Secondary 2 English Tuition | One-to-One vs Small-Group Tutoring

Secondary 2 English tuition can work well one-to-one or in a genuinely small group. The useful question is not which format wins in the abstract. It is which format gives this learner the right amount of diagnosis, explanation, practice, feedback, discussion, independence and transfer for the English problem that actually exists.

A one-to-one English tutor can observe every decision, change the task immediately and give the learner a very high share of instructional attention. A small-group English tutor can create comparison, discussion, peer explanation, multiple interpretations and social rehearsal that one learner alone cannot generate. Those are different instructional affordances. Either can be squandered by weak teaching, and either can be powerful when the lesson design uses the format deliberately.

This Secondary 2 guide therefore does not compete with SETC’s site-wide One-to-One vs Small-Group English Tuition comparison. That broader page owns the general format question. This page owns the Secondary 2 application: the consolidation year in which reading, writing, language, listening, oral response, timing and feedback transfer must become more stable before the learner moves into a more demanding Secondary 3 stage.

The short answer

Choose one-to-one more readily when the problem is narrow, highly individual, difficult to diagnose, emotionally sensitive, strongly dependent on tutor prompting, or in urgent need of intensive repair. Choose a small group more readily when the learner can already participate independently enough to benefit from peer comparison, discussion, multiple solutions, audience awareness and shared practice.

1. Start with the learner’s problem, not the product format.
2. Ask what kind of interaction the target skill requires.
3. Check whether the learner needs intensive diagnostic attention or useful peer variation.
4. Distinguish a real small group from a small class. Group size changes what the tutor can observe and respond to.
5. Do not assume one-to-one is automatically superior because it is more personalised.
6. Do not assume a group is automatically better because it is social.
7. Track student talk, decision ownership, feedback latency, prompt dependence and fresh-task transfer.
8. Change format if the learner’s needs change.
9. Judge the format by independent school transfer, not by how busy the lesson looks.

1. The evidence does not support a universal winner

The Education Endowment Foundation reports positive average effects for both one-to-one and small-group tuition. Its one-to-one review estimates a somewhat larger average impact, while its small-group review highlights lower delivery cost and strong average effects when support is targeted. Importantly, EEF also notes that direct comparisons are mixed: some studies favour one-to-one, while others find pairs or groups of two or three equally or more effective. That should immediately make families cautious about absolute claims.

A 2026 review of tutoring research from the National Student Support Accelerator similarly describes a trade-off. Lower student–tutor ratios often produce larger academic effects and more individualised interaction, but small groups of roughly two to four students can still generate meaningful learning. Recent experimental work also suggests that smaller groups can outperform slightly larger groups under some conditions, yet this evidence comes mainly from mathematics and school-based programmes. It should inform English tutoring decisions without being treated as a direct verdict on every English lesson.

The evidence therefore supports a design principle rather than a slogan: the format matters because it changes the teaching interactions that are possible. Secondary 2 English parents and tutors should ask which interactions the learner needs most.

2. The format is not the intervention

“One-to-one” tells you the student–tutor ratio. It does not tell you whether the tutor diagnoses accurately, explains clearly, chooses good texts, uses feedback well, fades prompts, protects authorship or tests transfer. “Small group” tells you several learners share the tutor. It does not tell you whether everyone participates, whether tasks are differentiated, whether one student dominates, or whether peer discussion actually improves English reasoning.

Two one-to-one lessons can be radically different. In one, the tutor speaks for most of the hour, corrects every line and guides the learner through familiar questions. In another, the learner attempts independently, explains reasoning, receives one precise intervention, then completes fresh work with less support. The ratio is identical; the learning design is not.

The same is true for groups. A three-student tutorial can function like a small lecture, with each learner waiting for the tutor to talk. Or it can use structured comparison: one learner explains an inference, another challenges the evidence, a third proposes a narrower claim, and all three revise independently. Group size creates opportunities. Teaching quality determines whether those opportunities become learning.

3. Secondary 2 changes the comparison

At Secondary 2, the learner is no longer simply adjusting to secondary-school English. The central problem becomes stability. Can the student use a reading strategy across unfamiliar genres? Can writing structure adapt to purpose? Can grammar survive longer original writing? Can feedback transfer after a delay? Can oral ideas withstand follow-up questions? Can the learner manage time without the tutor acting as an external control system?

These goals make format choice more interesting than “more attention versus less attention.” One-to-one can expose hidden dependence quickly because the tutor sees every hesitation and can vary support precisely. A small group can expose whether a learner’s explanation is robust enough to survive comparison, challenge and alternative interpretations. Secondary 2 often needs both kinds of evidence.

4. Ask what the learner must do more of

If the student needs more diagnostic conversation, more immediate correction of a narrow misconception, more supported rehearsal of a fragile routine or more carefully controlled pacing, one-to-one may create useful conditions. If the student needs more discussion, more exposure to different reasoning, more audience awareness, more oral challenge or more practice evaluating other people’s choices, a small group may create useful conditions.

This question is stronger than asking what the tutor can provide. The learner’s activity should drive the choice. A format is valuable when it causes the student to perform the cognitive work that the target skill requires.

5. One-to-one gives the tutor more observational bandwidth

In a one-to-one lesson, the tutor can watch the entire problem-solving path. They can see where the student pauses, what they reread, how they choose evidence, which sentence they abandon, when they change an answer and what kind of prompt restarts progress. This can be especially valuable during diagnosis because subtle failure points are easy to miss in a group.

For example, two students may both give an overbroad inference. One misread a qualifier. The other chose the correct evidence but used language that was too certain. Their final errors look similar; the repair differs. One-to-one questioning can trace the path quickly.

The risk is that observational bandwidth turns into intervention overload. Because the tutor can notice everything, they may correct everything. That can make the lesson efficient in the short term and dependency-producing in the long term. Good one-to-one tutoring must deliberately leave some errors untouched while the learner completes the current reasoning job.

6. Small groups create comparison bandwidth

A small group offers something one-to-one cannot manufacture perfectly: simultaneous access to other learners’ choices. One student may select different evidence. Another may interpret a tone shift differently. A third may structure the same writing prompt around a different paragraph sequence. Those contrasts can sharpen judgement when the tutor makes the comparison explicit.

English contains many decisions where multiple plausible responses exist. The learner benefits from seeing that “different” is not the same as “wrong,” and that alternatives can be compared by relevance, evidence, clarity, purpose and audience. A small group can make these criteria visible.

The risk is passive observation. A learner can sit in a strong group and mentally outsource the reasoning to faster peers. Good group design therefore requires individual first attempts, named participation roles, personal written decisions and fresh independent transfer after discussion.

7. Attention share matters, but attention quality matters more

One-to-one tutoring offers the highest possible share of tutor attention. That matters when the learner’s process is difficult to read or when errors need immediate, precise diagnosis. But a high attention share can become counterproductive if the learner is never allowed to struggle productively, wait, self-correct or evaluate someone else’s reasoning.

In a well-run small group, the tutor’s direct attention is divided, yet the student may spend more total time thinking because peers create useful tasks while the tutor works with someone else. A learner can annotate independently, compare a partner’s paragraph, prepare an oral response or identify a disagreement in evidence. The important measure is not “minutes of tutor gaze.” It is the quality of the learner’s cognitive activity across the lesson.

8. Feedback latency changes with format

One-to-one allows feedback to arrive immediately. This can be powerful when a misconception should not be rehearsed repeatedly. It can also be dangerous if the student becomes unable to tolerate uncertainty. If every hesitation triggers tutor help within seconds, the learner never develops an internal error-detection threshold.

Small groups naturally create some delay. While the tutor listens to another student, the learner may have to continue thinking, reread, compare or mark uncertainty. That delay can support independence when tasks are well designed. But delay becomes harmful if the group is too large, the learner remains stuck for long periods, or feedback arrives after the reasoning has been forgotten.

For Secondary 2, the aim is controlled latency: enough time for independent thought, not so much that important misconceptions harden.

9. The first-attempt rule protects both formats

Whether tuition is one-to-one or small-group, many English tasks should begin with an individual attempt. The student reads before hearing the tutor’s interpretation. The student plans before seeing another learner’s plan. The student answers before a model is shown. The student edits before the error location is revealed.

This rule preserves diagnostic evidence. In one-to-one, it stops the tutor from accidentally leading the learner. In a group, it stops fast students from becoming the unofficial answer key. Once each learner has committed to an initial position, comparison becomes educational rather than contagious.

10. One-to-one is often strongest for narrow repair

Suppose a student has a recurring but specific problem: pronoun reference collapses in long sentences; inference answers overstate motive; a writing plan stalls because paragraph jobs are not distinguished; checking consumes too much time because every answer is reopened. These problems benefit from rapid cycles of observation, explanation, practice and adjustment.

One-to-one can keep the lesson tightly centred on the mechanism without asking other learners to spend time on a problem they do not share. The tutor can vary difficulty moment by moment and stop as soon as the repair becomes stable enough to test on fresh work.

The key phrase is narrow repair. If one-to-one is used simply to cover the whole Secondary 2 curriculum privately, the format’s diagnostic advantage may be wasted.

11. Small groups are often strongest for comparative judgement

Some English capabilities grow through seeing alternatives. Which of three pieces of evidence is strongest? Which paragraph opening creates the clearest reader expectation? Which oral answer responds most directly to the question? Which summary sentence preserves the qualifier? Which vocabulary choice fits the register?

A small group can provide a live bank of alternatives generated by real learners. The tutor can ask students to rank, justify, challenge and revise. This is especially useful at Secondary 2 because the learner is moving from rule-following toward judgement.

12. One-to-one can accelerate diagnosis

During the first sessions, one-to-one can be efficient when the tutor does not yet know what is wrong. The tutor can move through a reading sample, short writing task, editing sequence and oral explanation while asking follow-up questions tailored to every response. A working hypothesis may emerge quickly.

This is particularly useful when school marks hide the mechanism. A student may have middling results across several components, but one-to-one observation reveals that slow task interpretation is creating problems everywhere. The tutor can then test that hypothesis directly rather than starting five separate remedial programmes.

13. Small groups can improve diagnostic contrast

Diagnosis is not exclusively a one-to-one advantage. In a group, the tutor can compare how several learners interpret the same prompt. If one student finds the relevant evidence immediately while another searches widely, the contrast may reveal a question-reading problem. If two students build different but defensible interpretations, the discussion can reveal whether a third student treats one plausible answer as the only possible answer.

Group diagnosis is strongest when the tutor records individual patterns rather than assuming that everyone who made the same mistake needs the same explanation.

14. One-to-one can be better when the learner is far behind

A learner who cannot access current Secondary 2 work may need a temporary intensity that is difficult to provide in a group. If foundational reading, sentence construction or basic task interpretation is weak, the tutor may need to adjust almost every example, slow the pace and protect the learner from constant comparison with stronger peers.

This does not mean the student should remain one-to-one forever. Once the foundation becomes stable, a small group may become useful precisely because it introduces broader variation and social transfer. Format can change as the learner changes.

15. Small groups can be better when the learner is strong but narrow

A strong student may already understand the tutor’s explanations. Their next need may be exposure to alternative interpretations, arguments, examples and writing structures. A group of capable peers can create productive disagreement that one-to-one tutoring has to simulate artificially.

For such learners, the tutor’s role shifts from constant explainer to designer of comparison. The group can test whether each student can defend a claim, recognise a stronger argument, revise after challenge and maintain individual voice.

16. Small groups are not automatically collaborative

Putting three learners in a room does not create collaboration. The tutor must decide when learners work alone, when they compare, when they explain, when they challenge and when they return to independent work. Without structure, one learner may dominate, another may imitate and a third may disengage.

Useful group roles are intellectual rather than administrative: evidence challenger, scope checker, audience reader, counterexample finder, summariser, questioner. Roles should rotate so no learner becomes permanently “the good writer” or “the weak reader.”

17. One-to-one is not automatically personalised

A tutor can sit with one learner and still use a generic worksheet sequence unrelated to the learner’s evidence. Personalisation means adapting the target, difficulty, support, feedback and transfer test because the student’s data justify the change. Ratio makes personalisation easier; it does not guarantee it.

A well-run small group can sometimes be more personally useful than generic one-to-one tuition if each learner has a clear target and the tutor deliberately uses shared tasks in different ways. One student may track inference scope, another evidence ranking, another vocabulary precision while all read the same text.

18. Think in terms of instructional density

Instructional density asks how much useful learning activity occurs per minute. In one-to-one, density can be high because transitions are fast and every example can target the learner. It can also be low if the tutor overexplains or corrects line by line. In a small group, density can be high when learners alternate independent work, peer comparison and tutor feedback. It can be low when students wait passively for their turn.

Families should observe what the learner actually does during a lesson rather than infer quality from group size alone.

19. Think in terms of decision ownership

Secondary 2 tutoring should steadily move English decisions back to the learner. Who decides what the question is asking? Who chooses evidence? Who selects the paragraph point? Who notices the sentence problem? Who decides whether an answer needs changing?

One-to-one can accidentally transfer ownership to the tutor because help is always available. A group can accidentally transfer ownership to stronger peers. The correct format is the one whose teaching design returns ownership to the student over time.

20. A first format decision rule

Use one-to-one first when the tutor needs to discover the learner’s process, repair a narrow high-cost mechanism or manage support so precisely that peer variation would initially obscure the diagnosis. Use a small group first when the learner can already work independently enough to benefit from comparison, explanation, dialogue and shared challenge.

Then review. Format is not an identity. A learner can move from one-to-one repair into small-group transfer, from group learning into a short one-to-one diagnostic block, or between formats across the year. The tutoring architecture should follow the learner’s current job.

Part II | How the Format Changes Each English Component

The one-to-one versus small-group decision becomes clearer when it is tied to the actual English component. Reading, writing, grammar, vocabulary, listening, oral response and editing do not all require the same interaction pattern. Secondary 2 is precisely the stage where a single format can be excellent for one task and inefficient for another.

21. Reading comprehension: one-to-one exposes the hidden path

Reading errors often appear only at the end, in the written answer. One-to-one tutoring can reconstruct the path that produced the answer. The tutor can ask what the learner thought the question required, which line first attracted attention, what alternative evidence was rejected and how certain the learner felt. This is valuable when the error is diagnostic rather than simply factual.

For a learner whose comprehension weakness is poorly understood, one-to-one can therefore accelerate discovery. The tutor can stop after one question and investigate deeply instead of moving with a group pace. That flexibility is particularly useful when the student’s school scripts contain several apparently unrelated errors.

22. Reading comprehension: groups make competing interpretations visible

Once the learner has enough reading control to make an independent attempt, a small group can make interpretation more rigorous. Three learners may choose three pieces of evidence for the same inference. The tutor can ask which is most direct, which only supports a possibility, and which is merely related. Students learn that evidence quality is comparative.

This is powerful for Secondary 2 because harder comprehension increasingly requires weighing clues, preserving qualification and distinguishing plausible interpretation from textual overreach. A group supplies real alternatives that force the learner to articulate criteria.

23. Reading danger in one-to-one: guided inference that looks independent

A skilled tutor can ask such good questions that the student appears to reason brilliantly. “What word limits that claim?” “Does this clue prove motive or only suggest it?” “What alternative remains?” These questions are useful while teaching. They become dangerous if the learner never has to generate them internally.

One-to-one reading therefore needs deliberate silent intervals and fresh questions without cues. The tutor should sometimes record the number and type of prompts. If comprehension quality improves but prompt count stays high, the format may be supporting performance more than independence.

24. Reading danger in groups: copying the fastest reader

In a group, a fast reader can anchor everyone else’s interpretation before slower students have completed their own reasoning. This produces apparent consensus and poor diagnostic visibility. The repair is simple but important: individual read, individual answer, individual evidence mark, then discussion.

When each learner commits first, peer ideas become comparison data instead of substitutes for thought.

25. Writing: one-to-one can trace the composition process

Writing is not one skill. It is task interpretation, idea generation, selection, planning, paragraph sequencing, evidence or detail, sentence construction, register, revision and proofreading. One-to-one tutoring can locate which stage first begins to fail. A tutor can watch the learner interpret the prompt, reject ideas, start a paragraph, pause, revise and explain decisions.

This is especially useful for a learner whose final writing looks weak for several possible reasons. A thin essay may begin with weak ideas, poor paragraph jobs or an inability to elaborate after examples. One-to-one observation can distinguish them before the tutor prescribes a generic writing framework.

26. Writing: groups create real audiences

Writing improves when students can see how different readers respond. In a small group, one learner can read another’s paragraph and explain what they think the main point is. If the writer intended something different, the gap becomes concrete. Audience awareness is no longer an abstract teacher instruction.

Groups also allow students to compare architectures. Two strong answers may use different paragraph orders. This can weaken template dependence and teach that structure follows purpose. Secondary 2 learners often need exactly this transition—from reproducing a taught pattern to making defensible organisational choices.

27. Writing danger in one-to-one: the tutor becomes co-author

The closer the tutor sits to one writer, the easier it is to become part of the composition engine. A tutor suggests the stronger word, asks the perfect development question, offers the missing example and rearranges the paragraph. The final piece improves. The student may not.

One-to-one writing should preserve version history and ownership. The learner drafts before intervention. The tutor marks the type of problem rather than rewriting immediately. The learner decides what to change. A fresh prompt later tests whether the same improvement appears without co-authorship.

28. Writing danger in groups: convergence toward one voice

Groups can create a different dependence: students copy the most fluent peer’s examples, phrases and structures. Over time, everyone’s writing begins to sound similar. The tutor should use comparison to reveal principles while protecting individual decisions.

A useful rule is “borrow the move, not the sentence.” If a peer’s paragraph works because it moves from example to implication, the learner should reproduce that relationship with different content and language. The transferable unit is the function.

29. Grammar: one-to-one can target the learner’s real error families

Grammar programmes become more individual by Secondary 2. One learner struggles with reference, another with clause boundaries, another with tense sequence under narrative load. One-to-one can use the learner’s own writing to build a small active risk map and practise precisely the patterns that recur.

This is efficient when the errors are idiosyncratic. It avoids asking a whole group to practise a rule that only one learner needs. The tutor can also test whether correction transfers from isolated sentences into original writing.

30. Grammar: groups can strengthen explanation and error discrimination

Small groups become valuable when grammar learning moves beyond rule recall. Give students several plausible sentences and ask them to decide which meaning each structure creates. One student explains; another challenges; a third offers a counterexample. The group learns that grammar is connected to meaning, not merely error hunting.

Peer explanation can also expose shallow knowledge. A learner may produce the correct answer but be unable to explain why. Another student’s question reveals the gap more naturally than another teacher correction.

31. Vocabulary: one-to-one can build a personal precision map

One-to-one is useful when vocabulary needs are tightly tied to one learner’s reading and writing. The tutor can collect recurring imprecise words, awkward collocations and register problems from actual work, then build contrast sets around them. The programme becomes a personal language corpus rather than a generic weekly word list.

32. Vocabulary: groups create richer lexical contrast

Small groups can generate more language options quickly. Ask three learners to replace a vague verb, then compare connotation, intensity and register. Ask each to use the same word in a different context. Discuss which collocation sounds natural and why. Vocabulary becomes a judgement task.

This is especially useful for strong Secondary 2 learners who know many words but need more precision. The group provides variation without requiring the tutor to manufacture every alternative.

33. Listening: one-to-one can diagnose where the spoken model collapses

A tutor can pause after a listening segment and ask the learner to reconstruct what changed. Did the student miss vocabulary, lose speaker identity, fail to notice a qualifier, or remember the first decision after the speaker revised it? One-to-one allows detailed diagnosis without the answers of peers contaminating recall.

34. Listening: groups can test interpretation before consensus forms

After individual listening notes, group comparison can be powerful. Students may disagree about the speaker’s final intention or which detail mattered. The tutor can ask them to reconstruct the audio evidence rather than simply reveal the answer. This trains listening as interpretation and updating.

The individual-first rule remains crucial. If students compare notes during the first listen, the tutor loses information about what each learner actually heard and retained.

35. Oral communication: small groups have a natural advantage in interaction

Oral English is not only producing a prepared answer. It includes listening to another person, responding to the actual point, clarifying, extending, conceding, disagreeing and recovering when challenged. A small group can generate these interactions authentically.

One learner gives a position. Another asks for an example. A third identifies an exception. The first learner must adjust. This creates a speaking environment that resembles real conversation more closely than repeated tutor–student question-and-answer exchanges.

36. Oral communication: one-to-one can help a hesitant learner enter the conversation

For a student who freezes, gives one-word answers or cannot organise a response under social pressure, one-to-one may be a useful temporary bridge. The tutor can slow the task, teach a simple structure, rehearse recovery and gradually increase unpredictability.

The exit condition matters. If oral confidence and structure improve, the learner may benefit from moving into a small group where interaction becomes less predictable and more socially real.

37. Editing: one-to-one can build a personal risk map

Editing becomes much more efficient when the tutor knows which error families recur in the learner’s own writing. One-to-one can track three active risks, remove them as they stabilise and test detection without location cues. The tutor can also notice whether the student edits by rule or by nervous rewriting.

38. Editing: groups can teach prioritisation

Give a group a flawed paragraph containing several possible improvements. Ask each learner to choose the first change they would make and justify why it has the highest cost. Students see that editing is not simply finding the largest number of errors. It is prioritising task, structure, sentence meaning and language risk.

This kind of editorial judgement can be especially valuable in Secondary 2, when students need to move beyond correction toward self-managed revision.

39. Summary and synthesis: one-to-one can isolate the learner’s compression error

A weak summary can fail because the learner cannot identify the main claim, keeps examples that should be removed, drops qualifiers, combines distinct points incorrectly or paraphrases by swapping synonyms. One-to-one questioning can identify which compression decision is failing.

40. Summary and synthesis: groups can expose what different readers considered essential

After each learner creates an independent summary, compare what was retained and omitted. The disagreement itself is useful. Why did one student keep a condition that another removed? Did one combine two points that were actually distinct? The group turns summary into a discussion about information hierarchy.

41. Literature and narrative reading: groups can widen interpretation without abandoning evidence

Narrative and literary texts often support more than one interpretation. A small group can help students experience that plurality while still requiring textual support. The tutor can insist that every reading answer two questions: what evidence supports it, and what evidence limits it?

One-to-one can do the same work, but the tutor must generate the alternative readings. In a well-matched group, peers naturally supply them.

42. Research and note-making: groups can divide perspectives, not responsibility

When Secondary 2 learners research an issue, small groups can assign different sources or viewpoints and then require each learner to explain the evidence. The activity teaches synthesis across perspectives. But every learner should still produce an individual final note set or response so responsibility does not disappear into group production.

43. Timing: one-to-one provides cleaner process data

If the learner’s problem is timing, one-to-one makes it easier to observe exactly where seconds and minutes disappear. The tutor can record start-up delay, rereading, planning, answer changes and review behaviour without confusing one learner’s process with another’s pace.

This makes one-to-one particularly useful for Leonie-type problems: sound knowledge but unstable execution. The tutor can teach a closure rule, run a short retest and see whether the changed behaviour actually saves time.

44. Timing: groups can normalise different efficient paths

Once timing is no longer a severe individual bottleneck, a group can show that efficient learners do not all work identically. One student may plan briefly but revise carefully. Another may annotate more and draft faster. The tutor can compare which choices preserve accuracy and which create avoidable cost.

45. Component choice can justify a hybrid format

The evidence above points to a practical possibility: the best Secondary 2 programme may not use one format for every purpose. A learner might begin with several one-to-one sessions to diagnose and repair answer scope, then join a small group to test interpretation against peers. A strong writer may use group discussion for ideas and one occasional individual conference for a recurring sentence-control problem.

Hybrid use should still be simple. Too many formats can fragment ownership and increase workload. The point is not to build a complicated schedule; it is to recognise that format is a teaching tool. Use the tool that fits the current job, then change it when the job changes.

Part III | Learner Fit, Group Composition and the Conditions for Good Small-Group English Tuition

Format decisions become much more precise when the learner is described by what happens during English work rather than by broad personality labels. “Quiet,” “bright,” “weak,” “visual learner,” “introvert” and “competitive” are too blunt. The useful questions are operational: Can the learner make an independent first attempt? Can they explain reasoning? Do they benefit from comparison? Do they lose attention while waiting? Do they copy stronger peers? Do they need a tutor to observe a very specific failure point?

46. A struggling learner is not automatically a one-to-one learner

Some struggling students benefit enormously from hearing another learner think aloud. A peer may explain a question in language that feels more accessible, and seeing another student make a similar mistake can reduce the sense that difficulty is private or shameful. A small group can also provide more examples of how a method is used.

The deciding factor is whether the learner can still participate meaningfully. If the gap is so large that most shared tasks are inaccessible, one-to-one or a more tightly matched group is likely to be more useful.

