The Tutor Handbook · Volume 0194 · Series ID THB-0194
The Tutor Handbook: Complete Series Index
A practice can be useful and unpopular
A tuition programme introduces a new routine.
Learners must attempt one problem privately before discussion. The purpose is to preserve individual thinking before the quickest student reveals the route.
Some learners dislike it.
They say the old discussion felt easier. One parent reports that the child "doesn't like being made to sit there when someone else already knows the answer". One tutor thinks the silent first minute feels awkward. Another tutor values the routine because it reveals who can actually begin independently.
What should the programme conclude?
It would be weak to say, "Students dislike it, so stop."
It would be equally weak to say, "Resistance is normal, so ignore them."
The reactions contain evidence—but not necessarily evidence about whether the practice improves learning.
They may reveal misunderstanding, workload, embarrassment, access barriers, weak explanation of purpose, poor timing, loss of autonomy, a genuine mismatch with some lesson conditions, or the ordinary discomfort of a task that now exposes independent thinking more honestly.
The Implementation-Acceptability Gate asks how a tuition programme interprets tutor, learner and family responses to a new or established practice without turning acceptability into a popularity vote and without dismissing negative reactions as attitude.
Acceptability is evidence about how a practice is experienced, understood and judged by the people expected to live with it.
That matters because an educational practice that nobody can accept may not be implemented, sustained or used honestly.
It does not mean that the most liked practice is the most effective.
Quick answer
Treat acceptability as one implementation outcome among several.
Ask whether the people affected by the practice understand its purpose, regard its demands as proportionate, can use it without unnecessary embarrassment or access barriers, and believe the trade-offs are reasonable enough to participate honestly.
Then keep acceptability separate from other questions.
Effectiveness: does the practice improve the educational outcome it is meant to improve?
Feasibility: can ordinary tutors realistically deliver it?
Fidelity: when it is delivered, are the load-bearing features present?
Reach: do eligible learners and tutors actually receive it?
Sustainability: can it continue over time?
Acceptability: how do the people expected to use or experience it judge its appropriateness, burden and fit?
Negative acceptability data should trigger investigation, not automatic abandonment. Positive acceptability should not be presented as proof of learning.
Look for the mechanism behind the reaction.
If learners dislike the practice because it exposes uncertainty publicly, redesign privacy. If tutors dislike it because it creates duplicate records, remove the duplicate burden. If parents resist because the purpose is unclear, improve explanation. If the practice remains disliked because it deliberately removes answer-giving help, decide whether the discomfort is educationally justified and manageable.
Acceptability changes the implementation route. It does not replace educational judgement.
Why this is not the Learner-Voice Interpretation Gate
The Learner-Voice Interpretation Gate asks how a tutor uses student feedback about difficulty, helpfulness and fit without treating satisfaction as proof of learning.
The Implementation-Acceptability Gate works at the programme-practice level.
It asks whether an adopted routine, tool, reporting system, discussion structure, AI workflow, feedback practice or coaching process is considered appropriate enough by the affected stakeholders to be implemented honestly and sustainably.
The Family-Evidence Integration Gate owns how parent and caregiver observations enter the learner model without replacing direct learning evidence.
Here, family reactions are evidence about the implementation of a programme practice.
The Implementation-Feasibility Gate asks whether the routine fits ordinary operating conditions. A practice can be feasible and still poorly accepted. It can also be highly acceptable and operationally impossible.
Keeping these owners separate is what makes the diagnosis useful.
AERO now names acceptability explicitly
The Australian Education Research Organisation's current Staying on track: Monitoring implementation outcomes, published and updated on 15 September 2026, explicitly treats feasibility and acceptability as implementation outcomes alongside fidelity, reach and sustainability.
It is research-informed school implementation guidance, not a private-tuition validation study.
The value of the distinction is practical. A programme may see non-use and assume tutors lack skill. Acceptability data may reveal that tutors regard the routine as duplicative or learners experience it as public error exposure. Those perceptions do not prove the routine is ineffective, but they change what the implementation problem might be.
The Education Endowment Foundation's A School's Guide to Implementation, third edition published 24 April 2024, similarly emphasises engagement, contextual factors, behaviours and opportunities to reflect and learn during implementation.
