eduKateSG Learning Node Series · 0076
How Instructional Coaching Works | Turn Observation, Feedback and Rehearsal Into Better Teaching
A teacher attends a workshop on questioning.
The ideas are sensible. The examples are good. The teacher leaves with notes, slides and intention.
On Monday, thirty students arrive. Time compresses. One explanation overruns. A difficult question lands badly. The teacher falls back to the familiar routine.
The professional development was not useless. It was incomplete.
Instructional coaching closes the distance between knowing a teaching idea and being able to execute it reliably in a real classroom.
The 50-Second Read
- Instructional coaching is sustained, practice-connected professional learning rather than a single workshop or appraisal event.
- A coach and teacher usually work around a focused instructional goal, gather evidence, identify a small next move, rehearse or plan it, implement it and review what happened.
- Observation is useful only when it produces discriminating evidence rather than a vague judgement that the lesson was “good” or “weak.”
- Feedback should be narrow enough to use, specific enough to act on and close enough to classroom reality to survive implementation.
- Rehearsal matters because teachers need execution practice, not only conceptual agreement.
- A major meta-analysis of 60 causal studies found meaningful average effects of teacher coaching on instructional practice and smaller positive effects on student achievement.
- Scale creates difficulty: programmes that work intensively with a small number of teachers may lose effectiveness when expanded without enough coaching quality, time or trust.
- Coaching should develop teacher judgement, not convert teachers into script-followers.
Canonical Owner Boundary
This page owns instructional coaching: the repeated observation-feedback-rehearsal-implementation cycle used to improve classroom practice. How Education Works | Teacher Professional Learning owns the wider system of continuous educator development. Teacher Appraisal & Professional Review owns formal evaluation and review. How Teacher Noticing Works owns the ability to see evidence of student thinking. How Implementation Fidelity Works owns preservation of active ingredients during delivery. Instructional coaching asks: how does a teacher convert evidence about current practice into a better next performance?
1. Teaching Knowledge and Teaching Performance Are Not the Same
A teacher can understand wait time, retrieval practice, questioning, modelling or checking for understanding and still struggle to execute the technique at the right moment.
Classrooms are live systems. Teachers make rapid decisions about explanation, pacing, behaviour, examples, misconceptions and who needs attention next. That creates an implementation gap between professional knowledge and fluent classroom use.
2. Coaching Narrows the Improvement Target
“Improve your questioning” is too broad to train.
“After asking a high-value question, wait three seconds before selecting a respondent and then ask one follow-up that requires evidence” is observable, rehearsable and reviewable.
Coaching works best when the improvement target is small enough to enter tomorrow’s lesson but important enough to change learning.
3. The Coaching Loop
goal → observe → identify leverage point → model or rehearse → implement → collect evidence → reflect → refine → repeat
The power comes from recurrence. A workshop can introduce an idea. Coaching keeps returning until the idea becomes usable practice.
4. Observation Must Be Selective
A lesson contains too much information to observe everything equally. A coach who tries to record behaviour, questioning, explanations, student talk, pacing, error correction, transitions and curriculum alignment simultaneously may produce a long report and little usable diagnosis.
Strong observation begins with a question. Did students have enough thinking time? Which students answered? What evidence showed that the explanation changed understanding? How often did the teacher rescue the task before students had a chance to think?
5. Evidence Beats Impression
“The lesson felt rushed” can become: eight of twelve questions were answered by the same four students, average wait time was under one second, and the final independent task began with six minutes remaining.
The second description gives the teacher something to work with. Coaching becomes stronger when it turns general impressions into observable patterns.
6. Feedback Needs an Actionable Next Move
Feedback that contains twelve recommendations is often feedback that contains no priority.
A coach should identify the highest-leverage reachable change: the move that is likely to improve learning and is realistic enough for the teacher to implement now.
The next step can be demanding without being vague.
7. Rehearsal Converts Advice Into Motor and Verbal Readiness
Teachers speak, gesture, scan, write, question and respond under time pressure. Many instructional moves therefore contain a performance component.
Rehearsal lets the teacher practise a question stem, board sequence, transition, explanation or response to a likely misconception before the class supplies real complexity.
This is the educational equivalent of simulation: reduce consequence while increasing execution quality.
8. Modelling Shows What the Move Looks Like
A coach can describe a technique, show a video, demonstrate with students, co-teach or role-play the move. Modelling helps when language such as “be more responsive” hides the actual behaviour required.
The model should expose the decision rule, not just the performance. The teacher needs to know not only what the coach did, but what cue triggered the move and why.
9. Trust Is Operational Infrastructure
Coaching depends on teachers being willing to reveal imperfect practice. If every observation feels like hidden appraisal, teachers rationally protect themselves: they select safer lessons, conceal uncertainty and avoid experimentation.
That does not mean coaching lacks accountability. It means the function of coaching should be clear enough that development is not constantly confused with judgement.
10. The Evidence Base Is Strong Enough to Matter
Matthew Kraft, David Blazar and Dylan Hogan reviewed 60 causal studies of teacher coaching. Their meta-analysis found an average effect of 0.49 standard deviations on instructional practice and 0.18 standard deviations on student achievement.
The authors also warned that effects were smaller in larger effectiveness trials than in smaller efficacy studies, highlighting the challenge of scaling coaching while maintaining quality.
Source: Kraft, