Why a Composition Lesson Should Not Have the Same Structure Every Week
Consistency is useful in tuition—but repeating the same lesson shape regardless of what the student currently needs can become a form of stagnation. The strongest composition programmes keep the instructional logic stable while allowing the proportion of modelling, planning, writing, feedback, repair and independent transfer to change.
Direct answer: the invariant should be identify the learning job → teach or model what is missing → student attempts → inspect the result → repair → reduce support → transfer. What changes from week to week is how much lesson time each stage deserves.
What This Page Owns
The page How a Primary Composition Tuition Class Should Work owns the standard tutorial workflow: what a strong class generally needs to contain.
This page owns a different question:
When should the lesson shape change because the learning problem has changed?
The distinction prevents two opposite failures: a random class with no repeatable teaching logic, and a rigid class that repeats the same recipe even when the student needs something else.
Stable Logic, Variable Proportions
A composition lesson can keep the same deep architecture while looking very different on the surface.
| Stable logic | Variable lesson proportion |
|---|---|
| There is a clear learning target | Target may be planning, sentences, revision, transfer or timing |
| Student must perform | Performance may be one sentence, one paragraph, a plan or a full composition |
| Feedback changes the next attempt | Feedback may happen live, after a draft or during a rewrite lesson |
| Support should eventually reduce | Some lessons may begin highly scaffolded; others almost unsupported |
| Transfer must be tested | Transfer may occur immediately, later or under exam conditions |
Structure is therefore not the same as sameness.
Evidence Base: Writing Requires Different Writer Jobs
The U.S. Institute of Education Sciences writing toolkit separates planning, drafting, evaluating, revising and editing as distinct parts of writing instruction and supports gradual movement toward independent use. The Education Endowment Foundation’s feedback guidance likewise treats feedback as useful when it informs what the learner should do next rather than functioning as commentary alone.
If those writer jobs are different, it follows that a lesson focusing on one job should not always allocate time identically to a lesson focusing on another.
Lesson Mode 1: New Concept or New Writing Process
When a student has never understood a process, the lesson may need more explicit teaching and modelling.
For example, the learning job might be:
- how to interpret a Continuous Writing prompt,
- how to build causal sequence,
- how to decide what belongs in a paragraph,
- or how revision differs from proofreading.
A sensible shape is:
explain → model → guided attempt → student attempt → short feedback → second attempt
Forcing a full composition into the same lesson may create unnecessary load before the process is understood.
Lesson Mode 2: Diagnosis
Sometimes the tutor does not yet know why the composition is weak.
The lesson then needs more observation than instruction.
Possible diagnostic tasks include:
- interpret a new prompt aloud,
- generate three possible routes,
- write one paragraph independently,
- revise an old paragraph,
- or complete a short timed section.
The tutor is looking for the earliest unstable dependency. See The Composition Skill Dependency Map.
A diagnosis lesson should not disguise uncertainty by teaching everything at once.
Lesson Mode 3: Sentence Repair
If the child has ideas and structure but sentence production repeatedly collapses, the lesson may temporarily reduce the full-composition load.
The sequence could become:
identify recurring sentence failure → model repair → controlled sentence practice → paragraph integration → changed paragraph → later full composition
This is not “doing less writing”. It is concentrating effort on the layer that currently prevents useful full writing.
Lesson Mode 4: Planning and Route Construction
A student who writes grammatically but repeatedly produces weak plots may need a planning-heavy lesson.
That lesson can spend more effort on:
- prompt interpretation,
- alternative idea generation,
- selection,
- causal sequence,
- paragraph jobs,
- and where the main development should occur.
The final written product might be only one key scene if the main learning job is planning quality.
Lesson Mode 5: Drafting and Integration
After isolated skills have improved, the lesson needs to recombine them.
This is where full or extended writing earns its place.
The tutor can now observe whether:
- the student remembers the planning process,
- sentence accuracy survives the larger load,
- vocabulary remains precise,
- paragraph development holds,
- and the student can sustain the writing without constant rescue.
Integration lessons reveal whether the separate repairs can report for duty together.
Lesson Mode 6: Feedback and Rewrite
A feedback lesson should not be squeezed into the last few minutes merely because every lesson must follow the same sequence.
If a recent composition contains high-value evidence, the next lesson may need to centre on:
- student rereads the draft,
- tutor identifies one or two highest-value problems,
- student explains the issue,
- student rewrites,
- tutor checks whether the repair worked,
- student applies the same principle to a changed sentence or paragraph.
Feedback pays rent only when it changes student performance.
Lesson Mode 7: Transfer
A transfer lesson should deliberately feel less supported.
The student receives:
- a new prompt,
- a different picture set,
- an unfamiliar theme,
- or the same underlying skill hidden inside a different writing problem.
The tutor should resist unnecessary help. If the student can perform only when the lesson resembles the teaching example, the capability is not yet portable.
