Spaced Practice: What Classroom Research Says About Spreading Learning Over Time

Spaced practice tends to beat massed practice in classroom studies. A 2025 meta-analysis in Behavioral Sciences pooled 22 classroom studies and found an average advantage of d = 0.54 for practicing in separate sessions, roughly the gap between a student at the 50th percentile and one at the 71st. Results varied widely from study to study.

None of the studies in that analysis tested retention beyond 42 days. That matters for anyone planning a unit that stretches across a semester, because a semester is a longer span than the evidence covers.

How the portfolio can support spacing

The Rapunzl curriculum scales from a 3-week unit up to a 28-week, year-long course, and the simulator uses live Nasdaq pricing with simulated $10,000 portfolios. A teacher can use a weekly portfolio check as a hook for bringing earlier concepts back: a holding checked in week two can be revisited in week six with new prices.

Spacing studies test scheduled return to specific content, and no study reviewed here tested a portfolio check as a spacing intervention. Treat the portfolio as a hook for planned review, one the teacher still has to schedule. A short unit can teach a concept; a longer course leaves more room to space the review.

What does the classroom evidence on spaced practice show?

Mawson and Kang (2025) screened over 3,000 articles for studies run in real classrooms with curriculum-relevant material, and kept those that compared spaced and massed practice with equal study time. The table puts their review next to the other studies this post relies on.

Mawson and Kang describe d = 0.54 as a moderate effect. Their studies differed a lot from one another (heterogeneity was above 90%), so the average hides a wide spread. Small studies in the analysis also tended to report bigger effects, which can inflate an average. Math was among the better-covered subjects: its estimate came out at d = 0.46, with a confidence interval from 0.09 to 0.84, and the authors read the math results as generally positive. Reading was the one subject that did not favor spacing (d = -0.11), though its interval spans zero. Differences between subjects were not statistically significant.

Cepeda and colleagues reviewed more than a century of research and concluded that separating learning episodes by at least a day is extremely useful for long-term retention. In the four studies in their review with a retention interval longer than a month, gaps of weeks or months did better than a one-day gap.

Does spacing work for school-age students?

Several small school-age studies point the same direction.

Carpenter's team had 8th graders review U.S. history facts after one week or after 16 weeks, by quiz with feedback or by restudying. A final test came 36 weeks after each group's review, about nine months later. Testing beat restudying, and the 16-week group scored higher than the 1-week group, though that gap did not reach statistical significance. This study had no massed condition, so it compares review timing and not spaced against massed. Sobel's fifth graders learned unfamiliar words in a massed session or spaced a week apart, and the spaced group recalled more five weeks later. Vlach and Sandhofer taught early elementary children science concepts on massed, clumped or spaced schedules, and spacing led to better generalization of concepts, which goes beyond rote recall. Both experiments are small.

In the Mawson and Kang review, primary school studies produced the most negative effects, and secondary school studies were more consistently positive. The authors could not separate age from the way schools are organized at each level.

How long should the gaps be?

No study reviewed here gives a formula for classroom gaps. Cepeda's analysis found the best gap between study sessions grows as the time you need to remember the material grows, so material needed in June benefits from longer gaps than material needed on Friday.

In the Mawson and Kang review, fixed gaps of seven days between returns gave the most consistently positive results, and expanding gaps tended to produce smaller effects, though that rests on a few studies that mostly shared one procedure. Studies with fewer returns also tended to show larger effects, so more returns are not automatically better. Most studies with three returns after the first lesson showed moderate to large effects. Larger effects also tended to come with longer retention intervals and older learners. These are descriptive patterns from a small set of studies, a starting point to try and adjust.

What does the research not show about spaced practice?

The evidence for spacing is encouraging in general and thinner for what a semester-long finance course needs.

One study found no benefit. Amabile and colleagues compared two cohorts of physical therapy students who took the same 45 hours of anatomy lecture and 90 hours of lab, one spread over 30 weeks and one compressed into 10. Eighteen months later, the cohorts did not differ significantly on the retention test. It was a prospective cohort study, not a randomized trial, so the groups may have differed in other ways. Still, spacing does not always show up in long-term results.

University STEM data are mixed. Bego's team found a significant spacing effect in two of nine introductory courses, and a significant overall effect only when calculus was included.

The classroom meta-analysis has its own limits. It found no studies with retention intervals longer than 42 days, so a 28-week course is supported in direction but not tested at that length. It defined spacing by timing alone, which let very different activities count as practice.

Cepeda's 2006 synthesis covers verbal recall tasks, and about 85% of its data came from young adults. The authors wrote that there was not enough data on children at retention intervals of a day or more to make strong claims.

Health-professions training shows a similar small benefit (a 2024 review of 23 studies found SMD 0.32 for knowledge in the nine that compared spaced and massed online education), but most of those studies were rated unclear or high risk of bias, so we leave them out of the argument.

Finally, the classroom meta-analysis counted only comparisons with equal study time in each condition. A one-week unit and a 28-week course differ in total hours as well as spacing, so a semester will teach more than a week for reasons that have nothing to do with the spacing effect. No study reviewed here tested personal finance content.

How do you build spaced practice into a finance unit?

1. Plan about three return dates for each core idea after you first teach it. Compound interest on day one could return in week two, week five and a semester review. More returns are not automatically better.
2. Start with a fixed gap of about a week. Mawson and Kang found fixed one-week gaps most consistently positive and expanding gaps smaller. If students need to remember for months, a longer gap is worth testing: Carpenter's 16-week delay did at least as well as a 1-week delay.
3. Use any recurring check-in as the hook. A weekly portfolio check is one option if you have one; ask students to explain one change using an idea from an earlier unit.
4. Put old questions in new places. Add two items on earlier topics to each week's homework or warm-up, so spacing arrives as a few questions with no extra lessons.
5. If you only have a short unit, split the practice across days. Even separating sessions by a day beats cramming everything into one sitting, according to the Cepeda review.
6. Pair spacing with quizzing. In Carpenter's study, review by quiz with feedback beat restudying.

For a concept worth returning to, teaching dollar-cost averaging can run in a single class period, and teaching students to read company financials builds toward a capstone thesis at the end of a unit.

Spreading a finance unit across a term is worth talking through with our team: request a Rapunzl demo and ask where the review could go.

By Maria Rodriguez, Curriculum Designer at Rapunzl.