The Tutor Handbook Vol No.0169 | Behaviour

The Tutor Handbook · Volume 0169 · Series ID THB-0169

Series route: The Tutor Handbook — Complete Series Index.

The tutor asks a question. The learner looks away, jokes with a classmate and says, “I don’t know.” The tutor repeats the question. The learner shrugs. A minute later the worksheet is still blank.

It is tempting to name the problem quickly: lazy, unmotivated, oppositional, careless, not serious, bad attitude.

Those labels feel explanatory because they convert an uncomfortable interaction into a stable property of the learner. They also close the investigation too early.

The same visible behaviour can emerge from very different educational conditions. The task may be unclear. The learner may not know the first step. The work may be far below the learner’s level and offer no meaningful challenge. A peer may be carrying the discussion. The tutor may be prompting so rapidly that refusal is the only remaining way to control the pace. The learner may have been told to attend tuition but may not yet understand what the work is for. A routine may be inconsistent, so every transition becomes a negotiation. Or the behaviour may have little to do with the academic task at all.

The Behaviour–Learning Differential is the tutor’s discipline of describing the observable behaviour first, keeping several educational causes alive, changing one relevant condition at a time, and responding with the smallest proportionate teaching or participation move before turning behaviour into a judgement about the learner.

This is not a clinical diagnostic framework. Tutors should not infer ADHD, anxiety, learning disorders, trauma or another medical or psychological condition from ordinary tuition behaviour. Nor is it a licence to tolerate unsafe conduct. The job is educational: before deciding that a learner “will not engage”, check whether the tutoring environment is making useful engagement possible.

Quick Answer

When a learner is off-task, refusing, distracting others or repeatedly disengaging, do not begin with punishment or motivation theory. Begin with a precise description of what happened and the instructional condition in which it happened.

Then test the most plausible educational explanations: unclear expectations, inaccessible or poorly pitched work, excessive help, insufficient help, low opportunity to respond, group dynamics, transitions, lack of bounded choice, or a routine that changes too often to become predictable.

Current school-based evidence supports several low-intensity practices that can improve participation: clear expectations, reminders, positive acknowledgement of expected behaviour, instructional choice, frequent opportunities to respond, and student self-monitoring. AERO guidance similarly emphasises rules and routines that make participation predictable and support engagement.

These sources are not direct evidence for every private three-student tutorial. They provide useful educational principles, not a validated tuition behaviour protocol. Apply them proportionately, keep the learner’s academic target central and preserve the possibility that the behaviour is telling you something about the task.

1. What This Volume Owns

The Attendance Differential owns missed sessions and the need to separate logistics, load, relationship and avoidance before calling absence a motivation problem. The Relationship Repair owns rebuilding working trust after friction. The Relationship Quality Check asks whether trust and rapport are supporting learning. The Individual Accountability Gate protects each learner’s own thinking inside group collaboration.

This volume owns the live moment inside tuition when behaviour interferes with participation and the tutor must decide what the behaviour justifies changing.

The distinctive problem is causal ambiguity. “Refusal” is an observation only after it has been defined carefully. It does not tell us why the refusal occurred. “Off-task” tells us that attention is not currently on the intended activity; it does not identify the mechanism. “Disruptive” tells us that another learner’s work may be affected; it does not explain the source.

The professional task is therefore not to find a nicer label. It is to make the behaviour educationally interpretable enough to choose the next move.

2. Describe Before You Explain

Compare two notes.

“Denise was unmotivated and disruptive.”

“During the ten-minute independent algebra set, Denise began three side conversations, wrote no first step on questions 2–4, and re-entered the task after the tutor reduced the set to one question and asked her to identify only the method.”

The second note is longer, but it is much more useful. It separates observation from inference and contains a possible clue: the learner re-engaged when the task was narrowed to a first decision.

Good behaviour notes need not become bureaucratic. A short structure is enough: condition → observable action → tutor response → immediate return.

This prevents labels from hardening before the tutor has tested what can change.

3. Keep Several Educational Causes Alive

The first hypothesis should rarely be “motivation”. Motivation is real, but it is broad and difficult to act on when used as a catch-all.

