ეს სტატია ინგლისურ ენაზეა.
How many hours of tutoring a week does a child need? Sometimes none
If your child needs tutoring, the research points one way: short and often beats long and weekly. The Education Endowment Foundation, which reviews school evidence for England, finds that one to one tuition does best in "short, regular sessions (about 30 minutes, three to five times a week) over a set period of time (up to ten weeks)". A review of 96 randomised trials found that effects grew with the number of days a week, and saw "little evidence of once-weekly tutoring sessions generating large effect sizes". So the answer the studies support is about an hour and a half a week, in three pieces, for about ten weeks, and then a look at whether it worked.
Two things change that answer for a lot of families. The studies behind it measured tutoring that schools ran or organised, mostly during the school day, mostly for children who were already behind. None of them measured what most families buy, which is an hour a week after school with no end date. And for many children the right number of hours is zero: when the trouble is the order a new school teaches things in, when the child has been taught a topic and only needs practice, or when nobody at school has said there is a problem.
One disclosure before the detail. Cairn sells no tutoring, and this piece names no individual tutor. Our catalogue does list tutoring services: 71 one-to-one services taught in English and 102 marketplaces where a family picks a tutor or a class. On 16 November 2026, 11 of those one-to-one services and 93 of the marketplaces were marked as affiliate links, so a tutoring link elsewhere on this site may earn us a commission. No link in this piece is an affiliate link.
This is about a child who goes to school. It is not about home-schooling, and it is not about coaching for a university entrance exam.
How often should a child be tutored?
The Education Endowment Foundation's page on one to one tuition rests on 123 studies and was last reviewed in July 2021. It puts the average gain at about five months of extra progress. Behind that average, the gain was about six months in literacy and two in maths, and larger in primary schools than in secondary ones. The dose it found worked best is the one quoted above: about 30 minutes, three to five times a week, for up to ten weeks. Its page on small group tuition, for groups of two to five children and 62 studies, says much the same: "Frequent sessions, three times a week or so, lasting up to an hour over about 10 weeks typically show the greatest impact."
The review of 96 trials is by Andre Nickow, Philip Oreopoulos and Vincent Quan, published by the National Bureau of Economic Research in 2020. Overall, its effects grow with the number of tutoring days a week. It is not a straight line for every age, though. For children in grades 2 to 5 in the United States, which is roughly Year 3 to Year 6 in England and Class 2 to Class 5 under CBSE, three days a week did better than four or five. One example from the paper: a volunteer reading programme for 8 and 9 year olds in Northern Ireland found no significant effect at 30 minutes once a week, and a modest effect when the same programme ran twice a week.
In the United States the same idea goes by "high-dosage tutoring". A brief from the Annenberg Institute at Brown University, written in February 2021 by four researchers at Stanford, Brown and the University of Virginia, sums up the evidence as "three or more sessions per week with sessions that last for about 30-60 minutes per day". It adds that "Studies have found little evidence that once-a-week tutoring is sufficient to generate meaningful gains", that most programmes that worked lasted at least ten weeks, and that younger children may do better with "20 minutes, five times a week".
The best-known single study of that kind is a pair of randomised trials in Chicago by Jonathan Guryan and colleagues, published in the American Economic Review in 2023. Students in Grades 9 and 10, which is Year 10 and Year 11 in England and Class 9 and Class 10 under CBSE, had 50 minutes of maths tutoring every school day, two students to a tutor, for a school year, on top of their usual maths class. For the students who took part, maths test scores rose by 0.18 to 0.40 standard deviations. That is a little over four hours a week, as a class on the timetable.
What the studies measured, and on whom
This is the part most pages on the question leave out.
| Source | Who was tutored | Tutored by | How often, for how long | What was measured |
|---|---|---|---|---|
| Education Endowment Foundation, one to one (123 studies) | Children aged 3 to 18 in nurseries and schools; the page says children with low attainment benefit most | Teachers, teaching assistants or other adults | Best results: about 30 minutes, three to five times a week, up to ten weeks | Attainment tests: about five months of extra progress on average, two in maths |
| Education Endowment Foundation, small groups (62 studies) | The same, in groups of two to five, often children who are falling behind | Teachers, trained teaching assistants or tutors | Best results: about three times a week, up to an hour, about ten weeks | Attainment tests: about four months on average, three in maths |
| Nickow, Oreopoulos and Quan, 2020 (96 randomised trials) | Preschool to Grade 12, but almost half the studies involve first graders and 7% involve children in sixth grade or above | Teachers, teaching assistants, volunteers or parents | One to five days a week; most sessions 30 to 60 minutes | Test scores, within three months of the end of the programme |
| Guryan and colleagues, 2023 (two randomised trials, 5,343 students) | Grades 9 and 10 in Chicago public high schools; the authors describe them as economically disadvantaged | Paid tutors, mostly recent graduates, with about 100 hours of training | 50 minutes every school day, for a school year, as an extra class in school hours | Maths test scores and school grades |
Three things in that table matter for a family deciding what to buy.