47. A strong learner is not automatically a group learner

Strong students can benefit from peer challenge, but only if the group contains enough intellectual variation to create it. If one learner consistently finishes first, supplies the explanation and waits while others catch up, the group may become inefficient. One-to-one stretch can be useful when the learner needs highly specialised challenge, advanced feedback or a pace that no available group can match.

48. A quiet learner may benefit from a group if participation is designed

Quietness should not be treated as evidence that a student cannot learn in a group. Some learners speak little in unstructured discussion but participate well when given preparation time, a specific question and a turn-taking routine. Written-first discussion can be especially useful: each learner records a position, then explains it.

The relevant measure is not how extroverted the student appears. It is whether the format produces enough meaningful spoken and written participation to support the target skills.

49. A very verbal learner may need a group more than they expect

A student who talks confidently in one-to-one lessons may dominate the conversation and receive constant tutor engagement. In a group, that learner has to listen, wait, respond to someone else’s idea and distinguish contribution from airtime. These are useful communication skills.

The tutor must prevent domination, but the constraint itself can improve oral discipline: answer the question asked, build on the previous point, concede when another interpretation is stronger, and speak with enough precision that others can respond.

50. A learner with attention-regulation difficulty needs task design, not a stereotype

Some learners focus better one-to-one because transitions are immediate and the tutor can redirect quietly. Others focus worse because one-to-one becomes an uninterrupted stream of adult attention. A well-designed small group can create short cycles of independent work, discussion and response that vary the cognitive demand.

Families should observe which environment supports sustained English thinking. If significant attention concerns extend beyond tutoring, the format decision should not substitute for appropriate professional assessment or school support.

51. A learner who copies easily needs strong individual-first rules

Some students rapidly adopt another person’s wording, plan or interpretation. They can appear to improve in a group while their own decision-making remains weak. This does not make group tuition impossible, but it raises the design requirement.

Require silent individual work before comparison. Keep original answers visible. Ask the learner to explain any change after discussion. Use a fresh individual task at the end. If the student still cannot produce the method independently, a temporary one-to-one block may be useful.

52. A learner who resists feedback may benefit from peer contrast

Some students interpret every tutor correction as a judgement. In a small group, comparison can depersonalise the issue. Instead of “your paragraph is weak,” the class compares three anonymous paragraphs and identifies which one explains evidence most clearly. The learner sees the criterion before applying it to their own work.

This can make feedback easier to accept because the standard becomes visible across examples. One-to-one can achieve the same effect with model contrasts, but peers often make the comparison more authentic.

53. A learner who needs privacy may need one-to-one temporarily

English weaknesses can feel exposing, especially when a student is far behind, has recently changed schools, is anxious about speaking, or is embarrassed by repeated writing errors. One-to-one can provide a quieter diagnostic space where the learner is more willing to reveal confusion.

Privacy should not become permanent protection from challenge. Once the learner has a workable method and enough confidence, carefully chosen group interaction may strengthen transfer.

54. A learner who needs challenge may need peers with different strengths

The strongest small groups are not necessarily groups of identical students. Useful asymmetry can create learning. One student may be a precise reader, another an inventive writer, another a strong speaker. If the tutor structures the lesson well, each learner encounters models of strength without being assigned a permanent hierarchy.

The difference must remain small enough that shared work is meaningful. Extreme gaps turn productive asymmetry into waiting and dependency.

55. Group size changes the instructional problem non-linearly

Moving from one learner to two is not the same as moving from five learners to six. In very small groups, each student can still receive frequent turns, the tutor can remember individual goals and the lesson can pivot quickly. As the group grows, waiting, generic instruction and behaviour management consume more time.

Ofsted’s review of tutoring in schools noted reduced effectiveness once groups became larger, with larger sessions tending to become more generic. EEF similarly treats small-group tuition as a distinct intervention whose strength depends on targeting. Families should therefore ask the actual group size and how individual targets are maintained, not accept “small group” as a marketing label.

56. A group of two behaves differently from a group of three

Two learners can compare directly and receive a high share of tutor attention. The risk is that one pairing becomes fixed: stronger and weaker, faster and slower, speaker and listener. A group of three creates more possible comparisons and can prevent one relationship from defining the lesson, but it also divides tutor attention further.

Neither ratio is inherently superior. The better choice depends on task design, learner match and tutor skill. Recent research on student–tutor ratios suggests that smaller ratios often help, yet the differences are not uniform enough to turn 1:1, 2:1 or 3:1 into universal laws.

57. Three learners can be an unusually productive English group

Three creates useful structures for English: writer, reader and editor; claimant, challenger and judge; speaker, questioner and observer; evidence selector, qualifier checker and summariser. Roles can rotate quickly, creating multiple perspectives without the scale of a conventional class.

The group works only if each learner also completes independent work. Three students producing one shared answer is not three students learning to answer.

58. Four learners can still be small, but design becomes more important

Four learners allow pair work, paired comparison and richer discussion. They also increase the chance that one learner can hide. The tutor needs routines that make thinking visible: individual written starts, randomised explanation, brief exit tasks and records of each learner’s active target.

59. Beyond a genuinely small group, the tutoring claim should be examined carefully

As the group grows, the tutor’s ability to observe individual reasoning and adapt immediately declines. The lesson may still be an excellent class, but the mechanism is changing. Families should ask whether the service is still tutoring in the sense they need or whether it is structured enrichment, exam preparation or supplementary teaching.

The distinction is not moral. A larger class can be useful. It simply should not be evaluated using the same promises as highly targeted tuition.

60. Match by current task demand, not only by school mark

Two learners with the same mark may be poor group matches. One may understand texts but write slowly; the other may read weakly but write fluently. Shared tasks can still work if the tutor differentiates goals, but the lesson becomes harder to design.

A useful match considers the kinds of decisions each learner is practising, how much support they need, and whether they can work on a common text or prompt without one student spending most of the lesson waiting.

61. Match by support level

A learner who needs a prompt every two minutes and a learner who can work independently for twenty minutes are difficult to teach together. The first needs high tutor availability; the second may become frustrated by interruptions. Similar support levels often matter more than identical grades.

62. Match by pace—but do not confuse pace with ability

Fast and slow workers can learn together if tasks have natural extension and reflection points. A fast student can compare alternative evidence while a slower student completes the first answer. But if the pace gap is extreme, one learner’s processing time can shape the entire lesson.

Speed should also be interpreted. A slow student may be careful rather than weak; a fast student may be shallow rather than strong. The tutor should match the actual learning process, not labels.

63. Match by willingness to think publicly

Groups work best when learners are willing to expose unfinished reasoning. They do not need to be confident speakers, but they need to be able to say, “I chose this evidence because…” or “I am not sure whether this word is too strong.” If a learner hides all uncertainty, the group loses diagnostic value.

Good tutors build this willingness through norms: errors are data, disagreement is about evidence, and changing one’s mind after stronger reasoning is a success rather than a loss.

64. Avoid fixed status roles

Groups can become harmful when identities harden: the good writer, the weak reader, the funny one, the slow one. Once students expect a particular person to supply certain answers, thinking becomes socially distributed in the wrong way.

Rotate task leadership. Use anonymous examples. Ask different learners to explain different strengths. Occasionally choose material that reverses the usual hierarchy. The group should expand each learner’s capability, not turn differences into permanent roles.

65. Friendship can help or hurt

Friends may feel safe challenging each other and speaking naturally. They may also distract one another, protect each other from criticism or rely on shared background that weakens explanation. The effect depends on whether the tutor can maintain intellectual norms without making the lesson socially tense.

Do not assume friends should be separated or kept together. Observe the learning behaviour.

66. Competition can motivate or distort

Some learners become more attentive when peers are present. Others start optimising for being first, sounding clever or avoiding visible mistakes. English judgement suffers when speed and performance status replace careful reasoning.

Use competition sparingly and on tasks where it does not distort the target. A race to locate evidence may train haste. A challenge to produce the most defensible explanation can be healthier if quality criteria remain explicit.

67. Group safety is an instructional condition

Students need to be able to offer an interpretation, make a language mistake or revise a position without humiliation. If the group climate punishes uncertainty, students will hide the very thinking the tutor needs to see.

Safety does not mean low challenge. It means disagreement is tied to evidence, criticism is specific, and every learner is expected to revise rather than defend ego.

68. One-to-one also needs psychological safety

Constant individual attention can feel intense. Some learners experience one-to-one tutoring as an hour in which every mistake is visible. A strong tutor uses wait time, normalises uncertainty and distinguishes diagnosis from judgement. They do not fill silence merely because only one student is present.

69. Learner fit can change across the year

A student may begin Secondary 2 needing one-to-one repair after a difficult transition, then move into a group once the foundation is stable. Another may thrive in a group for months but need a short individual block when a specific writing problem appears. Format should respond to evidence rather than become part of the learner’s identity.

70. The best group composition is dynamic enough to preserve challenge

A group that was well matched in January may not remain well matched. One learner can accelerate, another can develop a new bottleneck, and another may need a different writing emphasis. Tutors should review whether shared tasks still create useful work for everyone.

Dynamic does not require constant reshuffling. It requires willingness to acknowledge when the instructional logic has changed.

71. A learner can receive personalised tuition inside a group

Personalisation in a group is possible when the common task supports different targets. All three students may read the same article. Maren tracks how evidence is developed in writing; Iona tracks inference scope; Leonie tracks decision time and checking. The tutor asks different follow-up questions and assigns different transfer tasks.

This is more demanding than giving every student the same worksheet, but it preserves the efficiency and comparative value of the group while respecting individual diagnosis.

72. A learner can receive generic tuition one-to-one

The inverse is also true. If a tutor follows a fixed booklet regardless of the learner’s evidence, the lesson may be private without being meaningfully personalised. Parents should ask what changed in the lesson because of this student. If the answer is “nothing except pace,” the format’s potential is underused.

73. Small-group tutoring should record individual progress

Group work can make progress reporting vague: “the class covered inference this month.” That is not enough. The tutor should know whether each learner can perform the target independently, how many prompts are still needed and whether the improvement appears in fresh or school work.

A group programme remains accountable to individual learning.

74. One-to-one tutoring should still use external benchmarks

Because the tutor controls every task, one-to-one can create a closed ecosystem in which the learner becomes excellent at that tutor’s materials. Use school work, unfamiliar texts, alternative prompts and delayed tasks to ensure the programme is not measuring its own teaching too generously.

75. A second format decision rule

Ask two questions. First: How much individual diagnostic control does this learner need right now? Second: How much would this learner gain from seeing, explaining and challenging other people’s English decisions? High diagnostic need points toward one-to-one. High comparative-learning value points toward a small group. When both are high, a hybrid sequence can be justified.

Part IV | Instructional Design: What Good One-to-One and Small-Group Lessons Actually Do

76. A strong one-to-one lesson should not feel like permanent rescue

Because the tutor is always available, one-to-one lessons can become a chain of micro-rescues. The learner hesitates; the tutor hints. The learner writes a vague sentence; the tutor asks the question that fixes it. The learner chooses weak evidence; the tutor redirects. Every intervention may be individually sensible, yet together they can prevent the learner from experiencing enough uncertainty to build self-correction.

A strong one-to-one lesson therefore alternates high attention with deliberate withdrawal. The tutor observes closely but does not always intervene immediately. The student completes a meaningful unit of work, marks uncertainty, and only then receives support. The ratio stays one-to-one; the cognitive ownership shifts toward the learner.

77. A strong small-group lesson should not feel like a compressed classroom

If a tutor simply explains to three or four students at once and then assigns identical work, the group may be efficient but it is not using the format’s strongest possibilities. Small-group tutoring should exploit comparison, individual visibility, rapid turn-taking and targeted feedback.

A typical cycle might be: individual attempt, paired comparison, tutor sampling, short explicit teaching, individual revision, group challenge, fresh independent exit task. The tutor moves between individual and social learning rather than treating the group as one learner with several faces.

78. Begin both formats with a visible learning target

The learner should know the English decision being trained. “We are doing comprehension” is too broad. “We are learning to preserve qualifiers when we turn evidence into an inference” is actionable. In a group, the shared target may be common while each learner has a personal risk. In one-to-one, the target can be even narrower.

Visible targets also make format evaluation possible. If nobody can state the learning job, families cannot tell whether one-to-one attention or peer interaction is helping.

79. Use baseline tasks that do not flatter the format

A one-to-one tutor should not diagnose only while prompting. A group tutor should not diagnose only after discussion. Baseline evidence must include an independent first attempt under conditions that reveal what the student owns without the format’s support.

This principle prevents a common measurement error: comparing a one-to-one student’s heavily guided performance with a group student’s independent performance and concluding that the ratio caused the difference.