For ordinary tutoring, several neighbouring explanations are often more discriminating:

Expectation problem: the learner does not know what behaviour or output is expected in this part of the lesson.
Entry problem: the learner cannot identify a valid first step and escapes the task instead.
Challenge problem: the work is so easy, repetitive or disconnected from the learner’s current frontier that engagement collapses.
Support problem: the tutor is giving too much help, too little help, or help at the wrong moment.
Opportunity problem: the learner receives too few genuine turns to think, answer or make decisions.
Group problem: peer dynamics make joking, copying, silence or dominance more rewarding than the intended task.
Transition problem: the learner loses the route when the lesson moves between explanation, practice, correction or independent work.
Agency problem: the learner has no meaningful choice about process, sequence or participation and uses refusal to recover control.

None of these should be assumed automatically. They are hypotheses that can be tested through small changes.

4. Clear Expectations Are an Instructional Tool

The 2024 What Works Clearinghouse practice guide on teacher-delivered behavioural interventions for K–5 pupils gives strong evidence to co-establish, model and teach clear behavioural expectations. AERO’s current rules-and-routines guide similarly argues that predictable rules and routines help students understand what is expected and support focus and engagement.

A private tutorial is smaller and more flexible than a classroom. It does not need a wall of behaviour rules. But ambiguity can still create avoidable friction.

“Do the worksheet” may hide several decisions. Are learners working silently? Can they ask a peer? Should they show the first step before continuing? What happens when they are stuck? Is discussion allowed after an attempt? When is the set complete?

A clear routine can reduce negotiation: “First two minutes: everyone attempts question 1 privately. If stuck after a genuine first attempt, underline where you stopped. Then we compare methods.” The rule is not about obedience. It makes the route visible.

When behaviour improves after expectations become concrete, the tutor has learned something important: at least part of the problem was environmental ambiguity, not simply lack of motivation.

5. Check Whether the Task Has a Viable Entry Point

A learner can refuse a task because the task feels impossible before it begins.

This is especially common when adults see the work as revision. “You have done this before” can make the learner’s inability to start feel like a moral failure rather than a diagnostic clue.

Test the entry point. Ask for the smallest decision that belongs to the target: identify the known quantities, choose the representation, state what the question is asking, locate the evidence sentence, name the relationship, or write the first transformation.

If the learner can perform the first decision once the task is narrowed, the tutor should investigate whether whole-task complexity was masking available knowledge. If the first decision also collapses, a prerequisite or representation problem becomes more plausible.

Do not reward every refusal by doing the first step for the learner. The diagnostic move is to reduce ambiguity while preserving learner action.

6. Check Whether the Work Is Too Easy to Require Thought

Behaviour problems are not always created by difficulty. Repetition without intellectual demand can also produce disengagement.

Suppose Alicia completes ten routine equations accurately and then begins talking through the next ten. The tutor can interpret this as poor discipline and insist on silence. Or the tutor can test whether the practice has stopped paying rent.

One mixed question that removes the method label may immediately restore attention because it requires a decision again. If so, the issue was not simply that Alicia dislikes work. The task had become mechanically predictable.

The Challenge Readiness Gate remains the owner for when harder work is appropriate. This volume adds one point: behaviour can sometimes be a signal that the current task no longer requires enough learner thinking.

7. Opportunity to Respond Is Part of Engagement

The WWC behavioural practice guide gives moderate evidence for providing frequent and varied opportunities for students to respond and engage in activities. OECD work on classroom interaction likewise emphasises questioning, responding and equitable participation.

In a three-learner tutorial, a quiet student can remain physically present while another learner supplies most of the thinking. The tutor may then complain that the quiet learner is passive.

Count opportunities before judging participation. Did the learner receive a question that genuinely belonged to them? Was enough time protected before another student answered? Did every discussion begin after a private first attempt, or could the fastest learner continuously pre-empt the group?

Increasing opportunity does not mean calling on the learner every thirty seconds. It means designing enough consequential turns that participation is structurally possible.

8. Bounded Choice Can Increase Agency Without Surrendering the Learning Goal

The same WWC guide gives moderate evidence for instructional choice as a way to increase engagement and agency in elementary classrooms.

A tutor can use choice without turning every lesson into negotiation.

“We need to practise inference today. Do you want to start with the shorter passage or the visual text?” “You need to demonstrate three examples. Choose which one you explain aloud.” “We have to repair these two algebra errors. Which one do you want to diagnose first?”

The target remains fixed. The learner gains meaningful control over route, order, representation or response form.