A transfer lesson tests what the student owns after the lesson design stops carrying them.
Lesson Mode 8: Timed Exam Integration
Near an examination, the student must integrate capability under realistic timing. That requires a different lesson shape again.
A timed lesson may include:
- bounded prompt analysis,
- self-calibrated planning,
- continuous independent writing,
- protected checking time,
- and post-task analysis of where time and quality were lost.
This lesson should not stop every two minutes for live correction. The point is to observe what survives uninterrupted examination load.
The Adaptive Lesson Matrix
| Current learning job | More lesson weight on | Less lesson weight on |
|---|---|---|
| New concept | Explanation, modelling, guided attempt | Long unsupported writing |
| Diagnosis | Observation and contrasting tasks | Premature correction of everything |
| Sentence repair | Reduced tasks and reintegration | Repeated full essays |
| Planning weakness | Prompt interpretation and route construction | Decorative language |
| Feedback/revision | Student repair and changed reattempt | New content for its own sake |
| Transfer | Unfamiliar task, low scaffolding | Tutor prompting |
| Exam execution | Uninterrupted timed integration and analysis | Frequent mid-task teaching interruptions |
What Should Stay Predictable for Students?
Adaptive lessons should not feel chaotic. Students benefit from predictable norms even when the lesson mode changes.
- There is always a clear learning job.
- The child will eventually have to perform, not only listen.
- Important feedback requires action.
- Previously taught skills can return later.
- Support will reduce when capability improves.
- Unfamiliar transfer will eventually be tested.
The predictable element is the learning logic—not the exact worksheet sequence.
When Repetition Is Useful
Variation does not mean every lesson must be novel.
Repeating a structure is useful when:
- the process itself is still being learned,
- the student needs retrieval and fluency,
- comparable practice is needed to measure improvement,
- or a routine reduces unnecessary task-management load.
The important question is whether repetition is strengthening a process or merely reproducing the same performance environment.
When Routine Becomes Stagnation
- Every lesson uses the same prompt-analysis template even after the student no longer needs it.
- Every student writes a full composition even when one needs sentence repair.
- Feedback is always rushed because “new content” must occupy most of the lesson.
- Strong students receive more of the same work rather than a higher transfer ceiling.
- Weak students remain permanently scaffolded because the programme never changes mode.
- Timed papers appear every week even when the underlying skill is not yet secure.
How a Maximum-Three-Student Tutorial Can Vary Without Fragmenting
In a three-student group, students do not have to perform identical actions every minute for the lesson to remain coherent.
For example:
- one student can revise a paragraph while the tutor conferences with another,
- two students can compare planning routes while the third completes an independent writing section,
- a stronger student can attempt unfamiliar transfer while a Catch Up student works through a guided repair,
- and all three can reconvene to compare one shared principle.
The class stays unified by a common English domain and shared teaching principles, while task size and support can differ.
Evidence That the Lesson Shape Is Correctly Calibrated
- The stated learning target improves during the lesson.
- Students spend meaningful time performing, not only receiving instruction.
- The amount of tutor help matches the current need.
- Feedback produces a visible repair.
- Old skills are retrieved rather than forgotten.
- The student can eventually perform in a changed context.
- Lesson structure changes when the bottleneck changes.
Catch Up, Keep Up and Move Ahead
| Student state | Likely lesson pattern |
|---|---|
| Catch Up | More explicit instruction, smaller tasks, successful repair, gradual reintegration |
| Keep Up | Balanced application, feedback, retrieval and scaffold reduction |
| Move Ahead | Less modelling, more unfamiliar transfer, ambiguity, precision and independent revision |
A Parent Audit for Class Structure
- Can the tutor explain the learning job for this lesson?
- Does the lesson structure change when my child’s bottleneck changes?
- Is full composition used when integration needs testing rather than simply every week by default?
- Is there protected time for revision when feedback is the learning job?
- Are scaffolds eventually reduced?
- Does my child face unfamiliar transfer tasks?
- Is timed practice introduced for execution rather than as a substitute for teaching?
- Does routine make learning efficient without making it static?
- Can different students receive different levels of support without the class becoming chaotic?
- Is the student becoming increasingly capable across changing lesson shapes?
At eduKate Punggol
eduKate Punggol uses maximum-three-student, 1.5-hour tutorials. The tutorial length is fixed; the educational use of that time should not be. A lesson can become planning-heavy, feedback-heavy, transfer-heavy or timed when the student evidence calls for it. What remains consistent is the expectation that the child learns a process, performs it, repairs it and progressively owns more of it.
Continue
- How a Primary Composition Tuition Class Should Work
- Primary Composition Teaching Techniques: Match the Method to the Writing Bottleneck
- How Feedback Should Work in Primary Composition Tuition
- The Composition Skill Dependency Map
Lesson-architecture principle: preserve the logic, change the proportions, and let the student’s current learning job decide what the next 1.5 hours are for.