80. Explicit teaching should be brief enough to leave room for use

Both formats can be ruined by explanation-heavy teaching. A clear model or worked example is useful, but the learner must then perform the decision. In one-to-one, the tutor should resist turning every response into another explanation. In a group, the tutor should resist giving a mini-lecture because explaining to several students feels efficient.

A practical test is whether the lesson contains multiple moments when the tutor genuinely does not know what the learner will produce next. If every answer is tightly led, there is too little independent performance data.

81. Model thinking, not just polished answers

One-to-one tutors can model a decision aloud: “This evidence is vivid, but it does not directly answer the question, so I am rejecting it.” Group tutors can ask several learners to model different reasoning paths and compare them. In both cases, the valuable content is the selection process, not only the final answer.

Secondary 2 learners need to see expert criteria becoming decisions. This is how fixed formulas gradually become adaptable judgement.

82. Use the minimum-help principle

Support should begin with the smallest intervention likely to restart useful thinking. A question before a hint. A hint before a worked step. A worked step before a full model. In one-to-one, this prevents overhelping. In a group, it prevents the tutor from giving the whole class an answer because one student is stuck.

The minimum-help principle also generates data. If one light prompt is enough, the mechanism is closer to independence than if the learner requires a full demonstration.

83. Fade scaffolds differently in the two formats

One-to-one scaffolds often take the form of tutor questions, checklists, partial models and real-time reminders. Fade them by delaying the prompt, removing one checklist item, changing the surface and asking the learner to self-prompt.

Group scaffolds can include peer explanation and visible examples. Fade them by requiring individual answers before discussion, hiding peer work during a retest, changing the task after comparison and asking each learner to state the criterion independently. Peer support must also have an exit path.

84. Feedback in one-to-one should be selective

The tutor can notice many things, but the learner cannot act on all of them at once. Select feedback that is high-leverage and connected to the current goal. If the lesson target is evidence development, do not interrupt every sentence for minor vocabulary polish unless the language blocks meaning.

Selective feedback protects working memory and makes the next action clear.

85. Feedback in groups should remain individually attributable

“Everyone needs to elaborate more” is convenient group feedback but may be inaccurate for half the learners. A strong group tutor can teach a common principle while recording different personal applications. One learner may need evidence interpretation; another may need idea selection; another may already be stable and need qualification.

Shared teaching should not erase individual diagnosis.

86. Peer feedback needs criteria before opinion

Small groups can generate useful peer feedback only when learners know what to look for. “I like this paragraph” has little instructional value. “Your example is relevant, but I cannot see what it proves about the claim” is actionable because it uses a criterion.

The tutor should model one dimension at a time and audit peer comments. Students should not become substitute teachers responsible for correcting one another’s entire work.

87. Wait time matters more in one-to-one than many tutors realise

Silence can feel uncomfortable when only two people are present. Tutors often fill it. That steals retrieval time. A learner who pauses for ten seconds may be searching memory, testing wording or evaluating evidence. Intervening too quickly can turn healthy processing into dependence.

Good one-to-one teaching learns the difference between productive silence and lost silence. The tutor can ask the student to signal whether they are thinking, uncertain or stuck rather than assuming every pause needs rescue.

88. Productive waiting can be designed into groups

While the tutor works with one learner, others should have cognitively meaningful work: annotate evidence, rank alternatives, revise a paragraph, prepare a challenge, retrieve the current method or complete a micro-transfer task. Waiting becomes harmful when students have nothing purposeful to do.

This is one reason genuine small groups can work well: the independent intervals are short enough to remain connected to the lesson, yet long enough to test self-management.

89. Homework should not compensate for weak lesson design

A one-to-one tutor may assign large amounts of homework because the private lesson feels too short to cover everything. A group tutor may assign identical packets because individual practice time is limited during the session. Both responses can inflate workload without solving the underlying design problem.

Between-session practice should target the active mechanism, be short enough to complete well, and be checked in the next cycle. Quantity should not be used to make a format look rigorous.

90. Use school work to test transfer, not to consume the lesson

One-to-one tutoring can easily become a weekly review of whatever school assignment is due. Small groups can ignore school evidence because learners attend different schools or have different tasks. Both extremes are weak.

Use marked school work selectively to identify recurring mechanisms and later verify transfer. Then teach on fresh material. The tutor should not spend the whole lesson finishing or correcting school work line by line.

91. Use authentic texts differently in one-to-one and groups

One-to-one can choose texts tightly around one learner’s interests, background knowledge or current reading weakness. Groups can choose texts that support multiple viewpoints and discussion. A news feature, essay, short story or speech becomes more valuable when the tutor knows whether the aim is personal depth or comparative response.

92. Model answers should not erase the format’s advantages

In one-to-one, showing a model too early can replace the learner’s diagnostic attempt. In a group, one polished model can end discussion by establishing an apparent “correct” voice. Use models after independent work, analyse decisions rather than phrases, and then remove the model for fresh transfer.

93. AI can distort both format comparisons

If students use generative AI heavily between lessons, one-to-one tutors may see polished work that hides the learner’s actual state. Group tutors may compare outputs produced with different levels of AI assistance. Baseline conditions become unreliable.

For important diagnostics, require an AI-off independent attempt. When AI is used, define its job clearly and verify learning with fresh unaided work. The format decision should be based on the learner, not on tool-assisted output.

94. Online delivery interacts with group size

Online one-to-one can preserve high interaction density because tutor and learner share one screen, annotate together and switch materials quickly. Online small groups can work well, but participation management becomes more demanding. Recent experimental research in middle-school mathematics found suggestive evidence favouring one-to-one over 3:1 online tutoring, with tutors reporting greater challenges in personalisation, relationships, participation and behaviour in the small-group online condition.

That evidence is not an English-specific law, but it highlights a real design issue: small-group interaction that feels natural in a room may need more explicit structure online.

95. In-person delivery can make small-group interaction easier to read

In a room, the tutor can notice who is looking at whose paper, who has stopped writing, who wants to speak and who is quietly confused. Physical texts can be moved and compared easily. These signals can support group orchestration.

In-person one-to-one also benefits from shared physical space, but it does not automatically justify the format. Delivery mode and student–tutor ratio are separate decisions. See SETC’s site-wide Online vs In-Person English Tuition owner for the broader delivery comparison.

96. Cost changes the decision because opportunity cost is real

One-to-one usually costs more per learner because one tutor serves one student. Small-group tuition can distribute teaching cost across several learners. Research reviews often describe small groups as more cost-effective even when one-to-one has a somewhat larger average effect.

Families should not treat price as a proxy for quality. The relevant value question is what learning problem is being solved per unit of time, money and student effort. An expensive one-to-one programme that creates dependence can have poor value; a well-designed group that produces transfer can have high value.

97. Frequency and format should be decided together

A narrow one-to-one repair may benefit from a short period of relatively frequent contact. A group focused on discussion, writing range and transfer may work well with more independent practice between sessions. The evidence on tutoring often links effectiveness to regularity and sustained implementation, but private English tuition should still match frequency to the learner’s actual cycle.

More frequent one-to-one is not automatically better if the learner has no time to practise independently. More frequent group tuition is not better if each session repeats the same generic material. SETC’s site-wide How Often Should You Have English Tuition? page owns the general frequency question.

98. A one-to-one tutor needs excellent diagnosis

When all lesson time belongs to one learner, weak diagnosis becomes expensive. The tutor can spend an hour intensively teaching the wrong thing. Strong one-to-one tutors form hypotheses, test them, revise them and choose tasks that reveal the next uncertainty.

They should be able to explain why today’s task exists and what evidence would make them change the plan.

99. A small-group tutor needs excellent orchestration

Group teaching adds another layer of expertise: the tutor must manage attention without flattening individual needs. They decide when students work separately, when they compare, who speaks first, how to prevent copying, when to interrupt, and how to keep a stronger student challenged while another receives help.

A tutor who is excellent one-to-one is not automatically excellent with a group. The pedagogical demands overlap but are not identical.

100. One-to-one needs a deliberate independence metric

Because support is abundant, one-to-one programmes should track something that proves the learner is needing less of it. Prompt count, independent planning, unaided first attempts, delayed transfer and school evidence are useful. Without an independence metric, the format can hide its own dependency cost.

101. Small groups need a deliberate participation metric

Because attention is shared, group programmes should track whether each learner is doing enough of the thinking. Useful measures include independent answer completion, explanation turns, peer questions, revision decisions and exit-task performance. Attendance in a discussion is not the same as participation in learning.

102. Both formats need a transfer metric

The decisive test is whether the learner can use the skill on fresh, school-like work without the supportive features of the lesson. One-to-one support must disappear. Group comparison must disappear. The new task should preserve the underlying mechanism while changing the surface.

If the student succeeds only in the tutoring format, the programme has not yet completed its job.

103. Both formats need a stopping or switching rule

One-to-one may stop being the best format once a narrow repair becomes stable and peer variation would create more value. A small group may stop being the best format if a learner develops a highly specific gap, falls far behind the group or becomes too able to hide behind others.

Format review should be normal, not treated as evidence that the original choice failed. The learner changed; the instructional need changed with them.

104. A high-quality programme can recommend a cheaper format

If a one-to-one learner no longer needs intensive individual attention, a responsible tutor may suggest moving to a group, reducing frequency or pausing. If a group learner only needs one narrow repair, the tutor may suggest a short individual block rather than selling another full term of group lessons.

The ability to recommend less support is evidence that the programme is organised around learning rather than retention.

105. A third format decision rule

Observe the lesson through three lenses: Who is doing the thinking? How quickly does useful feedback arrive? What disappears when the tutor or peers disappear? One-to-one is strong when individual attention produces sharper diagnosis and then fades. Small-group tuition is strong when peer presence creates better comparison and then the learner can perform alone. Whichever format leaves the learner more independent on ordinary school English is doing the more important job.

Part V | Secondary 2 Format Casebook

Format decisions become easier when they are attached to a concrete learner state. The following cases are not rules that classify children permanently. They are decision tests. Each case asks what the format can reveal or train now, what risk comes with that format, and what evidence would justify switching later.

106. Maren: fluent writer, shallow evidence development

Maren writes easily, has enough vocabulary and rarely struggles to start. Her weakness is that examples are relevant but under-analysed. One-to-one can help initially because the tutor can stop at the exact sentence where explanation disappears and ask Maren to identify what the example proves. After the mechanism is visible, a small group may be more valuable: Maren can compare how different writers develop the same evidence and judge which explanation adds genuine meaning rather than extra words.

Format path: short one-to-one diagnosis, then group comparison and fresh independent writing.

107. Iona: strong evidence finder, overbroad inference

Iona finds textual clues quickly but writes claims that are too certain. One-to-one is useful if the tutor needs to reconstruct how she moves from evidence to conclusion. Once she can preserve qualifiers, group work becomes powerful because peers generate alternative interpretations. Iona must compare which claim is safest and why.

Format path: one-to-one to expose the overreach mechanism, then group interpretation to strengthen calibration.

108. Leonie: good methods, unstable timing

Leonie understands the methods but rereads, reopens secure answers and checks too broadly. One-to-one gives cleaner timing data because the tutor can observe exactly where the process stalls. A group becomes useful later when Leonie compares efficient decision paths and practises maintaining her closure rules while other students work at different speeds.

Format path: individual execution repair first, then group robustness if useful.

109. Case: the learner is far below the group’s reading level

The student needs extensive vocabulary support and loses the meaning of dense sentences before higher-level questions begin. In a shared group text, the tutor spends disproportionate time repairing basic access while stronger students wait.

Decision: one-to-one or a more closely matched group is likely to be better temporarily. The aim is to restore enough access that the learner can later benefit from peer comparison rather than remain isolated indefinitely.

110. Case: the learner is ahead of the group in reading but weak in discussion

The student reads difficult texts well but gives terse oral responses and rarely engages with alternative views. Moving to one-to-one because the learner is “advanced” could remove the very interaction they need.

Decision: keep or seek a strong small group with peers capable of challenging ideas. Use advanced texts if necessary, but preserve dialogue as part of the learning job.

111. Case: the learner needs immediate correction to avoid rehearsing a misconception

A recurring sentence error changes meaning and the student cannot yet detect it. Repeating the pattern in a group while waiting for feedback may strengthen the wrong form.

Decision: a short one-to-one block may be efficient. Once detection improves, move the skill back into whole writing and, if appropriate, group work where the learner must monitor it without constant correction.

112. Case: the learner understands better after explaining to peers

The student often gives the right answer but cannot articulate why until another learner asks a question. Explaining forces the method to become explicit and reveals gaps in reasoning.