If bounded choice restores participation, the tutor has evidence that agency was part of the behaviour pattern. That does not prove all future work should be chosen by the learner. It shows that a total absence of control was creating friction.

9. Do Not Let Choice Become Avoidance by Design

Choice has a boundary.

If the learner can always choose the easiest task, skip the weak area or select only preferred response forms, the programme may increase satisfaction while narrowing evidence and capability.

The Task-Choice Evidence Gate already warns that learner-selected practice can become a comfortable set mistaken for mastery.

Use choice around a non-negotiable learning job. “You need to show the reasoning; choose whether to write or explain first.” “Both weak links need repair; choose the order.” “The mixed set is fixed; choose which question you review with me.”

Agency should help the learner enter the work, not remove every encounter with difficulty.

10. Positive Acknowledgement Should Name the Productive Behaviour

The WWC guide gives strong evidence for acknowledging expected behaviour through positive attention, praise and rewards in K–5 classroom contexts. The tutoring translation should be specific and modest.

“Good job” is pleasant but vague. “You stayed with the question after the first method failed and tried a second representation” tells the learner what productive behaviour mattered. “You waited for your own attempt before checking Beatrice’s answer” reinforces individual accountability. “You told me exactly where the task became unclear instead of stopping the whole problem” reinforces useful help-seeking.

The aim is not constant praise or a reward economy. It is to make productive participation visible enough that the learner can reproduce it.

Praise should never be used to pretend weak academic work is strong. Behaviour acknowledgement and learning feedback can coexist: “Your method is not correct yet, but the way you kept the variables separate made the error easy to diagnose.”

11. Constructed Case: Alicia Keeps Talking During Routine Practice

This is a fictional teaching case.

Alicia begins side conversations every time the group enters routine algebra practice. The tutor initially interprets the behaviour as loss of seriousness.

The tutor changes one condition. Instead of another page of labelled equations, Alicia receives four mixed questions requiring method selection. She becomes quiet and works immediately.

The conclusion is not that easy work always causes disruption. It is that, in this episode, the routine set no longer demanded enough thought to sustain Alicia’s engagement. The programme moves routine fluency work to a short independent block and uses live tuition for method selection and explanation.

The behaviour changed because the task changed. That is much more actionable than calling Alicia “chatty”.

12. Constructed Case: Beatrice Says “I Don’t Know” Before Reading

Beatrice repeatedly says “I don’t know” the moment a long comprehension question appears. She has not yet read the relevant paragraph.

The tutor could insist, “Try harder.” Instead, they test the entry point: “Do not answer yet. Underline the noun the pronoun refers to.” Beatrice does so correctly. Next: “Find the sentence that changes the character’s decision.” Again correct.

The refusal appears at whole-question scale, not at every component skill. The tutor restructures the next few tasks so Beatrice must identify the evidence before composing the answer, then gradually removes the substeps.

Later, a full fresh question is presented without decomposition. Beatrice begins independently.

The behaviour was not ignored. It was used to locate the point at which the task stopped being operable.

13. Constructed Case: Ciara Refuses to Explain in Front of the Group

Ciara completes Science questions privately but refuses when asked to explain first in front of two peers. The tutor has been interpreting this as weak confidence.

A different design is tested. All three learners write one sentence privately before anyone speaks. The tutor then asks Ciara to read her own sentence and add one causal link. She participates.

The explanation demand remains. The public improvisation demand has changed.

The tutor later increases spontaneity gradually because oral reasoning is part of the goal. But the initial success shows that the earlier refusal did not mean Ciara lacked the scientific idea. The participation format was carrying an extra demand.

14. Constructed Case: Denise Derails Every Transition

Denise works well once she starts, but every transition produces five minutes of argument: from review to new material, from explanation to practice, from practice to correction.

The tutor notices that transitions are improvised. Sometimes practice lasts ten minutes, sometimes thirty. Sometimes corrections happen immediately; sometimes they wait until the end. Denise cannot predict when one demand will finish or what comes next.

A simple visible sequence is introduced: review, new example, private attempt, group compare, independent return. The content still adapts, but the transition cues stay stable.

Arguments fall sharply. The tutor has not “fixed behaviour” through control. They have reduced uncertainty in the lesson architecture.

The Session Structure Gate owns the broader question of predictable lesson rhythm. This case shows how a structure problem can present as conduct.