The broader eduKateSG owner How Evidence-Informed Decision Making Works makes a compatible distinction: stakeholder voice is evidence about experience, feasibility and implementation conditions, not a referendum on causal effectiveness.
This Tutor Handbook article turns that distinction into a small-group tuition decision.
Popularity is a weak proxy for learning
Learners often like things that feel fluent.
A familiar worksheet can feel productive because answers come quickly. Immediate tutor hints can feel helpful because frustration falls. Re-reading can feel reassuring. A game can feel engaging. A tutor who explains every difficult step can feel exceptionally clear.
None of those reactions is useless.
None proves durable learning.
Likewise, an evidence-informed practice can be temporarily unpopular because it creates desirable difficulty, requires independent retrieval, removes unnecessary prompts or asks the learner to revise work rather than simply receive the corrected answer.
The tutor should therefore keep two evidence streams.
"What was the learner's experience?"
"What changed in the learner's capability?"
The streams can inform each other without being merged.
A practice that improves learning but creates unnecessary humiliation should be redesigned.
A practice that feels wonderful but produces no transfer should not be protected by satisfaction scores.
Resistance can be information about the design
The word "resistance" is often too broad.
A tutor does not use a routine. Why?
They may disagree with the educational rationale.
They may understand it and believe it does not fit the current subject.
They may lack the skill to use it.
They may lack time.
They may be using it in a different form.
They may think the required documentation has no purpose.
They may have seen learners shut down when the routine is used publicly.
They may simply prefer the old way.
Each explanation points to a different response.
"Resistance" becomes useful only after it is decomposed.
The Training Need Gate already protects against assigning training when the real problem is material, workload or system design.
Acceptability adds another cause: the practice may be judged inappropriate or unreasonable by the people carrying it.
That judgement deserves inspection rather than automatic obedience or dismissal.
Composite case: learners hate the "no hints first" rule
The following case is fictional and constructed for teaching.
A programme introduces a short independent-attempt rule. For selected tasks, learners spend ninety seconds making a first move before asking for a hint.
Several learners complain.
They say tuition is supposed to help them, not make them struggle.
Parents repeat the concern.
The programme could abandon the rule to preserve satisfaction. It could also declare the complaints proof that learners are over-dependent and become stricter.
Instead, tutors investigate.
They find three different experiences.
Alicia understands the purpose and finds the pause useful.
Beatrice experiences the ninety seconds as manageable on current-level work but overwhelming on prerequisite gaps.
Ciara thinks she is not allowed to ask any question, even when she does not understand the instruction.
The acceptability problem is partly communication and partly eligibility.
The programme clarifies that the rule applies when the task is understood and the learner has enough knowledge to attempt. Instruction clarification remains available. Active Repair tasks can use a shorter supported start. Tutors explain that the pause is not punishment; it protects a sample of independent thinking before help enters.
The practice survives, but the implementation becomes more humane and precise.
Acceptability data improved the design without turning preference into the learning standard.
Who should be asked?
A programme can create false acceptability by listening only to the easiest voices.
Confident tutors attend meetings. Highly engaged parents respond to surveys. Articulate learners explain concerns clearly. Families under greater time pressure remain silent.
A broad open link does not guarantee representative voice.
For a small tuition centre, formal sampling machinery is unnecessary. The programme can still ask whether it has heard from:
new and experienced tutors;
younger and older learners;
learners who use access supports;
families with different schedules;
groups where the practice works smoothly;
groups where it is often skipped.
The point is not demographic bureaucracy.
The point is to avoid concluding "everyone is fine with it" because only the people closest to the programme were asked.
The broader eduKateSG article on education-policy consultation makes the same system-level point: affected groups, implementers and underrepresented groups can see different parts of implementation. This Tutor Handbook gate applies the principle at programme scale.
Ask about the experience, not just whether they "like" it
"Do you like the new routine?" creates weak data.
A learner can dislike effort and still find the routine fair. A tutor can like a practice but find the workflow impossible. A parent can dislike fewer worksheets but value clearer evidence of independence.
Better questions target mechanisms.
"Do you understand what this routine is for?"
"Which part feels unnecessary or confusing?"
"Does it make it easier or harder to show what you know?"
"Does it add work outside the lesson?"
"When does it fit well, and when does it not?"