Decision: a small group is valuable because peer questions create retrieval and explanation demands that one-to-one tutoring can simulate but not always reproduce naturally.

113. Case: the learner is too embarrassed to read aloud in a group

The student’s oral reading contains phrasing and self-correction problems, but social anxiety prevents useful participation. Forcing group performance may turn the lesson into avoidance rather than learning.

Decision: begin one-to-one or with a very safe pair. Teach the skill, build credible success, then widen the audience gradually if oral participation is part of the long-term goal.

114. Case: the learner is comfortable only one-to-one

The student has attended private tuition for years and performs well there but struggles to contribute in class. The familiar format may now be part of the problem.

Decision: test a structured small-group environment. The learner may need to practise waiting, listening, disagreeing and responding to less predictable people. One-to-one success should not become a reason to avoid social transfer forever.

115. Case: the learner disappears in groups

The student completes written work but rarely speaks, asks questions or reveals uncertainty. The tutor has little evidence of the reasoning process and the learner benefits mainly from listening to others.

Decision: first improve group design—individual roles, prepared responses, randomised explanation and exit tasks. If the learner still cannot participate enough for the tutor to diagnose or teach effectively, a temporary one-to-one block may be warranted.

116. Case: the learner dominates groups

The student answers first, interrupts, supplies peers with wording and treats speed as evidence of strength. Moving one-to-one may reward the domination by giving constant airtime.

Decision: keep group learning if the tutor can enforce listening and response constraints. Ask the learner to speak last, summarise another person’s argument before adding their own, or identify what changed their view. The group can teach disciplined interaction.

117. Case: two friends distract each other

Their shared humour consumes transitions and discussion drifts away from the text. The tutor repeatedly redirects them.

Decision: change the group composition or use stronger task routines before assuming both need one-to-one. The issue is interaction quality, not necessarily learning ability.

118. Case: two friends challenge each other productively

The same social familiarity can produce the opposite result. They are willing to say, “That evidence does not prove your point,” and neither treats disagreement as personal.

Decision: preserve the pair or group if independent transfer remains strong. Productive challenge is an instructional asset.

119. Case: the learner needs highly specialised writing feedback

A strong writer is working on subtle control of qualification, paragraph movement and sentence emphasis. Available groups are focused on more basic composition problems.

Decision: one-to-one may offer better fit because the tutor can spend time at the learner’s actual ceiling. This is a case where individual intensity supports stretch rather than repair.

120. Case: the learner needs more ideas, not more correction

The student writes accurately but generates predictable arguments and examples. Private correction is unlikely to solve a limited idea pool.

Decision: a strong discussion group may be more useful. Exposure to different viewpoints, examples and objections can widen the learner’s conceptual range, provided final writing remains independent.

121. Case: the learner has specific school-method confusion

The student has been taught one method at school and another in tuition. They cannot tell when each applies and performance has become less stable.

Decision: one-to-one may be useful briefly because the tutor can compare the exact methods and reconcile terminology. The goal is simplification, not building a third system.

122. Case: the group uses one school’s worksheets for everyone

Students from different contexts complete material designed around one school’s sequence. Some learners are ahead, others lack prerequisite teaching, and the tutor treats the worksheet as the curriculum.

Decision: the problem is not small-group tuition itself but weak targeting. Ask whether the tutor can select common skills and differentiate tasks. If not, a different group or one-to-one may be better.

123. Case: the one-to-one tutor follows a fixed workbook

The learner receives private attention but every lesson advances to the next page regardless of school evidence. The ratio is personalised; the curriculum is not.

Decision: do not pay for format while ignoring pedagogy. Ask how tasks are selected and how the plan changes when the learner’s evidence changes.

124. Case: the learner improves in group discussion but not school writing

The student has better ideas during lessons and contributes intelligently, yet written school work remains thin. The group is producing useful reasoning but the return path to independent writing is missing.

Decision: keep the group if it adds conceptual value, but add individual post-discussion writing and delayed transfer. One-to-one is not necessarily required unless the writing bottleneck remains unclear.

125. Case: one-to-one progress is strong but class participation remains weak

The learner writes and reads well privately but avoids discussion and oral response in school. The remaining problem is social execution, not comprehension.

Decision: consider a small group or other structured speaking environment. More private explanation may have diminishing returns.

126. Case: the learner needs recovery after a poor term

Several components have deteriorated and the student feels overwhelmed. A group may provide social normalisation, but the tutor first needs to distinguish backlog from one high-leverage bottleneck.

Decision: begin with a short individual diagnostic phase if necessary, then decide whether recovery can continue in a matched group. Do not assume the entire recovery must remain one-to-one.

127. Case: the learner is strong and bored one-to-one

The tutor asks good questions, but the student can predict the conversational pattern. There are few genuine surprises or competing viewpoints.

Decision: a strong small group may add productive unpredictability. Peers can introduce examples and interpretations the tutor did not pre-plan, forcing the learner to adapt rather than perform a familiar dyadic routine.

128. Case: the learner is weak and intimidated by stronger peers

The student stops attempting after hearing quick answers from others and begins saying “same as them.” The group is reducing independent evidence.

Decision: first use protected individual-first work and tasks with differentiated challenge. If intimidation continues or the gap is too large, move to one-to-one or a better-matched group.

129. Case: the learner is weak but energised by peers

Another struggling student becomes more willing to attempt difficult work when peers are present. They ask questions, compare answers and persist longer than in private lessons.

Decision: keep the group if the tasks remain accessible and the tutor records individual progress. Motivation can be a real format advantage when it results in more useful thinking rather than social distraction.

130. Case: the student needs frequent oral rehearsal

The learner can plan answers but needs repeated practice responding to follow-up questions. One-to-one provides many turns but only one interaction partner.

Decision: combine the strengths. Use one-to-one for initial structure if needed, then a small group for varied questioning and response. If only one format is possible, choose the one that best matches the current weakness.

131. Case: the learner has a narrow editing weakness

Everything else is stable, but ambiguous reference recurs in original writing. A full group programme would spend too much time on unrelated skills.

Decision: a short one-to-one intervention or consultation may be more efficient than ongoing group tuition. The support should end after detection and transfer stabilise.

132. Case: the learner has several moderate weaknesses

No single gap is severe. Reading, writing and oral work all need consolidation. The student can work independently and benefits from examples.

Decision: a well-matched small group may provide better breadth and value, provided the tutor still tracks individual targets. One-to-one intensity may be unnecessary.

133. Case: the learner’s needs change before Secondary 3

Early in Secondary 2, the student needed grammar repair. By the final term, grammar is stable and the main need is argument, unfamiliar texts and discussion.

Decision: reconsider format. A group may now offer more value than continuing the same individual arrangement. The correct answer can change because the learner changed.

134. Case: the group is excellent but scheduling causes constant absences

The teaching quality and peer match are strong, but the learner repeatedly misses sessions. Continuity breaks and make-up arrangements create extra burden.

Decision: practical fit matters. A slightly less ideal instructional format that the learner can attend consistently may outperform an excellent group attended irregularly. Do not separate pedagogy from implementation.

135. Case: one-to-one is affordable only at very low frequency

The family can choose one private lesson every few weeks or a high-quality small group weekly. The student has ongoing consolidation needs rather than a rare specialist question.

Decision: compare instructional dose and quality, not prestige. A regular, targeted group may provide more useful practice and feedback than infrequent one-to-one. Conversely, if the learner only needs occasional expert diagnosis, the low-frequency private format may fit perfectly.

What the casebook shows

The same principle keeps returning: choose the format that exposes and trains the current mechanism with the least unnecessary support. One-to-one is not a permanent label for struggling, shy or ambitious learners. Small-group tuition is not a permanent label for sociable, average or strong learners. Format is a variable in the tutoring design. The learner’s evidence should be allowed to change it.

Part VI | How to Trial, Compare, Switch and Stop Formats

136. Do not choose from a brochure; choose from a hypothesis

A useful format decision begins with a sentence that can be tested. “I think one-to-one may help because my child’s main problem is a narrow inference-scope error that is difficult to diagnose in class.” Or: “I think a small group may help because my child already understands the methods but needs more discussion, comparison and independent response to other viewpoints.”

A hypothesis protects the family from vague promises. If the expected advantage does not appear, the format can be changed without turning the decision into a judgement about the child.

137. Define the baseline before the trial

Use one or two tasks that represent the actual concern. Record accuracy, support level, time, self-correction and the learner’s explanation of the difficulty. If writing is the problem, keep an independent draft. If comprehension is the problem, use an unfamiliar passage. If oral interaction is the problem, record how the learner responds before coaching.

Without a baseline, families may confuse novelty, rapport or easier material with improvement.

138. Define what the chosen format is supposed to add

For one-to-one, the added value might be precise diagnosis, faster adjustment, reduced embarrassment or intensive support fading. For a group, it might be peer comparison, varied interpretations, richer oral interaction, audience awareness or cost-effective regularity. State the expected advantage before the programme begins.

139. A one-to-one trial should test tutor restraint as well as tutor skill

Families naturally notice whether the tutor can explain. They should also notice whether the tutor can wait, observe and let the learner think. Does the tutor allow an independent attempt? Do they distinguish uncertainty from ignorance? Do they give the smallest useful prompt? Can they let a student make a recoverable mistake?

A tutor who knows a lot but cannot stop helping may be a poor fit for a learner whose main goal is independence.

140. A group trial should test visibility

Can the tutor see what each student thinks, or only hear the fastest speaker? Does every learner make an independent first attempt? Are quieter students asked to explain? Are stronger students challenged rather than used as unpaid assistants? Does the tutor remember individual targets?

If the group makes one learner invisible, the format is not delivering its promised advantage.

141. Test the same mechanism across both formats when possible

If a family is genuinely unsure, compare how the learner responds to the same underlying target. In one-to-one, practise evidence ranking with tutor questioning. In a trial group, practise the same target through peer comparison. Then give a fresh individual task. Which route produced clearer independent control?

This is more informative than comparing two entirely different lessons and attributing every difference to format.

142. Do not compare enjoyment alone

A student may enjoy one-to-one because they receive constant attention. Another may enjoy group lessons because friends are present. Enjoyment matters for sustainability, but it should be paired with evidence of learning. Ask whether the learner participated meaningfully, understood the target, performed fresh work and needed less support.

143. Do not compare immediate marks alone

One-to-one can produce fast visible improvement because feedback is immediate and tasks are tightly adapted. Group tuition may initially look slower while the learner develops comparison and independence. Conversely, a group may raise motivation quickly while a one-to-one programme is still diagnosing.

Use marks over time, but include prompt dependence, transfer, retention and school evidence before deciding the format is working.

144. Use a four-week format trial when the problem is reasonably narrow

A practical short trial can have four stages. Week 1: baseline and working hypothesis. Week 2: explicit teaching and supported practice. Week 3: reduced support and changed surface. Week 4: fresh transfer task plus review. The exact calendar is flexible, but the logic is useful because it prevents indefinite “trying” without a decision point.

Complex or broad problems may need longer, but every trial should still have reviewable evidence.

145. Use a longer trial when the goal is discussion, writing range or confidence under challenge

Some group advantages require repeated interactions before they become visible. A learner may need several weeks to become willing to challenge peers, explain unfinished reasoning or carry discussion insights into writing. Do not abandon a well-designed group after one quiet lesson.

The question is whether participation and independent transfer are trending in the right direction, not whether the learner became socially comfortable immediately.

146. Use a switch trigger rather than vague dissatisfaction

• Switch toward one-to-one if the learner’s problem remains hard to diagnose, the group pace repeatedly misses the need, support requirements diverge sharply, or the learner cannot participate enough to make thinking visible.
• Switch toward a small group if the learner is stable enough that private attention has diminishing value and peer comparison, discussion, audience or varied challenge would create more learning.
• Reduce tutoring altogether if independent transfer is stable and the format is continuing mainly from habit.

147. Switching format is not the same as changing tutor

A strong tutor may recognise that the learner now needs a different environment. The same organisation or tutor might offer both individual and group work. Conversely, a format can be theoretically appropriate while the specific tutor is weak. Separate the questions: Is the ratio wrong, or is the teaching wrong?

148. Changing tutors is not the same as changing goals

If the goal is vague, no format will solve the problem cleanly. Before moving from group to private tutoring, ask whether the original target was defined well enough. A student doing “general English improvement” may drift in either format. Rebuild the goal first if necessary.

149. Changing frequency can solve a format problem

A one-to-one lesson may be effective but too frequent, leaving no time for independent practice. A group may be excellent but too infrequent for a fragile new mechanism. Before abandoning the format, test whether dosage is the real issue.