The tutoring was run by schools. The Education Endowment Foundation includes only studies done in "a normal educational setting", such as a nursery or a school. In the review of 96 trials, about 80% of studies took place during school hours. The Chicago tutoring was a class on the timetable.
The children were mostly young, or behind, or both. A first grader is in Year 2 in England and Class 1 under CBSE. Few studies looked at older children: the Chicago authors describe theirs as the first large-scale tests of this kind with students of that age.
Nobody in these studies paid for it themselves. The review of 96 trials mentions privately paid tutoring only as a growing "shadow education system", and the Brown brief notes that "Wealthy families are also more likely to hire professional tutors". Neither measures what that tutoring does. We read these four sources and did not find a study of privately bought weekly tutoring to put beside them. That does not mean one hour a week after school does nothing. It means the evidence above was not about it, and the honest move is not to borrow its numbers for it.
Is 1 hour of tutoring a week enough?
On the evidence above, it is the dose with the least behind it. Both the review of 96 trials and the Brown brief say, in nearly the same words, that they found little evidence for once a week.
It is also the most common shape in our catalogue. We record a typical weekly time for each service, and for 43 of the 71 one-to-one services taught in English it is one hour a week. Of the 49 that charge by the session, the middle session is 60 minutes and the middle price is 30.69 US dollars. Those figures were read from our catalogue on 16 November 2026.
So the shape the market sells most and the shape the research found works are different shapes. You can ask for the second. Shorter sessions exist: 15 of those 49 services price a session of 25 to 50 minutes. Some arithmetic on the catalogue's middle price, for a tutor who charges the same by the hour: one hour a week for a 36-week school year is 1,104.84 US dollars. Three half-hour sessions a week for ten weeks is 15 hours, or 460.35 US dollars. Fifteen hours is also the block in which schools in England bought subsidised tuition through the National Tutoring Programme in 2020 to 2021, according to the Education Endowment Foundation.
When is the answer none?
Here are four cases where we think paying for tutoring is the wrong first move.
Nobody has named the gap. The Education Endowment Foundation finds tuition works best when it is "additional to and explicitly linked with normal lessons". If the worry is general, a tutor has nothing to link to. Ask the school which topic, and which skill inside it, before you pay anyone.
The gap is the order things are taught in. A child who moves between school systems often looks behind for a term because the new school teaches the same topics in a different order. We set out one case in CBSE to British curriculum at age 9: a nine year old moving from Class 4 under CBSE into Year 5 in England (Grade 4 in the United States) usually needs a short bridge in a few strands, not a weekly tutor in "maths".
The topic has been taught and only needs practice. Making times tables or fraction arithmetic automatic is practice a child does alone, and much of it is free. We compared two of the tools for it in IXL or Khan Academy.
The week is already full. The review of 96 trials expects more sessions to help "unless and until the tutoring sessions begin to crowd out other learning inputs", and asks readers to count what tutoring after school replaces: "homework time, recreation, or other developmentally beneficial activities". An hour of tutoring at the end of a long school day is not free time made useful. It is time taken from something else.
When tutoring earns its fee
The case for paying is strongest when the school has named the gap and it is more than one topic, when a short bridge at home moved nothing in six weeks, or when what your child lacks is a person who expects them at a set time. That last one is real, and no app supplies it.
If you do pay, the studies point to a shape you can ask any tutor for:
- A set period. Ten weeks, then stop and compare with where your child started. The Education Endowment Foundation says "up to ten weeks"; the Brown brief says most programmes that worked lasted at least ten. Either way, a date to stop and check, not a standing order.
- Short and often. Three sessions of 30 to 60 minutes a week, or for a younger child 20 minutes on most days, rather than one long session.
- Tied to the class. Ask the school what the class is doing this term and give it to the tutor. The school programmes had that link built in; a tutor a family books has to be given it.
- The same tutor. The Brown brief notes that many programmes keep one tutor with the same child for the whole programme.
The trade-off, said plainly: three short sessions a week ask more of a family's diary than one long one, and some tutors will not sell them. One hour a week, booked on purpose for ten weeks with a named goal, is a reasonable choice. One hour a week with no goal and no end date is the shape with the least behind it.
Cairn is software that writes a weekly learning plan for a child aged 4 to 18 from real classes, courses and apps, free ones included, and explains each choice. The plan fits the hours a week you give it. You can see how it works.