"Do you feel able to ask for clarification without being given the answer?"
"Does the routine create embarrassment or pressure that changes how you participate?"
"If we removed one part, which part would you remove and what would be lost?"
These questions turn acceptability into design evidence.
Acceptability is not consent to lower the target
Learner agency matters.
So do educational standards.
A learner may prefer not to write extended responses. A parent may prefer every lesson to focus on the next school test. A tutor may prefer a familiar explanation style over a demanding diagnostic process.
Those preferences deserve understanding.
They do not automatically set the route.
The Priority Arbitration Gate already protects against allowing the loudest stakeholder to determine learning priorities.
The acceptability gate uses stakeholder experience to improve implementation while preserving legitimate curriculum, access, safety and independence goals.
A strong programme can say:
"We understand why this feels harder."
"We have checked that the extra difficulty belongs to the learning goal."
"We will change the unnecessary part."
"We will keep the part that protects independent thinking."
That is engagement without surrendering professional responsibility.
Tutor acceptability can reveal hidden professional cost
Tutors are implementers.
They often see burdens before leaders do.
A new formative-assessment routine may require duplicate data entry. A new feedback protocol may create excellent learner action but need more preparation than the timetable protects. A discussion structure may look simple in a document but require advanced facilitation skill. An AI tool may promise time savings while tutors spend longer verifying the output.
If tutors report that a practice is unacceptable, leaders should ask what the judgement is based on.
Time?
Professional autonomy?
Lack of evidence?
Poor fit with subject?
Learner response?
Unclear ownership?
A duplicated process?
Fear that observation is becoming appraisal?
Some concerns are misconceptions. Some are implementation defects.
Treating every concern as reluctance can silence the people with the best view of the workflow.
Composite case: the progress dashboard everyone completes and nobody trusts
This case is fictional.
A programme introduces a dashboard that displays learner status after each lesson.
Managers like the visibility.
Tutors dislike the system. They say it reduces nuanced learner evidence to a green, amber or red label. They still complete it because it is required.
Parents appreciate the simple display but begin asking why a child moved from green to amber after one difficult task.
The programme initially interprets tutor dislike as resistance to accountability.
A closer acceptability review reveals a construct problem. The dashboard is asking tutors to make a broad status judgement from evidence that is often too narrow. Tutors distrust it because they know the colour overstates precision.
The solution is not an engagement campaign.
The programme redesigns the dashboard. It records the current target, latest evidence condition and next decision instead of one broad colour. Parent communication explains uncertainty more honestly.
Acceptability improved because the product became more educationally valid.
Sometimes negative reaction is a quality signal.
Learner discomfort can be productive, harmful or merely unfamiliar
Not all discomfort has the same meaning.
Productive effort: the learner has enough knowledge to attempt and must retrieve, discriminate or persist.
Unnecessary difficulty: the task adds confusing format, inaccessible language or irrelevant motor demand.
Social threat: the learner expects public embarrassment, comparison or ridicule.
Loss of control: the learner does not understand why support has changed or when help remains available.
Novelty: the routine is unfamiliar but becomes easier after repetition.
Mismatch: the practice truly does not fit the learner or task.
A tutor should not interpret all discomfort as productive struggle.
Nor should every report of discomfort end the practice.
The distinction requires direct observation and follow-up.
The existing Behaviour–Learning Differential similarly treats refusal and off-task behaviour as evidence to investigate rather than a ready-made motivation diagnosis.
Acceptability is one more lens on the same professional humility.
Family acceptability often concerns trade-offs outside the lesson
Parents see implementation costs the tutor may not.
A new homework routine may require printing. A digital platform may need a device at a particular time. A parent-reporting system may send too many notifications. A revision practice may collide with school homework. A new Saturday schedule may create transport problems.
These conditions can affect reach and sustainability.
Family reaction is therefore not merely customer satisfaction.
It can reveal hidden transaction costs.
At the same time, families may prefer high volumes of visible worksheets because output is easy to see. A lower-volume diagnostic lesson can feel less valuable despite being better targeted.
The programme should explain what the practice is doing while listening carefully for genuine access and burden problems.
Communication and design are both legitimate responses.
Ask about acceptability after the practice is understood
Early reactions can be misleading.
