By Callum Hay · 20 August 2026 · 9 min read

Families weighing up English tutoring are choosing between two structures more than two providers. One is a solo practice: a single practitioner who teaches every lesson himself. The other is a tutoring college: a business that delivers a program through a roster of tutors, usually in groups.

I run the first kind, so this article argues against an obvious interest. The only honest way to handle that is to put the evidence for the other model on the table in full. Where it favours colleges, I have said so.

The research does not say one-to-one is better

Start with the finding that most surprises parents.

The Education Endowment Foundation's Teaching and Learning Toolkit, the most widely used evidence summary in school education, rates one to one tuition at five months of additional progress over a year, from 123 studies, at moderate cost. It rates small group tuition, defined as one tutor with two to five students, at four months, from 62 studies, at low cost.

That is a one-month difference at a much lower price, and the EEF does not present even that month as secure. Its own summary says studies comparing the two show mixed results, that "tuition in groups of two or three has been equally or even more effective" in some cases, and that what matters may be "the particular type or quality of teaching enabled by very small groups" rather than the precise size of the group. The one firm threshold it offers sits higher up: once a group passes six or seven students, effectiveness drops noticeably.

The largest meta-analysis of randomised tutoring studies, Nickow, Oreopoulos and Quan (2020), points the same way. Across 96 randomised trials, the pooled effects by ratio are not even monotonic: 0.38 standard deviations for one-to-one, 0.29 for one-to-two, 0.36 for one-to-three or more. Smaller is not automatically better.

The same paper carries a caveat that applies to every number above, including the ones that flatter my model. It is overwhelmingly a study of younger children. Across grades 6 to 11 it found only eight studies, with a pooled effect of 0.16, the thinnest cell in the paper. Nobody, solo or college, can honestly tell you the headline tutoring effects are proven for VCE-age students. They are demonstrated in primary classrooms and extrapolated upward.

Where a college genuinely wins

Price, first. Group classes routinely run at around half the per-student cost of private lessons, and the EEF rates small group tuition as low cost against one-to-one's moderate. My own rate is $120 an hour. A family choosing a small group saves a substantial share of a year of tutoring, for a group size that sits inside the range the evidence supports, and the research above does not entitle me to tell them they are buying an inferior product.

Peer feedback, second, and this one matters specifically for English. A meta-analysis by Huisman and colleagues found that students who received peer feedback on their academic writing improved substantially more than students who wrote alone: an effect of 0.91, larger than any tutoring effect cited above. The same analysis found peer and teacher feedback produced similar improvements, though on that comparison the estimate is imprecise enough that it shows an absence of proven difference rather than proven equivalence. The studies are from higher education, not Year 12, so carry that caveat. But the mechanism is real: a class can put a student's draft in front of other students writing on the same text, and a one-to-one lesson structurally cannot. I approximate it with annotated exemplars on the readings pages. An approximation is what it is.

A cohort, third. Some students work harder in a room of peers doing the same subject than they ever will across a desk from an adult. I have no study for that; it is an observation from teaching, and colleges are built to exploit it.

Continuity of the business, fourth. When a college tutor is unwell, another tutor takes the class and the term continues. When I am unwell, the lesson moves. A solo practice is a single point of failure and there is no honest way to dress that up.

Where close attention wins

The strongest strand in the EEF toolkit is not a tuition model at all. It is feedback: six months of additional progress, at very low cost, on a larger and more secure evidence base than either tuition entry. The EEF adds that feedback aimed at the student as an individual, rather than at the work, tends to be the least effective kind.

That reframes the whole choice. What a family is buying, in either model, is the quality of feedback on actual work, not a seating arrangement. The question is which structure gets your child more of it.

A solo practice is built to concentrate it. The lesson is designed after the student's writing has been read, not before. When the problem turns out to be sentence construction rather than textual knowledge, the term changes shape that week, with no program running in parallel for thirty others to reconcile against.

And it accumulates. I read every piece a student writes across a year, so the twentieth lesson is informed by the nineteen before it: the way their arguments tend to collapse, the sentence pattern they overuse. Tutor continuity is listed as a design principle in Stanford's high-impact tutoring literature, but I should be straight about the evidence: no study has isolated and measured what continuity is worth. I believe it is the core of the model. I cannot prove it with a number, and neither can anyone selling you the opposite.

