A question parents ask

Is one-on-one tutoring better than a classroom?

The short answer

Yes, and by a wide margin. In 1984 Benjamin Bloom found that the average tutored student beat about 98 percent of students in a regular class, and later reviews found smaller gains that are still large.

The reasoning

Bloom compared students taught thirty to a room with students taught one on one by a tutor who made sure they understood each step before moving on. The average tutored student landed above about 98 percent of the classroom group. We've known this for forty years and mostly shrugged, because thirty desks are cheaper than a tutor.

Be careful with that number, though. Bloom's tutoring came with mastery learning, which means no one moved on until the last step was solid. Later work found the tutoring effect on its own is smaller. Kurt VanLehn's review of human tutoring put it at about 0.8 of a standard deviation. A 2020 review of tutoring programs for children from preschool through grade 12 found 0.37. Those figures are well under Bloom's two. They're still big.

So what makes it work? It isn't a magic person. It's the arrangement: one learner, someone paying close attention, a check for understanding before moving on, and feedback right away. A parent teaching one child already has all four. A school can't afford them.

In Sui Learning, a session has four moves: greet, review progress, teach one thing, then listen. You teach less than you want to, and you ask before you tell.

The research behind it

Students tutored one on one, with checks for understanding before moving on, did better than about 98 percent of students taught in a regular class. Benjamin Bloom, "The 2 Sigma Problem," Educational Researcher 13, no. 6 (1984).
Relationships, feedback, and review rank near the top of what moves learning. John Hattie, Visible Learning (2009), as cited in Sui Learning.
Human tutoring raised learning by about 0.8 of a standard deviation over no tutoring, in a review of studies from 1975 to 2010. Bloom's figure was 2.0. Kurt VanLehn, Educational Psychologist 46, no. 4 (2011).
Across experiments on tutoring for children from preschool through grade 12, the pooled gain was 0.37 of a standard deviation. Andre Nickow, Philip Oreopoulos, and Vincent Quan, NBER Working Paper 27476 (2020).

Try this

Tomorrow, teach one thing, then ask your child to explain it back before you move on. If they can't, teach it again a different way. That check for understanding is the part of Bloom's method any parent can start on right away.

Where this falls short

Tutoring takes time. A parent who is also working, or a child who learns well in a group, may need a mix. The size of the gain varies with the subject, the tutor and the program, so treat Bloom's number as the high end of what's possible and not a promise.

Also asked as

Is tutoring more effective than classroom teaching?
What is Bloom's two sigma problem?
Can a parent teach better than a school?

The ideas behind this answer

Two Sigma: you are the advantage
One on one beats the classroom by two standard deviations. You are the advantage. Act like it.
The four-move lesson
Every session is the same four moves: greet, review progress, teach one basic in context, then catch up and counsel.
Supply and demand: teach less
Teach less, so your words are worth more. Scarcity creates value; effort creates memory.