People often judge the surface before experiencing the routine enough to understand its purpose.
A tutor may dislike a structured questioning protocol during training and value it after seeing clearer learner evidence.
A learner may enjoy a new game initially and later find it repetitive.
A parent may reject a smaller homework dose until they see that the tasks are more targeted.
Acceptability should therefore be measured at more than one point when the decision matters.
Initial acceptability tells the programme about launch conditions.
Later acceptability tells the programme how the practice feels after novelty and learning cost have changed.
The Implementation-Sustainability Gate uses that later signal as part of long-run viability.
A practice can be acceptable because expectations are low
Positive reactions can hide weak challenge.
Learners love a routine because it makes every answer feel successful.
Tutors love it because it is easy to run.
Parents love it because the worksheet returns full of ticks.
If the practice avoids difficult retrieval, removes independent decision-making or provides strong answer cues, high acceptability can coexist with weak educational value.
This is why acceptability must remain one stream.
The programme should ask whether the practice is also preserving the target mechanism and producing the intended learner opportunity.
Popularity is not a defence against weak evidence.
A practice can be unacceptable because implementation is bad, not because the idea is bad
Suppose learners dislike peer explanation.
Observation shows that one fluent learner is repeatedly asked to teach the others, while listeners have no active role. The programme concludes that peer explanation is unpopular.
But the Peer-Explanation Boundary would identify the implementation defect: one learner has become an unofficial assistant and peers are borrowing understanding.
The acceptability signal is real.
The target of repair is implementation, not necessarily the underlying idea of learner explanation.
This is a general principle.
Before abandoning a disliked practice, check whether people are reacting to the intended practice or to a distorted version.
Fidelity belongs before verdict.
Acceptability can differ across stakeholders for legitimate reasons
A tutor may value a routine that a learner dislikes.
A learner may value a routine that a parent cannot see the purpose of.
A parent may value frequent progress messages that tutors experience as distracting from preparation.
There is no requirement that all groups have identical preferences.
The programme should identify the underlying interest.
Tutor: instructional clarity.
Learner: dignity and manageable challenge.
Parent: confidence and visibility.
Programme lead: consistency and evidence.
Then ask whether one design can respect several interests or whether a real trade-off exists.
A trade-off should be named rather than hidden.
For example, reducing parent-report frequency may free tutor preparation time while preserving one higher-quality monthly review. The family may prefer weekly messages; the programme may decide the educational trade-off favours fewer, more useful updates.
Acceptability informs the decision. It does not automate it.
Cultural and linguistic context can change acceptability
Participation norms differ.
Some learners are comfortable challenging peers publicly. Others need more preparation before public disagreement. Some families expect tutors to provide direct answers quickly. Others value independence and learner struggle. Some language choices sound warm in one context and rude in another.
The Cultural-Safety Adaptation Gate protects against stereotyping while adapting communication and participation conditions.
The acceptability gate uses the same discipline.
Do not infer a learner's preference from cultural category.
Ask and observe.
Then adapt the surface where the active educational ingredient can remain intact.
For example, a public challenge can become a written private comparison before discussion. The thinking demand remains. The social route changes.
Privacy and dignity are part of acceptability
A programme can create a technically effective routine that learners experience as exposing.
Public error boards, visible rank displays, forced disclosure of marks, recorded coaching clips or AI transcripts may create privacy concerns.
The fact that the data could improve teaching does not erase dignity.
The Professional Relationship Boundary and Learner-Data Retention Gate protect related professional boundaries.
Acceptability asks another question: can the educational job be achieved with a less exposing design?
Often it can.
A private first response can replace public guessing. An anonymised example can replace named error comparison. A coach can review a bounded clip rather than record every lesson.
The best implementation design earns trust rather than spending it unnecessarily.
AI creates a special acceptability problem
Learner-facing and tutor-facing AI can be rejected for different reasons.
A learner may worry that AI use means the tutor is not really teaching.
A tutor may worry that AI-generated materials will be unreliable or that professional judgement is being replaced.
A parent may worry about privacy, dependency or excessive screen time.
Some concerns are factual and can be addressed through clear boundaries. Some reveal genuine design risks.
The AI Material Verification Gate and Human-Supported AI Engagement Gate own accuracy and learner-facing use.
