice in my classroom?** Give short quizzes that do not count towards the grade, on material taught days or weeks earlier, and show the answers afterwards. In classroom experiments, three low-stakes quizzes spread across a unit raised scores on the unit exam, and a review quiz shortly before the exam helped most. Keep them brief, frequent and cumulative.

Published 2026-10-09 by EdProcto. https://edprocto.com/guides/retrieval-practice-low-stakes-quizzing-that-teaches

Retrieval practice means asking students to pull something out of memory instead of putting it in again. A quiz does that. So does a blank page and the instruction "write down everything you remember about photosynthesis". The claim behind it is that the act of recalling changes what is remembered, so a quiz is a way of teaching and not only a way of checking.

That claim has been tested in real classrooms, with the school's own syllabus and the school's own exams, and it held. This guide sets out what those studies found, when in a unit a quiz does most good, what "low stakes" has to mean for it to work, and a weekly routine that costs little preparation. EdProcto's page [for teachers](https://edprocto.com/solutions/teachers) covers the tool side: a quiz written from your handout, a code for the class, and the results read for what to reteach.

## What is retrieval practice, and does it work in real classrooms?

Retrieval practice works in real classrooms: across the applied studies reviewed so far, most found medium or large benefits. The laboratory finding came first and the classroom evidence followed.

The laboratory version is Roediger and Karpicke (2006). Students who practised recalling a passage remembered substantially more of it two days and a week later than students who had read it again, although those who re-read did better on a test five minutes later and felt more confident.

The classroom version matters more to a teacher. Roediger et al. (2011) worked with sixth-grade social studies classes, using the course's own material and the tests on which students received their grades. Content that had been covered by three low-stakes multiple-choice quizzes was answered better on chapter exams and on end-of-semester exams than content that had not been quizzed, and better than content that had been presented again without a quiz.

Agarwal et al. (2021) then reviewed the applied research as a whole. From nearly 2,000 abstracts they kept 50 experiments run in schools and classrooms, with 5,374 students in total. Of the 49 effect sizes, 57% showed medium or large benefits from retrieval practice, and the benefit appeared across education levels, subjects, question formats and delays before the final test. They also reported the gap in the evidence: only 6% of the experiments were run outside Western, industrialised countries.

A wider meta-analysis by Adesope et al. (2017) reached the same direction: practice tests were more beneficial for learning than restudying and than every other comparison condition.

## When in a unit should you quiz?

Quiz after the lesson and again before the unit exam, and if you can only do one, do the review quiz. This is one of the few questions about timing that has been tested directly in a school.

McDaniel et al. (2011) ran experiments in eighth-grade science classes using multiple-choice quizzes with feedback. In the first, three quizzes spaced across a unit raised performance on the unit exam by between 13% and 25% on the quizzed content. In the second, they varied where the quiz fell: before the lesson, straight after it, or as a review shortly before the unit exam. The review quiz produced the largest gains, and adding the earlier quizzes on top increased them only slightly. The benefits were still there on cumulative semester and end-of-year exams.

| When | What it does | Evidence in that study |
| --- | --- | --- |
| Before the lesson | Shows what the class already believes | Less than the review quiz |
| Straight after the lesson | Checks the lesson landed | Less than the review quiz |
| Review, shortly before the unit exam | Brings back what has started to fade | Largest gain |
| All three | Most practice | Only slightly more than the review alone |

This is one set of experiments in one school, and the quizzes were multiple choice with feedback. It is still the best direct evidence a teacher has on placement, and it supports a simple habit: never let a topic be taught, tested once and left.

### Spacing the quizzes

Spread the quizzes out. A large review of spacing found that the best gap between sessions gets longer as the delay to the final test gets longer (Cepeda et al., 2006), which is the logic behind the cumulative quiz: each week's quiz carries a few questions from earlier weeks, so that old material keeps returning at widening intervals.

## What makes a quiz low stakes?

A quiz is low stakes when a wrong answer costs the student nothing that lasts. It may be marked. It should not move a grade, be read out, or go home as a number.