15. Constructed Case: Emily Distracts the Group When the Tutor Helps Someone Else

Emily is productive when the tutor is speaking directly with her. When the tutor turns to Alicia, Emily starts a separate conversation with Ciara.

The behaviour may look attention-seeking. The orchestration audit reveals a simpler issue: Emily has no meaningful task during the tutor’s one-to-one branch. She finishes the assigned question in ninety seconds and then waits.

The tutor creates a branch routine: after the core item, each learner has one extension, one self-check and one “mark where you are stuck” instruction. Emily now has a next move when tutor attention is elsewhere.

The behaviour improves because idle time becomes purposeful work. The lesson did not need a stronger sanction. It needed better group orchestration.

16. Do Not Let the Fastest Learner Define “Good Behaviour”

Small groups create social comparison.

A fast learner who sits quietly and answers immediately can become the informal standard. A slower learner who needs to verbalise, sketch or ask clarifying questions may look less compliant even when their learning process is productive.

The Peer Baseline Trap warns against making the fastest student the standard for everyone else. Apply that principle to conduct as well as marks.

Expected behaviour should serve the learning job: listen when another learner’s explanation matters, make a genuine attempt, avoid preventing peers from working, use help appropriately, and return to the task after feedback. It should not require every learner to display the same temperament.

17. When the Tutor Is Part of the Behaviour Pattern

Professional humility requires one uncomfortable question: what does the tutor do immediately before the behaviour appears?

Does refusal follow long explanations with no learner turns? Does joking begin when one pupil waits while the tutor works privately with another? Does shutdown appear after public correction? Does argument emerge only when the tutor changes criteria mid-task? Does the learner repeatedly ask for clarification and receive “you should know this already”?

This is not about blaming the tutor. Interaction is a system. If changing tutor behaviour reliably changes learner participation, the system has revealed a lever.

A tutor coach can use this evidence constructively. Instead of “manage Denise better”, the coaching target can become “make transition expectations explicit before the independent block” or “protect three seconds of wait time before rephrasing”. Specific teaching moves are improvable; personality judgements are not.

18. Use Consequences Carefully

A tutorial still needs boundaries. A learner should not be permitted to prevent peers from learning, damage materials, threaten others or turn every task into an unlimited negotiation.

But consequences should be proportionate, predictable and connected to the learning environment. A brief reset, separated work space, temporary change in group interaction or parent contact may be appropriate depending on context and programme policy.

The tutor should not improvise humiliating punishments, public shaming or psychological explanations. Nor should the tutor use academic help as a reward that is withdrawn when behaviour is difficult.

If conduct creates a safety issue or moves beyond the tutor’s educational role, follow the programme’s established safety and escalation procedures. The Scope Boundary remains the owner for what must leave tuition.

19. The Behaviour–Learning Differential Card

Condition: what task, group and lesson phase was active?
Observable behaviour: what did the learner actually do or not do?
Impact: did it block the learner, peers or the tutor’s ability to teach?
Expectation: was the required participation clear and previously taught?
Entry point: could the learner identify a valid first step?
Challenge: was the task meaningfully demanding without being inaccessible?
Support: too much, too little or badly timed?
Opportunity: did the learner have enough genuine turns to think and respond?
Agency: was there any bounded choice over route, order or response form?
Group condition: did peer dynamics change the behaviour?
Small change: what one condition can be altered without changing the target?
Return: what happened to participation and learning after the change?

20. Parent Communication: Report the Pattern, Not a Character Verdict

Parents may ask, “Was she behaving?” That question can invite a yes/no judgement that teaches the family little.

A more useful report might be: “During independent practice, she left three questions blank and started talking to the group. When I reduced the task to choosing the method before calculation, she re-engaged and completed the next two. I am treating this as a task-entry issue first, not as evidence that she does not care. I will check whether the same pattern appears on a fresh set next lesson.”

This language preserves accountability. The learner did leave the work. The tutor is not excusing it. But the family receives a mechanism and a next check rather than a label.

21. Learner Communication: Give Behaviour a Repair Route

“Stop being disruptive” tells a learner what not to do. It does not necessarily tell them how to re-enter the work.

A repair route is concrete: “If you are stuck for more than a minute, mark the first unclear step and move to the next question.” “If you finish before I return, take the extension card.” “If you disagree with the method, show one alternative rather than refusing the whole task.” “If you need a break from speaking, write the first sentence, then we will return to oral explanation.”

The rule should preserve learner agency while keeping the learning target intact.