150. Changing group composition can solve a group problem

A small group may fail because the learner match is poor rather than because group tuition is wrong. If one student is far ahead, another needs constant support and a third rarely speaks, the tutor may be trying to solve three incompatible problems simultaneously.

Re-matching the group can restore participation and pace without moving everyone to private lessons.

151. A one-to-one learner should periodically perform without the tutor

Schedule complete tasks in which the tutor does not intervene until the end. This reveals whether the learner can initiate, monitor and recover independently. If the tutor cannot tolerate watching the student struggle, the programme cannot measure its own success honestly.

152. A group learner should periodically perform without peers

Discussion can make everyone sound more capable. Use individual exit tasks, delayed homework with clear authorship, and school work to see what survives. The group should create better independent thinkers, not only better group conversations.

153. Use a transfer ladder before declaring either format successful

1. Success with full format support.
2. Success with reduced prompts or reduced peer access.
3. Success on a fresh but similar task.
4. Success on a delayed task.
5. Success on a more unfamiliar task.
6. Success in ordinary school work.

The farther the improvement travels, the stronger the evidence that the learner owns it.

154. Compare the format’s failure mode, not only its strength

Every format has a characteristic risk. One-to-one risks overhelping, tutor dependence, narrow interaction and high cost. Small groups risk invisibility, peer copying, pace mismatch, generic feedback and uneven participation. The best choice is often the format whose main risk can be managed most reliably for this learner.

155. Ask what happens when the learner is wrong

In one-to-one, does the tutor correct immediately, ask a question or let the learner inspect the evidence? In a group, does the learner hear the answer from a peer before thinking, or does the tutor preserve individual reasoning? Error treatment reveals the real pedagogy faster than promotional descriptions.

156. Ask what happens when the learner is right

Correct answers also need diagnosis. Did the learner know why? Would the method survive a changed question? In one-to-one, the tutor can probe depth. In a group, peers can challenge the reasoning. A format should not treat correctness as the end of thinking when the goal is durable control.

157. Ask what happens when the learner finishes early

In one-to-one, early completion can lead to deeper variation immediately. In a group, the tutor needs extension work that remains connected to the target: compare two answers, generate a counterexample, improve an explanation, or prepare a question for peers. If early finishers simply wait, the group is wasting its strongest students’ time.

158. Ask what happens when the learner falls behind during the lesson

One-to-one can slow immediately. A group tutor must decide whether to support the learner briefly, adjust the shared task, provide a scaffold or continue while the student completes a reduced version. The answer reveals whether the group can handle individual variability without losing everyone else.

159. Ask whether the format can handle a changed diagnosis

Suppose the apparent comprehension problem turns out to be vocabulary access. Can the one-to-one tutor change direction? Can the group programme differentiate without forcing the whole group onto vocabulary? Flexible diagnosis matters more than fidelity to a pre-sold syllabus.

160. Ask how the programme protects student voice

One-to-one can make the learner’s writing sound like the tutor. Groups can make everyone sound like the strongest peer. Ask whether first drafts are preserved, whether revisions are explained, and whether fresh work remains recognisably the student’s own. This is especially important when models or AI are also used.

161. Ask how the programme uses parents

One-to-one programmes sometimes create detailed parent feedback after every lesson; groups may provide only general summaries. Neither extreme is necessary. Parents need enough information to understand the target, evidence and next review, but they should not become nightly assistant tutors.

Good reporting reduces uncertainty without increasing home correction.

162. Ask how the programme handles busy school weeks

One-to-one can adapt quickly to the learner’s workload, but that flexibility can turn the lesson into homework rescue. Groups provide a stable curriculum rhythm, but may ignore individual overload. Strong programmes in either format can reduce practice volume while preserving the active target.

163. Ask how the programme handles school holidays

A one-to-one holiday block can repair one persistent mechanism deeply. A group can use the extra space for longer reading, writing workshops and discussion. Neither format should automatically fill the holiday with more work. Recovery is part of sustainable learning.

164. Ask how the programme prepares for assessments

One-to-one can target a personal risk map precisely. Groups can run integrated tasks and compare execution strategies. Near assessments, both should reduce unnecessary novelty and stabilise known methods. Full-paper volume is not a substitute for diagnosis.

165. Ask what happens after an assessment

One-to-one can analyse a marked script closely, but should avoid line-by-line correction of everything. Groups can compare common mechanisms, but should not assume identical errors have identical causes. The result should update the tutoring plan: maintain, repair, retire, investigate or stretch.

166. A format review should include the learner’s voice

Ask what the student finds useful and why. “I like private tuition because the tutor tells me what to do” may actually reveal dependence. “I like the group because I hear other ideas” may reveal real comparative learning. “I hate the group because I never finish before everyone discusses” may reveal a pace mismatch worth fixing.

Student preference is evidence, not the only decision rule.

167. Use a format scorecard without turning it into a ranking

• Diagnosis: Is the learner’s actual weak link visible?
• Participation: Is the learner doing meaningful thinking?
• Feedback: Is feedback timely and actionable?
• Independence: Are prompts or peer supports reducing?
• Transfer: Does improvement appear on fresh and school work?
• Fit: Are pace, challenge and interaction appropriate?
• Workload: Is the format earning the time it consumes?
• Sustainability: Can the family and learner maintain it without crowding out sleep, school and independent practice?

The scorecard is not meant to produce a universal winner. It makes the reasons for the current choice explicit.

168. A good format trial ends with one of four decisions

• Continue: the format is producing useful progress and independence.
• Modify: keep the format but change group composition, frequency, task design or support level.
• Switch: another format better matches the learner’s current need.
• Reduce or stop: tutoring is no longer earning its place.

169. Do not confuse continuity with quality

Families often stay with a format because the student is used to it. Continuity has value, but familiarity can also hide stagnation. Review the evidence periodically. A long-running one-to-one arrangement may need more peer challenge. A familiar group may no longer fit a learner whose needs have become specialised.

170. The format should become less important as independence grows

The ultimate goal is not to produce a student who thrives only in the perfect tutoring environment. It is to produce a student who can read, write, listen, speak, revise and recover across ordinary conditions. As that happens, the one-to-one-versus-group question should lose importance because the learner can carry the method across settings.

Part VII | What the Research Can—and Cannot—Tell a Secondary 2 English Family

Research is useful when it improves the questions families ask. It is less useful when average effects are converted into guarantees for one learner. Much of the strongest tutoring evidence comes from school-based interventions, and recent experimental work on student–tutor ratios is often in mathematics rather than English. This does not make the evidence irrelevant. It means the evidence should guide design principles while the learner’s own English data decides the application.

171. EEF finds positive average effects for both formats

The Education Endowment Foundation’s Teaching and Learning Toolkit reports a positive average impact for both one-to-one tuition and small-group tuition. The one-to-one estimate is somewhat higher on average, while small-group tuition is generally less costly per learner.

The practical implication is not “always choose private tuition.” EEF itself notes mixed findings in direct comparisons and suggests that the quality of teaching enabled by very small groups may matter more than the precise ratio in some contexts.

172. Average impact does not describe your child

An average effect combines many programmes, ages, subjects, tutors and implementation conditions. Your Secondary 2 learner may have a narrow writing issue, a broad reading gap, strong oral skills, weak timing, unusual support dependence or a very specific school context. The evidence says both formats can work. Diagnosis decides which work is needed.

173. Recent ratio research often favours smaller groups—but cautiously

A 2026 review by Kraft, Schueler and Falken reports a pattern of larger effects for smaller student–tutor ratios, while also noting that many individual ratio comparisons are imprecise and do not always reach statistical significance. Their summary supports an important middle position: lower ratios often help, but pairs and small groups can still preserve meaningful tutoring effects.

That conclusion is more useful than a simplistic hierarchy. It encourages families to ask what the extra individual attention actually enables and what may be lost when peers disappear.

174. A 2025–2026 online ratio study highlights personalisation challenges

Kraft and Lovison’s experimental study of an online middle-school mathematics programme found suggestive evidence that one-to-one outperformed 3:1 tutoring, while tutors reported greater difficulty personalising instruction, building relationships, fostering participation and managing behaviour in small online groups.

For Secondary 2 English, the lesson is not that 3:1 online groups are bad. The study is not an English trial. The useful warning is that adding learners changes the tutor’s orchestration burden, especially online. A group programme should show how it protects participation and personalisation rather than assuming the technology will do so.

175. A 2026 field experiment suggests quality can outweigh extra frequency

Rohen Shah’s 2026 field experiment compared equal-cost tutoring arrangements in a middle school. Two-student groups meeting twice weekly produced a significant mathematics gain, while three-student groups meeting more frequently did not show a significant gain. This is one study in one subject and context, so it should not be turned into a universal 2:1 rule.

Its broader design lesson is valuable: more sessions do not automatically compensate for reduced interaction quality. Families comparing a larger group with more frequent lessons against a smaller group with fewer lessons should inspect what each format enables during the session.

176. Ofsted’s tutoring review warns about groups that become too large

An independent review of tutoring in schools in England observed that larger tutoring groups were more likely to become generic and less able to address misconceptions or specific knowledge gaps. The review noted a noticeable reduction in effectiveness once groups moved beyond a small-group threshold.

Private tuition is not identical to school tutoring, but the mechanism is intuitive: as group size rises, individual observation and responsive adaptation become harder. Families should ask how many learners are actually present, not rely on the word “small.”

177. Small groups can produce strong outcomes at scale

Evidence should not be read only as a warning about groups. EEF’s Tutor Trust work reports positive outcomes from tutoring delivered in groups of three in ordinary school conditions. Other experimental research has found strong results from two-student online tutoring. These examples show that small groups can be effective when the programme is structured, targeted and implemented well.

178. Subject matters

Mathematics tutoring often involves worked procedures, problem types and immediate correctness in ways that differ from English discussion, interpretation and writing. English may gain more from peer comparison in some tasks because multiple defensible answers exist. It may also require more individual feedback in extended writing because each learner’s draft is unique.

Do not transfer a ratio finding from mathematics directly into English without examining the mechanism.

179. Age matters

Younger learners may require more direct attention for foundational decoding or basic language routines. Secondary 2 students can usually sustain longer independent work, explain reasoning and benefit from comparison. That developmental difference changes what a small group can accomplish.

At the same time, adolescence can heighten social sensitivity. Group safety, status and embarrassment matter. The format must allow the learner to expose unfinished thinking without fear.

180. Tutor expertise can interact with format

An expert diagnostician may create enormous value one-to-one because they can detect subtle mechanisms quickly. An expert discussion facilitator may create extraordinary group learning by turning peer differences into evidence. A tutor who is strong in one format may be weaker in the other.

Ask for evidence of the tutor’s actual practice, not only qualifications: how do they diagnose, differentiate, manage turn-taking, fade prompts and verify transfer?

181. Alignment with school learning matters in either format

EEF consistently emphasises connecting tuition with normal teaching. For private Secondary 2 English tuition, that does not mean copying the school syllabus blindly. It means understanding the learner’s school demands, using marked work as evidence and checking whether the repaired behaviour returns to ordinary class tasks.

A private tutor can become too detached from school. A group centre can become too standardised to one internal curriculum. Both should maintain a return path.

182. Targeting matters more than format branding

Both EEF strands emphasise targeting support to specific learner needs. This should make families sceptical of products sold mainly by ratio. “Premium one-to-one” is not a diagnosis. “Elite small group” is not a learning plan. Ask what need the format is designed to address.

183. Cost-effectiveness is an educational question, not only a financial one

Money spent on tutoring also buys or displaces time. A family choosing one-to-one may be paying for more immediate adaptation. A family choosing a group may be buying more regular contact for the same budget. Compare whether the resulting dose creates enough independent practice and feedback.

SETC’s How Much Should You Spend on English Tuition? page owns the broader cost decision. For Secondary 2, the important point is that cost should be tied to the instructional job rather than prestige.

184. Opportunity cost can reverse the apparent winner

Suppose one-to-one produces slightly faster improvement but requires travel, a high fee and additional homework that crowds out reading and sleep. Suppose a nearby small group produces slightly slower improvement but is sustainable and leaves time for independent practice. The theoretically stronger ratio may not be the stronger whole system.

Families should evaluate tutoring inside the learner’s actual week.

185. Small groups should not use peers as unpaid tutors

Peer explanation is useful, but one strong learner should not spend the lesson reteaching content while the adult tutor withdraws. Every learner is there to learn. Roles should create mutual intellectual work rather than shift instructional responsibility onto students.