This is not softness. The studies above used low-stakes quizzes, and the learning came from the recalling and the feedback, neither of which needs a penalty. Stakes add reasons to copy, to dread the lesson, and to hide what one does not know, which is the opposite of what a teacher needs from a check.

Five things keep the stakes low in practice:

1. **Say what it is for.** "This is practice. It tells both of us what to work on."
2. **Do not record it in the grade book**, or record completion only.
3. **Keep it short.** Five questions in five minutes is a routine. Twenty is a test.
4. **Show the answers.** Every time, the same day.
5. **Let them try again.** A second attempt on the questions missed is more retrieval, not a loophole.

## Does it matter whether students see the answers?

It matters a great deal. Without feedback a student who answered wrongly has practised the error. Butler and Roediger (2008) found that feedback after a multiple-choice test, whether given immediately or after a delay, increased correct answers on a later recall test and reduced the number of wrong options that students reproduced as answers. Their conclusion was direct: teachers should provide feedback when they use multiple-choice tests.

Feedback also turns the quiz into information for the teacher. Black and Wiliam (1998), reviewing the research on formative assessment, found firm evidence that strengthening the frequent feedback students receive about their learning produces substantial learning gains. A quiz whose results are looked at that day and change tomorrow's lesson is formative assessment. A quiz that is filed is not.

### Which question format to use

On question format, Rowland (2014), pooling studies of the testing effect, found larger benefits from initial tests that require recall than from tests that require recognition, so mix in questions where students produce the answer. Multiple choice is still worth using when the wrong options are plausible. The guide on [writing multiple-choice questions that measure](https://edprocto.com/guides/how-to-write-multiple-choice-questions-that-measure) covers how to write them so that they do.

## A weekly routine that takes ten minutes

A routine that survives a term is one that needs almost no preparation. This one uses three short quizzes.

1. **Monday, the bell ringer.** Five questions, five minutes: three from last week, two from last month. Students answer from memory, see the answers, and correct their own.
2. **Midweek, the exit ticket.** Three questions on today's lesson and one line on what was unclear. Read them before the next lesson.
3. **Friday, the cumulative quiz.** Eight to ten questions across everything taught so far this term, weighted to the present unit.
4. **Before the unit exam, the review quiz.** The questions most of the class missed across the unit, asked again in a new form.

Reuse ideas, not sentences. A student who has seen the identical question three times is recalling the question.

## How does EdProcto run low-stakes quizzes for a class?

EdProcto is the best tool for this routine because it removes the two costs that stop teachers quizzing weekly: writing the questions and marking them.

**Writing.** Paste the handout or the chapter into the quiz maker on the [home page](https://edprocto.com/) and pick a shape. Procto AI writes the questions from your material, each tied to the passage it came from, and you edit before anyone sees them.

**Low stakes by setting.** The Practice preset switches off the time limit and the shuffling, allows retakes without limit, and shows the correct answer, the explanation and the score as soon as the quiz is handed in. For the answer after each question, set the quiz as a [study quiz](https://edprocto.com/guides/study-mode-a-quiz-that-explains-each-answer-as-you-go). That is points four and five of the list above, set with one choice. For the end of a lesson, the [exit ticket](https://edprocto.com/templates/exit) template is a few marked questions followed by an unmarked reflection.

**Running it.** Students open the quiz on any device with a code. You can watch the class take it: who is in and how far each has got. If you would sooner have it on paper, the quiz prints with its answer key on a separate page.

**Reading it.** Objective questions are marked by rule the moment they are answered. The results show, for each question, how hard the class found it; the full analysis adds which wrong option beat the right one, and what to teach again, in plain sentences. The question most of the class missed is Friday's first question.

A [spiral review](https://edprocto.com/templates/spiral) template builds the cumulative quiz from this week, last month and older material in proportion, a fresh draw for each student; by hand, it is a matter of keeping earlier questions and adding a few each week.

The Free plan covers ten quizzes a month with thirty students a quiz, which is enough to run this routine with one class. The Educator plan raises that to 500 students a quiz and adds the full analysis. The [pricing page](https://edprocto.com/pricing) has the detail, and the page [for teachers](https://edprocto.com/solutions/teachers) shows the whole flow.