Acceptability should not be used to market people past legitimate concerns.
Explain what data are used, what the human checks, what the tool is allowed to do and what it is not allowed to decide. Then ask whether the remaining concern is about values, privacy, access or educational fit.
A trustworthy implementation can survive questions.
Composite case: parents want more homework than the programme intends
This case is fictional.
A programme reduces routine homework volume and replaces it with shorter targeted retrieval and one fresh transfer task.
Several parents object.
They equate the thinner packet with less teaching.
Learners are generally positive because the work is shorter.
The programme could give more pages to restore acceptability.
Instead, tutors explain the learning job: the new tasks are selected to expose retrieval and transfer, not to maximise page count. They show parents how one fresh task is used at the next lesson to update the route.
After a month, some parents remain unconvinced.
The programme keeps the lower volume because the rationale, workload and learner evidence support it. It also improves visibility by including a brief note on the task purpose.
Acceptability improved somewhat, but not universally.
That is acceptable.
Implementation does not require unanimity.
Use disagreement to identify the contested value
When stakeholders disagree, ask what value is underneath the disagreement.
Efficiency?
Independence?
Challenge?
Dignity?
Visibility?
Fairness?
Consistency?
Choice?
Privacy?
Examination alignment?
Once the contested value is visible, the programme can reason more clearly.
A learner who dislikes cold calling may be objecting to public uncertainty, not to being asked to retrieve.
A tutor who dislikes scripted routines may be protecting professional adaptation, not rejecting consistency.
A parent who wants immediate correction may be protecting confidence, not opposing productive struggle.
The surface preference can often be redesigned while respecting the underlying concern.
Sometimes the values genuinely conflict. Then the programme should make the trade-off explicit and proportionate.
Acceptability should be sampled where implementation is weakest
If a routine has low reach in one set of groups, ask those groups.
If one tutor keeps modifying the practice, ask why.
If families using access supports opt out more often, investigate their experience.
This is more useful than collecting generic satisfaction from the whole programme.
The Implementation-Reach Gate can identify where the practice is disappearing.
Acceptability inquiry can then test whether perceived fit, burden or dignity is part of the cause.
Implementation outcomes should talk to one another.
Do not measure acceptability so heavily that it becomes unacceptable
Surveys create burden too.
A programme does not need a questionnaire after every lesson.
For most tutoring practices, a small set of mechanisms is enough: brief tutor debriefs during rollout, periodic learner questions, parent feedback already arising through normal communication, and targeted inquiry where reach or fidelity is weak.
The programme should collect acceptability data when a decision depends on it.
If nobody will change the practice, communication, support or eligibility regardless of the answer, the survey is probably ceremonial.
The Evidence-Capture Burden Gate applies to implementation evidence too.
A practical acceptability review
Choose one practice with a real implementation decision ahead.
State its purpose and active ingredient.
Identify the groups whose experience matters: tutors, learners, families or other implementers.
Ask about specific mechanisms rather than general liking.
What is clear?
What is burdensome?
What feels inappropriate or exposing?
What creates hidden work?
Where does the practice fit badly?
What do people think would be lost if it were removed?
Then triangulate the answers with feasibility, fidelity, reach and learner evidence.
Classify the issue.
Misunderstanding: purpose or boundary is unclear.
Surface burden: a non-essential part creates avoidable friction.
Access problem: some users cannot participate fairly.
Skill problem: the routine feels poor because implementation competence is weak.
Value conflict: stakeholders disagree about a legitimate trade-off.
Core mismatch: the active practice itself does not fit the context or intended users.
The response differs by category.
Communicate.
Redesign.
Provide support.
Narrow eligibility.
Protect the core despite discomfort.
Or stop the practice.
Acceptability becomes useful when it changes a real decision.
Failure modes
The popularity-vote failure. The most liked practice is treated as the best educational practice.
The resistance-dismissal failure. Negative stakeholder reaction is labelled attitude before workload, dignity, access or fit is investigated.
The satisfaction-equals-learning failure. Learners saying a lesson was helpful is treated as direct evidence of improved capability.
The one-voice failure. Highly engaged parents or confident tutors stand in for people whose experience differs.
The launch-only failure. Early reactions are assumed to remain valid after novelty, skill and workload change.
