Online vs in-person tutoring

What the evidence actually supports

The strongest evidence base for tutoring concerns one-to-one tuition in general rather than the medium through which it is delivered. The Education Endowment Foundation reports roughly five additional months of progress on average from one-to-one tuition, with the largest effects from short, frequent sessions over a defined period of up to ten weeks.

Those conditions are medium-neutral, and in one respect online delivery makes them easier to meet. A 30-minute session four times a week is logistically impossible for most families if it involves travel, and entirely feasible if it does not. Where online tutoring has a structural advantage, this is it — not the technology, but the fact that it removes the constraint preventing the evidence-based pattern.

Where in-person genuinely wins

Three situations favour physical presence, and they are more specific than the general preference people express.

Young children are the clearest case. Sustained attention to a screen is difficult below roughly ten years old, and a great deal of early teaching depends on watching a child's hand, not just their face.

The second is any subject with a physical component — instrumental music, laboratory practical technique, handwriting, art. Latency and a single camera angle make fine motor correction unreliable.

The third is students whose difficulty is primarily attentional or motivational rather than academic. A person in the room is harder to disengage from than a window that can be minimised, and for some students that difference is the entire intervention.

Where online genuinely wins

The decisive advantage is supply. In-person tutoring restricts you to people within travelling distance; online tutoring restricts you to people who speak your language and share enough of your timezone. For a common subject at a common level that difference is marginal. For IB Mathematics Analysis and Approaches at Higher Level, or preparation for a specific admissions test, it is the difference between a genuine specialist and whoever is nearby.

Online delivery also produces artefacts by default. A shared whiteboard leaves a record, sessions can be recorded with consent, and written work moves instantly in both directions. Parents can see what was covered without asking, and students can revisit an explanation rather than relying on memory of it.

Cost is modestly lower, since travel time leaves the tutor's cost base — but less than people expect, because the scarce input is expertise rather than the commute.

• Access to specialists regardless of location
• Short, frequent sessions become logistically possible
• Written records, shared whiteboards, optional recordings
• No travel time for either party

The conditions that decide it

Online tutoring fails for predictable, fixable reasons, and they are worth checking before concluding that the medium does not suit a student.

A student needs a stable connection, a device with a real keyboard and a screen large enough to hold a document and a video call simultaneously, and a quiet place that is not their bed. Mathematical and scientific subjects additionally need a way to write by hand — a graphics tablet, a second device pointed at paper, or a tutor who uses a shared editor well. Attempting IB Physics through a phone camera will fail, and the medium will get the blame.

The tutor's setup matters equally. Ask what they use to write during a lesson. A tutor who has taught online seriously will answer immediately and specifically, because they have solved this problem; one who has not will say the platform is fine.

Safeguarding differs by medium, not in importance

In-person tutoring raises questions about supervision, location and who else is present. Online tutoring raises questions about which platform is used, whether communication stays on it, and whether a parent can see the record.

The reasonable standard is the same in both cases: communication through channels a parent can see, an adult aware of when sessions happen, and no migration to private messaging with a child. Online makes that easier to enforce, since the record exists by default.