In the first week back, a nine-year-old sits a two-minute multiplication check. In June she answered 7 × 8 without pausing. Now she counts up in eights on her fingers, gets there, and runs out of time before the last row. Her reading is exactly where it was in June — same books, same pace, no wobble. That pattern, fast recall gone soft while comprehension holds steady, is the most common shape of summer slide, and it points to a fix far smaller than most families expect.
What the evidence supports, and what gets stretched
The popular version — children lose a fixed chunk of learning every summer, and it compounds year after year — is stronger than the research behind it. The narrower version is more useful.
Studies that test children in June and again in September do find a small average dip, and they find it unevenly. Computation and other fact-based, procedural skills drop more than reading comprehension. But the average conceals a huge spread: many children hold their level, some come back ahead, and the measured size of the drop depends heavily on which test was used and how its scores were scaled. When researchers rerun the analysis with newer assessments and different growth models, the effect tends to shrink and become less consistent. The related claim, that the summer is the main engine of long-run achievement gaps, is disputed for the same reasons.
Two conclusions survive all of that. First, the losses that do appear land on a predictable set of skills rather than everything at once. Second, they are recoverable, usually within the first weeks of term and without heroic effort. That makes the long break a maintenance problem, not an emergency.
How long your break actually is matters more than any of this. Six weeks off, as in much of the UK, is a different situation from ten or eleven weeks in many US districts. In Australia and New Zealand the long break falls across December and January. In a number of countries children come home with holiday workbooks and the question of adding anything barely arises. Count your weeks before deciding how much to do.
Why arithmetic slips and reading usually doesn't
Reading keeps happening whether or not anyone plans it: subtitles, street signs, cereal boxes, game menus, the comic a sibling left on the sofa, road signs on a long drive. Vocabulary and background knowledge feed on all of it. Nobody encounters 7 × 8 by accident on holiday.
That is the whole mechanism. Skills that depend on fast retrieval, or on rehearsed step-by-step procedures, fade without rehearsal. Skills built from broad exposure keep being fed.
| Skill | What it rests on | Risk over a long break | Cheapest maintenance |
|---|---|---|---|
| Multiplication and division facts | Fast recall from memory | High | 5 minutes of shuffled cards, 3 times a week |
| Written procedures (column subtraction, long division, fractions) | Rehearsed sequences of steps | High | Three problems, twice a week |
| Spelling and handwriting fluency | Motor habit plus memory | Medium | A short dictation or a postcard, weekly |
| Decoding, ages 5–7 | Still a procedure, not yet automatic | Medium to high | 10 minutes reading aloud, most days |
| Reading comprehension, age 8+ | Vocabulary and background knowledge | Low | Anything readable, including audiobooks |
| A second or home language | Daily use | High if unused | Conversation, not worksheets |
The exception worth watching is the five- to seven-year-old whose decoding is still new. At that stage sounding out behaves like arithmetic rather than like comprehension — it is a procedure being automated, and it needs turns. Ten minutes of reading aloud to a listening adult, most days, is enough.
How little practice actually holds ground
Maintenance is cheap. Ten to fifteen minutes, three or four days a week, will keep number facts and written procedures roughly where they were in June. Not forty-five minutes a day, and not a workbook for next year's syllabus.
What makes those minutes work:
- Ask, don't re-teach. Pulling an answer out of memory is what strengthens it. Pose the question, wait, and explain only if the child is genuinely stuck rather than merely slow.
- Spread it out. Four ten-minute goes across a week beat one forty-minute session on Sunday, comfortably, for the same total time.
- Watch the clock, not just the tick. A fact recalled in under about three seconds is remembered. One that takes six or seven is being rebuilt by counting. Both look correct on a page, and only one of them is safe.
- Shuffle. Practising the six times table in order lets a child ride the pattern rather than recall anything. Mixed cards from three tables at once force the real thing.
A workable week: Monday, five minutes of shuffled multiplication cards. Wednesday, three column-subtraction or long-division problems on the back of an envelope. Friday, cards again plus two fractions to put in order. Reading happens whenever it happens, unmanaged. Total commitment, about half an hour.
A routine that survives travel and childcare weeks
Routines fail on the days they were never designed for: the flight, the fortnight at grandparents', the three weeks of holiday club with a 6am start. Design for those days first.
Anchor to an event, not a clock. "After breakfast, before screens" crosses time zones and works in someone else's kitchen. "Ten o'clock" collapses on day one.
Write down the shrunken version. Have three sizes and name them out loud. A full day is fifteen minutes at a table. A small day is ten minutes of cards in the back of the car. A minimum day is one question at dinner — "what's seven eights?" — and it counts. Children accept a smaller version far more readily than a skipped day, and continuity is what keeps the habit alive.
Pack a kit, not an app. A deck of cards covers most arithmetic maintenance: take the picture cards out, each player turns two, multiply, higher product wins. Add one paperback and a small notebook. That works on a train with no signal and a flat battery.
Hand over one thing, not a schedule. If a grandparent or a childminder has the child for two weeks, give them an index card with three lines on it. Most people will do one specific thing willingly and nothing vague at all.
Book the amnesty in advance. Mark two or three weeks off on the calendar before the break starts — the actual holiday, the week the cousins visit. A break that was planned does not turn into guilt, and guilt is what turns one missed week into a missed month.
The fortnight before term starts
The last two weeks do more work than the middle six.
- Move bedtime earlier by about fifteen minutes every second or third day, rather than resetting it in one jolt the night before.
- Revisit last year's final topics, not next year's opening ones. Reactivating something half-remembered is quick and it feels good. Pre-teaching new material mostly makes the first lessons boring.
- Rebuild writing stamina: five to ten minutes of actual handwriting a few times — a postcard, a shopping list, a short account of the trip. Hands get out of practice, and a full page on day one is a shock.
- For early readers, daily reading aloud with an adult listening for the final ten days or so.
When maintenance isn't the right goal
Some situations justify more. A child who finished the year still struggling to decode. A child moving countries, or into a new language of instruction. A child who missed a long stretch of the year through illness. In those cases twenty to thirty minutes, four or five days a week, aimed at exactly one skill, is reasonable — and it should be one skill, chosen with the teacher's input where that is possible, not a general programme. Structured practice, whether that is a phonics scheme, a tutor, or an app such as Kid Genius World, earns its place here the way a physiotherapy exercise does: the same movement, often, for a defined period, with someone checking. None of it substitutes for the classroom.
Everyone else should aim lower and actually hit it. Half an hour a week, spread across the week, in a house where reading happens, is enough to walk back in with the times tables intact.