Neither model is regulated, so ask better questions

A 2025 study in the Australian Educational Researcher found "little to no concrete government policy regulating tutoring" in Australia, and noted that tutors are not required to follow any curriculum or hold qualifications. No government body licenses tutors or accredits tutoring businesses. "College" is a marketing word with no more legal standing than "solo practitioner".

The Australian Tutoring Association offers membership, a code of conduct and a complaints process to individual tutors and businesses on the same terms. So the real dividing line for accountability is member versus non-member, not solo versus college. Ask either kind of provider what happens when a parent complains.

On qualifications, the large Melbourne providers typically advertise their tutors' ATARs or study scores rather than teaching credentials. Whether that should worry you is genuinely unsettled. Nickow and colleagues found teacher-delivered tutoring strongest on average, but wrote that "it is far from clear that effectiveness differentials between trained teachers and paraprofessionals outweigh the cost differentials."

Results claims and testimonials, finally, are governed by Australian Consumer Law: fake or arranged reviews are illegal, and so is quietly removing negative ones. Read any provider's wall of praise, including mine, with that in mind.

Private tutor or group class: which suits which student

A student who needs structure and does not yet supply their own is often better served by a college. The timetable, the cohort and the set homework do work that a weekly hour cannot.

A student whose difficulty is specific and has persisted through ordinary teaching is usually better served one to one. If three years of classroom English have not resolved it, a fourth program covering the same ground at the same pace is unlikely to.

A student who is capable and unmotivated tends to do better in a room with peers than across a desk from an adult.

A student who is motivated and stuck is the clearest case for close attention, because what stands between them and better work is usually diagnostic rather than motivational. If you are not sure which of these describes your child, that question has its own page.

Three questions that separate providers within either model

Ask who teaches the lesson, and whether it is the same person each week. The answer varies more than the marketing suggests.

Ask what happens to a piece of writing after it is submitted. Written feedback a student can return to does different work from verbal comment in the room, and the feedback evidence above is the strongest evidence in this article.

Ask what the provider will tell you when it is not working. A provider with an answer has thought about the case where their own service is the wrong fit.

How Argo is arranged, and what it costs against the alternative

I teach every lesson myself, which is what makes the accumulation above possible and what limits the number of places each term. Lessons are $120 an hour, weekly. That is more than double the per-student cost of a good small group, and I have set out above what the difference does and does not buy. Who I am is on the about page; the free material, including the scholarship guide and the monthly sentence clinic, shows the teaching rather than describing it.

The discovery call and the trial lesson are free, and the trial begins with a reading of the student's writing rather than a description of a program. If the account of the college model above describes what your child needs, that is worth knowing before booking anything, and I would rather you found it out on a free call than after a term.

References

Australian Competition and Consumer Commission. (n.d.). Online reviews for products and services. Retrieved 20 August 2026, from https://www.accc.gov.au/business/advertising-and-promotions/online-reviews-for-product-and-services

Australian Tutoring Association. (n.d.). Membership. Retrieved 20 August 2026, from https://ata.edu.au/membership/

Education Endowment Foundation. (n.d.-a). One to one tuition. Teaching and Learning Toolkit. Retrieved 20 August 2026, from https://educationendowmentfoundation.org.uk/education-evidence/teaching-learning-toolkit/one-to-one-tuition

Education Endowment Foundation. (n.d.-b). Feedback. Teaching and Learning Toolkit. Retrieved 20 August 2026, from https://educationendowmentfoundation.org.uk/education-evidence/teaching-learning-toolkit/feedback

Education Endowment Foundation. (n.d.-c). Small group tuition. Teaching and Learning Toolkit. Retrieved 20 August 2026, from https://educationendowmentfoundation.org.uk/education-evidence/teaching-learning-toolkit/small-group-tuition

Huisman, B., Saab, N., van den Broek, P., & van Driel, J. (2019). The impact of formative peer feedback on higher education students' academic writing: A meta-analysis. Assessment & Evaluation in Higher Education, 44(6), 863–880. https://doi.org/10.1080/02602938.2018.1545896

Nickow, A., Oreopoulos, P., & Quan, V. (2020). The impressive effects of tutoring on preK-12 learning: A systematic review and meta-analysis of the experimental evidence (Working Paper No. 27476). National Bureau of Economic Research. https://doi.org/10.3386/w27476

Zunica, B., Reid-O'Connor, B., & Cartwright, K. (2025). Whose responsibility is it anyway? Exploring the tutoring policy landscape in Australia using the Peters framework. The Australian Educational Researcher, 52(6), 4511–4534. https://doi.org/10.1007/s13384-025-00908-3