In 1984, the educational psychologist Benjamin Bloom published a finding that has shaped thinking about teaching ever since. Students taught one-to-one performed about two standard deviations better than students taught in a conventional classroom — enough to move a median student to roughly the 98th percentile of the comparison group.
Bloom called it the “two sigma problem”, and the word problem was deliberate. He was not claiming that every student should have a private tutor; he was asking how group teaching might be improved to approach a result that one-to-one teaching achieved reliably.
Why one-to-one produces the difference
The mechanism is less mysterious than the headline figure suggests. A tutor working with one student can do several things a teacher with thirty cannot, and none of them depend on the tutor being exceptional.
- Diagnose the specific misunderstanding rather than the general topic a class is finding hard.
- Adjust pace continuously, slowing where a concept has not landed and moving quickly where it has.
- Require the student to explain their reasoning, which surfaces gaps that a correct answer can hide.
- Correct a method error at the moment it happens, before it is practised into a habit.
- Choose what to work on next based on this student's gaps rather than a scheme of work.
The third point deserves emphasis. A student can reach a right answer by a wrong method, and in a class of thirty that goes unnoticed until it fails on harder material. One-to-one, it surfaces in minutes.
Reading the number honestly
Two standard deviations is a striking figure, and it deserves context rather than repetition.
Bloom's studies were conducted under particular conditions, with tutors applying a specific mastery-learning approach over defined units of work. Later researchers have generally found smaller effects in ordinary settings, and the exact size remains debated. What has held up consistently is the direction and the mechanism: individualised instruction outperforms group instruction, and it does so because of the feedback loop rather than because of anything mystical about attention.
The honest summary: one-to-one teaching reliably outperforms group teaching, and the size of the advantage depends heavily on what the tutor actually does with the time.
What this means when choosing
If the value of tutoring lies in individualisation, then the useful question about any arrangement is how much individualisation it actually delivers.
A session where a student watches a tutor work through prepared material captures little of the advantage — that is a lecture with an audience of one. A session built around the student's own errors, where they do most of the talking and the tutor responds to what they produce, captures most of it.
Where group classes still make sense
Group teaching is not without merit. Cost, peer discussion, and the discipline of a regular timetable all count for something, and a confident student working through standard material may do perfectly well in a class.
The case for individual attention is strongest where a student has specific, identifiable gaps — the situation where a class necessarily moves at a pace set by someone else's understanding.
Common questions
What is Bloom's two-sigma problem?
A 1984 finding that students taught one-to-one performed roughly two standard deviations better than students in conventional classrooms. Bloom framed it as a problem because the challenge was to find group methods that approached the same result, not to argue that everyone needs a private tutor.
Is the two-sigma effect still accepted?
The direction is well supported; the exact magnitude is debated. Later studies in ordinary settings generally find smaller effects than Bloom's original figure. The consistent finding is that individualised instruction outperforms group instruction, largely because of faster, more specific feedback.
How many sessions before we see improvement?
It varies by subject and by how specific the gap is. Where a student has an identifiable misunderstanding, progress can be quick once it is diagnosed. Broader gaps built up over years take longer, and any tutor promising a fixed timeline regardless of the student is overselling.
Are group classes ever better than one-to-one?
They can suit a confident student working through standard material, and they cost less. The advantage of one-to-one is largest where a student has specific gaps, because a class necessarily moves at a pace set by the group rather than by them.
Ready to See the SmavyDifference? Free Trial Per Subject.
One free 1-hour trial session per subject. Not happy with the tutor? We find another — free — until you're satisfied.