186. One-to-one should not use the tutor as a permanent external executive function

The tutor can become the person who remembers every deadline, initiates every task, chooses every practice activity and decides when work is good enough. The student’s English may improve while self-management does not.

Secondary 2 should move toward learner ownership. The tutor can model planning and checking, but must eventually hand these functions back.

187. Examine the tutor’s use of questions

In one-to-one, high-quality questions should diagnose and then fade. In groups, questions should distribute thinking rather than merely test who knows the answer fastest. Ask whether questions reveal reasoning, require evidence and change according to learner responses.

188. Examine the tutor’s use of models

One-to-one tutors may customise models to one learner’s exact weakness. Group tutors can compare multiple models and let learners debate quality. In both settings, model exposure should follow a meaningful attempt when the first decision is part of the skill. The model should become less necessary over time.

189. Examine the tutor’s use of error

Does the lesson treat errors as something to erase quickly, or as evidence about the learner’s model? One-to-one creates rich error data. Groups create comparative error data. Good tutors use both to decide what to teach next rather than simply count mistakes.

190. Examine the tutor’s use of challenge

One-to-one challenge can be precisely calibrated: remove a prompt, change genre, increase abstraction or add time. Group challenge can come from peers: defend the claim, answer an objection, improve another version, or produce a different structure. Both formats should increase challenge by mechanism rather than merely by assigning harder material.

191. Examine what happens after successful guided work

The most revealing moment is often the next task. Does the one-to-one tutor keep helping in the same way? Does the group continue discussing before anyone works alone? Or does the programme remove support and ask for fresh transfer? Good tutoring treats guided success as a bridge, not the destination.

192. Examine how progress is reported

One-to-one reports can become overly detailed descriptions of lesson activity. Group reports can become overly generic summaries of topics covered. Useful reporting in either format states the target, evidence of change, current support level, transfer status and next decision.

193. Examine whether the learner can name the current target

If the student says only “we are doing English,” the programme may be too diffuse. A Secondary 2 learner should increasingly be able to say, “I am working on keeping my inference within the evidence,” “I am making my examples prove the point,” or “I am reducing unnecessary answer changes.” This metacognitive clarity matters in any format.

194. Examine whether the target retires

If the group covers the same topic every term because it is on the curriculum map, individual progress may not change the plan. If the one-to-one tutor keeps correcting the same pattern indefinitely, the repair may not be transferring. Targets should have exit criteria.

195. Examine whether the format creates evidence outside itself

A strong tutoring programme should eventually point to school scripts, independent drafts, oral performance or timed work that improved without the tutoring environment present. If all evidence comes from the tutor’s own materials, the programme may be measuring adaptation to itself.

196. Examine what the learner does between lessons

One-to-one can encourage excessive reliance if every difficult task is saved for the tutor. Groups can encourage passive dependence if learners wait for the next session to compare answers. Between lessons, students should retrieve methods, complete targeted practice and collect questions without outsourcing every uncertainty.

197. Examine whether the programme tolerates uncertainty

English often contains uncertainty: two plausible interpretations, several possible structures, a vocabulary choice that depends on register. One-to-one tutors should resist supplying certainty too quickly. Group tutors should resist forcing consensus merely to keep the lesson moving. Learners should practise holding alternatives and deciding from evidence.

198. Examine whether the learner is becoming more adaptable

Secondary 2 progress should increase range. A student should handle unfamiliar texts, varied prompts, different audiences and changing time conditions more reliably. One-to-one can engineer this variation deliberately. Groups can generate some of it naturally. Either way, the learner should become less dependent on familiar surface features.

199. Examine whether the format is still solving the original problem

Families often continue a successful arrangement after the original weakness has disappeared. At that point, the format may need a new goal, lower frequency or retirement. Do not let “it worked” become the reason it must continue forever.

200. What the evidence supports most strongly

The research base supports tutoring as a useful intervention when it is targeted and implemented well. It gives reasonable grounds to expect smaller student–tutor ratios to support more individualisation on average, while also showing that very small groups can produce meaningful and sometimes highly cost-effective gains. It does not support treating one-to-one as universally superior or treating every “small group” as equivalent.

For a Secondary 2 English learner, the evidence should lead to a disciplined local question: what kind of interaction will make this student’s next English decision more accurate, more transferable and more independent?

Part VIII | Parent Decision Toolkit, Provider Questions and Frequently Asked Questions

201. Start with the problem statement

Before contacting any tutor or centre, write one sentence describing the current English problem without naming a format. For example: “My child understands passages but writes inference answers that are broader than the evidence,” or “My child has good ideas but cannot organise them independently under time.” If the sentence instead reads “I need one-to-one tuition because it is more focused,” the solution has been chosen before the problem is defined.

202. Then state what one-to-one would need to accomplish

• Reveal a difficult-to-see process.
• Repair one narrow high-cost mechanism.
• Adjust difficulty or scaffolding rapidly.
• Protect a learner who cannot yet access shared tasks.
• Provide temporary privacy for a sensitive weakness.
• Deliver specialist stretch unavailable in a suitable group.

If none of these advantages is necessary, private tuition may still work, but the family should ask whether it is earning the additional intensity and cost.

203. State what a small group would need to accomplish

• Expose the learner to alternative interpretations and solutions.
• Create real discussion and oral-response practice.
• Build audience awareness through other readers.
• Require explanation and justification to peers.
• Increase task variation without losing tutor oversight.
• Provide regular high-quality tuition at a sustainable cost.

If the group cannot reliably produce these interactions, its theoretical advantages may exist only on paper.

204. Questions to ask a one-to-one tutor before starting

1. What evidence do you need before deciding what to teach?
2. How will you distinguish a knowledge gap from an execution gap?
3. How much of the first lesson will my child work independently?
4. How do you prevent one-to-one support from creating dependence?
5. How do you decide which errors not to correct immediately?
6. How do you use school work without becoming a homework-completion service?
7. What fresh task will show whether teaching transferred?
8. How will you measure prompt reduction?
9. What would make you recommend fewer lessons or a move to a group?
10. What would make you revise your original diagnosis?

205. Questions to ask a small-group tutor before starting

1. How many learners are actually in the group?
2. How are learners matched?
3. Does each student have an individual target?
4. How do you prevent faster learners from supplying answers too early?
5. How do quiet learners make their thinking visible?
6. What happens when one learner falls behind the shared task?
7. How do you differentiate without turning the lesson into several unrelated mini-lessons?
8. How is peer feedback taught and checked?
9. How do you test each learner independently after group discussion?
10. What would make you recommend temporary one-to-one support?

206. Ask to see the learning cycle, not just the materials

Beautiful notes and difficult worksheets do not show how learning happens. Ask what occurs before, during and after a task. Is there an independent attempt? How is the error diagnosed? What is taught? How is support faded? What fresh work tests transfer? A strong learning cycle matters in both formats.

207. Ask how progress changes the programme

If the learner improves, does the one-to-one tutor remove prompts and reduce intervention? Does the group tutor increase challenge or change the learner’s role? If progress does not alter the teaching, the programme may be following a curriculum rather than responding to evidence.

208. Ask what happens to a stable goal

A stable goal should retire. One-to-one programmes can otherwise keep revisiting the same strengths because the tutor knows the learner well. Group programmes can revisit them because the syllabus repeats. Retirement protects attention for current needs and creates a natural path toward less tuition.

209. One-to-one red flag: every lesson produces perfect work

If the tutor remains beside the learner through every decision, polished outputs may be inevitable. Ask for unassisted baselines and fresh tasks. A strong programme should sometimes produce imperfect work because the student is being allowed to reveal what they can do independently.

210. One-to-one red flag: the tutor is the student’s memory

The tutor reminds the learner of every strategy, deadline, error pattern and next step. The student no longer retrieves the system. Replace repeated reminders with student-generated checklists, delayed prompts and self-diagnosis. If the programme cannot fade these supports, independence is not improving.

211. One-to-one red flag: the same worksheet sequence would be used for anyone

Private tuition is expensive if only the seating arrangement is personalised. Ask what the tutor changed because of this learner’s evidence. Pace alone is sometimes a meaningful adaptation, but usually one-to-one should allow more precise changes to target, difficulty, support and feedback.

212. Small-group red flag: students hear the answer before attempting

If the fastest learner answers every question publicly, the tutor loses diagnostic data and slower students lose retrieval opportunities. Strong groups use individual thinking time before open discussion.

213. Small-group red flag: progress is reported only as topics covered

“We completed inference and discursive writing” says what the group did, not what this learner can now do. Ask for individual evidence: fewer prompts, stronger fresh answers, improved writing, better timing or clearer school transfer.

214. Small-group red flag: stronger students are always waiting

Occasional waiting is normal. Repeated waiting means the group is poorly matched or extension design is weak. Stronger learners should have deeper versions of the same intellectual job rather than unrelated extra worksheets.

215. Small-group red flag: weaker students are always copying

If a learner’s work improves mainly after seeing peers’ answers, the group may be masking rather than repairing the gap. Protect independent first attempts and use individual exit tasks. Persistent copying is evidence that the learner may need a different support level.

216. Red flag in either format: guaranteed outcomes

Tutoring outcomes depend on baseline, attendance, school context, learner effort, teaching quality and the nature of the problem. A responsible tutor can describe evidence and likely mechanisms without guaranteeing a particular grade. Certainty in marketing should not be confused with certainty in learning.

217. Red flag in either format: more work is the answer to every problem

If progress stalls and the response is always more worksheets, more homework or more sessions, diagnosis may be weak. Sometimes the learner needs a different explanation, a smaller task, a changed format, a lighter workload or a period of independent transfer.

218. Red flag in either format: no exit path

Ask what would make the tutor say, “This learner no longer needs this level of support.” If the answer is unclear, the programme may be organised around continuity rather than capability. Secondary 2 tutoring should build toward Secondary 3 readiness and greater independence.

219. The Secondary 3 transition can change the format decision

As Secondary 2 closes, some weaknesses become more expensive: shallow inference, template writing, unstable evidence use, weak sentence control, overchecking and heavy prompting. A learner with one unresolved bottleneck may benefit from a short individual repair before Secondary 3. A learner whose foundation is stable may benefit more from a strong group that increases unfamiliarity, discussion and performance range.

See How Should Tuition Prepare a Student for Secondary 3 English? for the transition owner.

220. A final format decision matrix

Current need	One-to-one may add more value when…	Small group may add more value when…
Diagnosis	The failure point is unclear and needs close process observation.	Comparing several learners’ approaches helps expose the misconception.
Reading	The learner needs precise cueing and evidence-boundary repair.	The learner benefits from alternative interpretations and evidence ranking.
Writing	The tutor needs to trace a highly individual composition bottleneck.	The learner needs audience, structural comparison and idea diversity.
Grammar/editing	Error families are highly individual or need intensive detection work.	Students can compare meaning, prioritise errors and explain rules.
Oral	The learner needs a safe bridge into speaking or very targeted rehearsal.	The learner needs unpredictable interaction, challenge and response.
Timing	Precise observation of decision cost is needed.	The learner is ready to compare efficient processes and perform amid variation.
Stretch	The learner needs specialist challenge unavailable in a suitable group.	Peers can provide genuine intellectual disagreement and varied solutions.
Recovery	The learner is far behind or needs a short intensive repair.	The learner can access shared work and benefits from peer examples and motivation.

The table is not a ranking. It maps needs to affordances. The same learner may move across columns during the year.

Frequently asked questions

Is one-to-one English tuition always better because the tutor gives full attention?

No. Full attention can improve diagnosis and adaptation, but it can also create overhelping and dependence. Small groups can add peer comparison, discussion and varied reasoning. The better format is the one that fits the learner’s current problem and produces independent transfer.

Is small-group tuition just a cheaper version of one-to-one?

No. A strong group has different pedagogical advantages: alternative interpretations, real audiences, peer questions, oral interaction and comparative judgement. If the group is merely a tutor lecturing several students, then those advantages are being wasted.

What is a good small-group size for English tutoring?

There is no universal number. Research generally supports very small groups and suggests effectiveness can decline as groups grow. For private English tuition, two to four learners can preserve frequent participation when the tutor is skilled and the group is well matched. The actual design matters more than attaching a magic number to the label.

Is a group of three too small to create discussion?

No. Three learners can create rich comparison because roles can rotate: speaker, challenger and evaluator; writer, reader and editor. The tutor still needs to protect individual first attempts and ensure no one becomes permanently dominant.

Should a struggling Secondary 2 student always start one-to-one?

No. If the student can access shared tasks and benefits from peer explanation, a well-matched small group may work well. One-to-one becomes more compelling when the learner is far behind, the failure point is unclear, or support requirements are too individual for the available group.