Over time, the learner should need fewer adult reminders because they can recognise the condition and choose the productive return themselves.

22. Self-Monitoring Can Become a Learner Skill

The WWC guide gives moderate evidence for teaching students to monitor and reflect on their own behaviour. In a tutoring context, this should remain simple and connected to learning.

A learner might track whether they made a first attempt before asking for help, returned after correction, waited for a peer’s explanation, or used the agreed stuck routine. These are observable participation behaviours, not personality ratings.

Avoid pseudo-precision. “Engagement: 7.4/10” may look scientific while hiding arbitrary judgement. One or two concrete behaviours are more useful.

The goal is for the learner to become capable of restarting themselves, not to become dependent on a behaviour scorecard.

23. Research Foundation and Its Limits

The strongest direct source used here is the What Works Clearinghouse 2024 practice guide on teacher-delivered behavioural interventions in grades K–5. Its recommendations are based on systematic evidence review and expert interpretation. The guide supports clear expectations, reminders and positive acknowledgement with strong evidence, and instructional choice, opportunities to respond and self-monitoring with moderate evidence.

AERO’s rules-and-routines practice guide, current to 2025, provides research-informed guidance that predictable expectations and routines can support focus and engagement in primary and secondary classrooms. OECD work on high-quality teaching also emphasises the complexity of equitable classroom interaction and student participation.

These are school and classroom sources. A private three-student tutorial differs in scale, relationship, age mix, stakes and flexibility. The recommendations should therefore be used as principles to test, not as proof that one behaviour strategy will produce a particular outcome in tuition.

Most importantly, educational behaviour guidance does not diagnose medical or psychological conditions. If a concern appears broader than tuition, follow appropriate family, school or professional channels rather than stretching tutor observation beyond its authority.

24. Common Failure Modes

Motivation as the first diagnosis: using a broad label before testing the task and support conditions.
Compliance equals learning: rewarding quiet completion even when the learner is copying or mechanically following.
Every refusal gets rescued: teaching the answer whenever the learner resists, which can make refusal an efficient route to more help.
Every refusal gets punished: escalating consequences before checking whether the task is operable.
Choice becomes escape: allowing the learner to choose away every weak area.
Fast learner as conduct standard: treating one temperament as the model of “good behaviour”.
Idle-time blindness: criticising side conversations while leaving learners without a next task.
Clinical inference: turning ordinary tutoring behaviour into a diagnosis the tutor is not qualified to make.
Parent label transmission: reporting “lazy” or “defiant” instead of the observed pattern and tested conditions.

25. The Thirty-Second Behaviour–Learning Differential

What exactly did the learner do, what task and group condition immediately preceded it, what educational explanations remain plausible, and what is the smallest change that would test one explanation while preserving the learning target?

That question does not eliminate the need for boundaries. It makes the boundary more intelligent.

26. The Independence Direction

A mature learner gradually learns to notice their own participation patterns without turning them into identity statements.

“I start talking when I finish before everyone else; I need a next task.” “I say I don’t know when I cannot see the first step; I should mark the part I do understand.” “I avoid explaining aloud if I have had no preparation time; I can write one sentence first.” “I argue at transitions because I do not know what happens next; I can check the lesson sequence.”

This is more powerful than “I have bad motivation”. It turns behaviour into a solvable interface between learner, task and environment.

The tutor’s long-term success is not perfect obedience. It is a learner who can enter, persist, seek help, recover and protect other people’s learning with less external management.

Evidence and Connected Reading

Final Compression

Behaviour is real evidence. It is not self-explaining evidence.

Describe the action before naming the learner. Check expectations, entry point, challenge, support, response opportunity, group conditions and agency. Use clear routines. Offer bounded choices. Acknowledge productive participation specifically. Protect peers and safety without turning every difficult interaction into a character verdict.

The useful question is not “What is wrong with this learner’s attitude?” It is “What does this behaviour tell us about the learner–task–tutor system, and what small change will reveal whether our explanation is right?”

That is the Behaviour–Learning Differential.

That is Tutor Handbook Volume 0169.

Small Group Tutorials

The Tutor Handbook Vol No.0169 | The Behaviour–Learning Differential — How a Tutor Responds to Off-Task, Refusal or Disruptive Behaviour Without Assuming the Problem Is Motivation When the Task, Support or Group Conditions May Be Driving It