The distorted-practice failure. Stakeholders reject a poorly implemented version, and the underlying practice is blamed without a fidelity check.
The unanimity failure. The programme believes implementation cannot proceed until every stakeholder prefers the change.
The communication-only failure. Real operational burdens are treated as public-relations problems.
The low-expectation acceptability failure. An easy, highly liked routine is protected even though it removes productive learning demand.
The survey-burden failure. Acceptability monitoring creates more friction than the practice being monitored.
Acceptability thresholds should depend on the seriousness of the concern
A programme should not require the same response to every negative reaction.
Minor irritation with a new layout is different from a privacy concern.
A preference for the old worksheet is different from a learner reporting repeated public embarrassment.
A tutor finding a routine initially awkward is different from the routine requiring unsustainable unpaid work.
The programme can use a simple seriousness test.
Does the concern involve safety, dignity, access, privacy or professional boundary?
Does it prevent meaningful participation?
Does it create a recurring burden large enough to threaten feasibility?
Does it indicate that the active ingredient has been misunderstood?
Is the concern likely to fade with ordinary familiarity?
High-consequence concerns deserve immediate investigation even if only one person reports them.
Low-consequence preference differences can often be monitored without redesign.
This avoids two extremes: majority rule and leadership deafness.
Minority reactions can reveal a majority-blind design
An implementation can be highly acceptable on average and still exclude a small group systematically.
Ninety per cent of learners like a digital annotation tool. The remaining ten per cent use an assistive technology with which the tool performs badly.
Most parents accept an online reporting system. A small number cannot access it reliably.
Most tutors find a timed discussion routine manageable. One subject area repeatedly requires different pacing.
An average satisfaction figure can hide these patterned problems.
Acceptability should therefore be inspected alongside reach and access.
The right question is not only, "How many people approve?"
It is also, "Who is having a different experience, and is the difference caused by a legitimate educational condition or by a removable barrier?"
A small subgroup can identify a design flaw that the majority never encounters.
A change in acceptability after redesign is an implementation receipt
If the programme changes a surface feature because stakeholders identified unnecessary friction, check whether the reaction changes.
Suppose tutors object to duplicate note entry. The programme merges two forms. Do tutors now use the practice more consistently?
Suppose learners dislike a discussion routine because they must answer publicly without thinking time. The programme adds private preparation. Does participation become more honest and less avoidant?
Suppose families find a new progress report confusing. The programme changes the explanation but not the evidence standard. Are follow-up questions clearer?
The change in acceptability does not prove improved learning.
It does help test whether the programme correctly identified the implementation friction.
Acceptability can therefore participate in a bounded improvement loop: hear the concern, classify the mechanism, change the relevant surface, and see whether the experience changes without damaging the educational job.
Evidence boundaries and sources
AERO's Staying on track: Monitoring implementation outcomes, published and updated 15 September 2026, explicitly treats acceptability as an implementation outcome. It is research-informed school guidance. It does not provide a universal acceptability threshold for private tutoring.
EEF's A School's Guide to Implementation, third edition published 24 April 2024, emphasises engaging people, attending to context and creating opportunities to reflect and learn throughout implementation. It is education implementation guidance, not a tutoring satisfaction model.
The broader eduKateSG owner How Evidence-Informed Decision Making Works separates stakeholder voice as evidence about experience and implementation conditions from research evidence about causal effectiveness. That owner remains generic; this Tutor Handbook gate applies the distinction to tutoring programme practices.
No claim is made that a particular satisfaction score, approval percentage or survey response predicts learning. Acceptability should be interpreted alongside feasibility, fidelity, reach, sustainability and learner evidence.
The end state
A good tuition programme should be neither governed by applause nor deaf to the people it serves.
Tutors need room to report when a routine creates impossible work or conflicts with subject reality.
Learners need room to say when a practice is confusing, exposing or inaccessible.
Families need room to identify costs and concerns the lesson itself does not reveal.
Programme leaders still need to protect educational purpose, evidence quality, curriculum standards and learner independence.
That is the Implementation-Acceptability Gate.
Listen to the reaction.
Do not worship it.
Do not dismiss it.
Find out what the reaction is evidence of, then change the part of the system that the evidence actually justifies.