Should a strong student always choose a small group for challenge?

No. Strong students often gain from peer disagreement, but a highly advanced or specialised goal may be better served individually. The question is whether the available group can genuinely challenge the learner rather than simply give them more work.

Which format is better for comprehension?

One-to-one is often useful when the tutor needs to trace the learner’s reading process closely. A small group is often useful when the learner is ready to compare evidence and alternative interpretations. Many students can benefit from an individual-diagnosis phase followed by group transfer.

Which format is better for composition?

One-to-one can diagnose a personal planning, development or sentence-control bottleneck. Groups can provide audiences, alternative structures and idea diversity. The learner should still draft and revise independently in either format.

Which format is better for oral English?

Small groups have a natural advantage for interaction because learners respond to different people and viewpoints. One-to-one can be useful as a safe starting point for a hesitant learner or for precise coaching. The long-term goal should include responding to less predictable interlocutors.

Which format is better for grammar?

One-to-one can target highly personal error families efficiently. Groups can improve grammar judgement through contrast and explanation when learners share enough common ground. The success test is transfer into original writing, not worksheet accuracy.

Can a child switch from one-to-one to group tuition?

Yes. This can be a strong progression when a narrow weakness has been repaired and the learner now needs variation, discussion and peer challenge. The switch should be based on evidence of readiness rather than a fixed timetable.

Can a child switch from group tuition to one-to-one?

Yes. A temporary switch can be useful when a learner develops a highly specific bottleneck, falls behind the group pace or needs more diagnostic privacy. The learner can return to a group after repair if peer interaction becomes useful again.

Should one-to-one tuition be more frequent because it is more effective?

Not automatically. Frequency should match the learning cycle. Too much private tuition can crowd out independent practice and create support dependence. Regularity matters, but the learner needs time to retrieve and apply methods without the tutor.

How can I tell if a small group is too large?

Look for practical consequences: long waiting, generic feedback, invisible learners, limited speaking turns, poor differentiation and difficulty tracking individual goals. The number matters because it changes these interactions, not because one threshold is magically correct everywhere.

What if my child says they learn better alone?

Ask what “better” means. They may value concentration, privacy or immediate help. Those are real factors. Also test whether the learner is avoiding useful challenge, discussion or uncertainty. Preference should inform the decision without becoming the only evidence.

What if my child says they learn better with friends?

Again, inspect the mechanism. Friends may increase persistence and discussion, or they may increase distraction and answer sharing. The relevant evidence is whether participation, understanding and independent transfer improve.

Can online small-group English tuition work?

Yes, but it needs strong participation design. Tutors should manage turns, individual first attempts, shared documents and breakout or paired work carefully. Recent ratio research in online mathematics highlights the additional orchestration challenge, which is a useful caution rather than a prohibition.

Can online one-to-one English tuition work?

Yes. Shared annotation, digital texts, recorded oral practice and rapid resource changes can support effective teaching. The same dependency safeguards still apply: the learner must eventually perform without real-time tutor rescue.

Does one-to-one guarantee faster progress?

No. It can accelerate diagnosis and provide more individual practice, but weak targeting, overhelping or poor tutor fit can waste the intensity. Progress depends on the mechanism, quality, dosage, learner participation and transfer.

Does group tuition build confidence better?

It can when learners feel safe, see peers struggle productively and experience successful discussion. It can reduce confidence when comparison becomes status competition. One-to-one can also build confidence through credible independent success. Confidence follows the quality of the learning experience, not the ratio alone.

How do I know if one-to-one is creating dependence?

Check whether prompt count, model use and tutor intervention are falling. Give fresh independent tasks. If school performance remains weaker than tuition performance and the gap does not close, the tutor may be carrying too much of the process.

How do I know if a group is hiding weakness?

Compare group discussion with individual exit tasks. Preserve first attempts. Ask the learner to explain changes made after hearing peers. If performance collapses when peer answers are unavailable, the group is not yet producing independent control.

Should siblings study in the same group?

Only if the instructional match works. Familiarity can support discussion or create fixed family roles. Age, level, pace and the ability to disagree productively matter more than the relationship itself.

Should schoolmates study together?

They may share useful school context, but the group should not become a place to complete common homework collaboratively. The tutor must preserve authorship and focus on transferable skills.

What if the group has mixed school levels?

Mixed levels can work when the core English decision is shared and tasks are differentiated. They work poorly when one learner cannot access the text or another has already mastered the target. Ask how the tutor designs common intellectual work across different starting points.

What if one learner is absent from a small group?

The tutor should be able to run the lesson without turning it into an entirely different product. Absence may change discussion possibilities, but the remaining learners still need individual targets and independent transfer. Repeated absence is an implementation problem worth addressing separately.

Is peer correction reliable?

Not automatically. Peer feedback should be constrained by criteria and checked by the tutor. It is most useful for making reasoning visible and developing judgement, not for outsourcing professional diagnosis to other students.

Should a one-to-one tutor teach everything the school teaches?

No. Private tuition should prioritise the mechanisms that need additional support or meaningful stretch. Duplicating every classroom lesson can waste the format’s ability to target precisely.

Should a small group follow a fixed curriculum?

A coherent curriculum can provide structure, but it should not override learner evidence. Strong group programmes combine common progression with room to differentiate targets, difficulty and feedback.

What is the strongest evidence that one-to-one is working?

The learner can perform the repaired decision on fresh and school work with fewer prompts than before. The tutor’s individual attention is becoming less necessary because the student has internalised the method.

What is the strongest evidence that small-group tuition is working?

The learner uses comparison and discussion to sharpen judgement, then reproduces the improved decision alone. Peer presence adds learning during the lesson without becoming necessary for later performance.

When should format stop being the main question?

When the learner is becoming stable and independent. At that point the important question is whether tutoring itself is still needed, not whether the next lesson should contain one student or three.

The Secondary 2 decision in one sentence

Choose one-to-one when intensive individual observation and adaptation are the missing ingredients; choose a genuinely small group when peer comparison and interaction are the missing ingredients; and change the format when the learner’s evidence says the missing ingredient has changed.

Further reading and evidence

• Education Endowment Foundation: One-to-one tuition
• Education Endowment Foundation: Small-group tuition
• National Student Support Accelerator: Five Years of Tutoring Research
• Kraft and Lovison: The Effect of Student–Tutor Ratios
• Kraft, Schueler and Falken: What Impacts Should We Expect From Tutoring at Scale?
• Shah: The Trade-off between Quality and Quantity
• UK Department for Education: Independent review of tutoring in schools
• Nickow, Oreopoulos and Quan: Systematic review and meta-analysis of tutoring

Where this page fits

This page owns only the Secondary 2 application of the one-to-one versus small-group decision. For the general cross-age comparison, use One-to-One vs Small-Group English Tuition. If you have not yet established whether tutoring is needed at all, begin with Does My Child Need an English Tutor?. For tutor quality, use What Makes a Good Secondary 2 English Tutor?. For personalisation inside either format, use How Should English Tuition Be Personalised?.

After choosing a format, use What Progress Should English Tutoring Produce? to measure whether the programme is working and When Should English Tutoring Stop? to keep the exit path visible.

Final principle

One-to-one and small-group tutoring are not opposing philosophies. They are different environments for producing learning. One concentrates adult attention; the other adds carefully managed peer variation. Secondary 2 students need both precision and adaptability, but not always at the same moment. Choose the format that makes the current weak link teachable, makes the learner do the necessary thinking, and makes the support progressively less necessary.

Final Format Audit | Twelve Last Checks Before You Commit

A final decision should survive more than preference and price comparison. Use these twelve checks to test whether the proposed format has a clear instructional reason. If several answers remain uncertain, the right next step may be a short diagnostic or trial rather than a long commitment.

1. Can you name the learner’s current English bottleneck?

If the answer is only “English is weak,” neither one-to-one nor small-group tuition has a sufficiently defined job. Gather school evidence, independent work and teacher feedback first. The more precisely the bottleneck is described, the easier it becomes to decide whether intensive individual observation or comparative peer work is the missing ingredient.

2. Does the format make the bottleneck more visible?

A one-to-one arrangement should reveal process that was previously hidden: hesitation, evidence choice, planning, checking, retrieval or error detection. A group should reveal comparative reasoning: how the learner’s choices differ from peers and whether they can explain, defend or revise them. If the format hides the mechanism, it is poorly matched.

3. Does the student still make an independent first attempt?

This is a non-negotiable diagnostic safeguard in many English tasks. Private tutors should not lead the first move simply because they can. Group tutors should not let the fastest peer lead it for everyone. Independent first attempts preserve authorship and make later support measurable.

4. Can the tutor explain how support will be faded?

One-to-one support may fade through longer wait time, fewer hints, removed checklists and complete independent tasks. Group support may fade through hidden peer answers, individual exit work and delayed transfer. If the support has no planned reduction, the learner may improve only inside the format.

5. Does the learner get enough useful struggle?

Good tuition prevents pointless frustration but does not remove every difficulty. One-to-one tutors must sometimes watch without rescuing. Group tutors must sometimes stop peers from helping too early. The learner needs enough unresolved space to retrieve, choose, test and recover.

6. Is feedback changing the next attempt?

Feedback should not accumulate as comments. The student should act on one or two important points, then demonstrate the principle on fresh work. One-to-one gives more feedback opportunities; groups give more comparison opportunities. Neither matters unless the next independent performance changes.

7. Is the tutor collecting evidence that could prove the current plan wrong?

A programme becomes stronger when it can revise its own hypothesis. If the tutor believes one-to-one is needed because the learner cannot infer independently, a fresh unaided task may show the real problem is vocabulary. If the group assumes peer discussion improves writing, school drafts may show no transfer. Evidence should be allowed to change the plan.

8. Does the format preserve the learner’s own voice?

Private tutors can unintentionally overwrite voice through constant sentence improvement. Groups can create imitation of fluent peers. Preserve original drafts, ask learners to justify revisions and use fresh tasks. Better writing should remain recognisably the learner’s thinking and language at an appropriately developing level.

9. Is the format sustainable inside the learner’s real week?

Include travel, cost, homework, school deadlines, sleep and independent reading. An ideal format that creates chronic overload is not ideal in practice. A slightly less intensive arrangement that the learner can attend consistently and follow with independent practice may produce stronger long-term learning.

10. Is there evidence of transfer beyond tuition?

Look at school scripts, class writing, timed work, oral responses and teacher comments. The tutor’s own worksheets are necessary but insufficient evidence. A format earns its place when the learner carries improved decisions into settings where the tutor and peers are not available.

11. Is there a reason to keep this format next term?

Do not renew automatically because the arrangement feels familiar. Ask whether the original need still exists. A learner may have outgrown intensive one-to-one and now need peer challenge. A group learner may have developed one specific bottleneck worth repairing individually. Or tutoring may no longer be necessary at the current intensity.

12. Can you describe the exit condition?

The most mature tutoring decision includes an end state. The learner can recognise the task, retrieve the method, make the decision, self-check, recover after difficulty and repeat the performance across fresh work and time. At that point, the format has succeeded precisely because it can become less important.

A practical conversation before enrolment

A parent can ask the tutor: “Our current evidence suggests the main problem is ______. Why does your format help with that problem specifically? What would my child do individually before receiving help? How will you know whether the problem is changing? What fresh task will you use? What would make you change the format or tell us that this level of tuition is no longer needed?”

A serious answer should describe teaching and evidence, not merely class size. It may include uncertainty. That is a strength. The tutor has not yet taught the learner and should be willing to update the plan as better information arrives.

A practical conversation with the learner

Ask the student: “Where do you think English becomes difficult? Do you want more time to ask individual questions, or do you learn from hearing other people explain? When someone gives you an answer quickly, does that help you understand or make you stop thinking? Do you feel comfortable showing unfinished work? What would make you feel that tuition was becoming less necessary?”

The student’s answers are not final verdicts, but they provide useful implementation evidence. A learner who understands the reason for the format is more likely to participate intelligently in it.

The final Secondary 2 standard

World-class Secondary 2 English tuition is not defined by maximum individual attention or maximum peer interaction. It is defined by whether the learner’s real bottleneck is diagnosed accurately, taught clearly, practised deliberately, transferred across variation and eventually controlled without the support that made the repair possible.

One-to-one tutoring is strongest when concentrated attention reveals and repairs something the learner could not otherwise stabilise. Small-group tutoring is strongest when carefully managed peers create useful comparison, communication and variation that deepen judgement. The mature decision is not to defend one format. It is to choose, test, modify and eventually reduce the format according to evidence.