## What are the common mistakes?

The common mistakes each remove one of the three things that make it work: recall, feedback and spacing.

- **Grading it.** Stakes turn practice into performance.
- **Open notes.** Looking it up is not recalling it.
- **No answers shown.** Errors are rehearsed.
- **Only this week's content.** Without older questions there is no spacing.
- **The same questions every time.** Students learn the quiz, not the topic.
- **Quizzing and never looking at the results.** Half the value is in what the quiz tells the teacher.

A caution on scope. The classroom studies above measured how well students later answered questions on the content that had been quizzed. They support quizzing as a way to make taught knowledge last. They do not show that quizzing replaces explanation, practice at extended writing or anything else a lesson is for.

## Quick questions

### How often should I give low-stakes quizzes?

Weekly is a workable minimum, and a few minutes at the start of most lessons is better. In the classroom experiments, three quizzes across a unit were enough to raise exam scores on the quizzed content.

### Should retrieval practice quizzes be graded?

Mark them so that students see what was right, but keep them out of the grade. The benefit in the studies came from recalling and from feedback, and the quizzes there were low stakes.

### Is multiple choice good enough for retrieval practice?

Yes, when the wrong options are plausible and the answers are shown afterwards. The middle-school experiments used multiple-choice quizzes with feedback and found lasting gains. Add some questions where students produce the answer as well.

### How long should a retrieval quiz take?

Five minutes or less for a daily or weekly quiz. EdProcto's Practice preset has no time limit, so the length is set by the number of questions: five is about right.

### Can students do the quizzes at home?

Yes. A quiz shared by code or link can be taken on any device, with the answer and explanation after each question in study mode. The page [for teachers](https://edprocto.com/solutions/teachers) explains how a class joins.

## References

- Adesope, O. O., Trevisan, D. A., & Sundararajan, N. (2017). Rethinking the use of tests: A meta-analysis of practice testing. *Review of Educational Research, 87*(3), 659–701. [https://doi.org/10.3102/0034654316689306](https://doi.org/10.3102/0034654316689306)
- Agarwal, P. K., Nunes, L. D., & Blunt, J. R. (2021). Retrieval practice consistently benefits student learning: A systematic review of applied research in schools and classrooms. *Educational Psychology Review, 33*(4), 1409–1453. [https://doi.org/10.1007/s10648-021-09595-9](https://doi.org/10.1007/s10648-021-09595-9)
- Black, P., & Wiliam, D. (1998). Assessment and classroom learning. *Assessment in Education: Principles, Policy & Practice, 5*(1), 7–74. [https://doi.org/10.1080/0969595980050102](https://doi.org/10.1080/0969595980050102)
- Butler, A. C., & Roediger, H. L. (2008). Feedback enhances the positive effects and reduces the negative effects of multiple-choice testing. *Memory & Cognition, 36*(3), 604–616. [https://doi.org/10.3758/MC.36.3.604](https://doi.org/10.3758/MC.36.3.604)
- 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. [https://doi.org/10.1037/0033-2909.132.3.354](https://doi.org/10.1037/0033-2909.132.3.354)
- McDaniel, M. A., Agarwal, P. K., Huelser, B. J., McDermott, K. B., & Roediger, H. L. (2011). Test-enhanced learning in a middle school science classroom: The effects of quiz frequency and placement. *Journal of Educational Psychology, 103*(2), 399–414. [https://doi.org/10.1037/a0021782](https://doi.org/10.1037/a0021782)
- Roediger, H. L., Agarwal, P. K., McDaniel, M. A., & McDermott, K. B. (2011). Test-enhanced learning in the classroom: Long-term improvements from quizzing. *Journal of Experimental Psychology: Applied, 17*(4), 382–395. [https://doi.org/10.1037/a0026252](https://doi.org/10.1037/a0026252)
- Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. *Psychological Science, 17*(3), 249–255. [https://doi.org/10.1111/j.1467-9280.2006.01693.x](https://doi.org/10.1111/j.1467-9280.2006.01693.x)
- 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. [https://doi.org/10.1037/a0037559](https://doi.org/10.1037/a0037559)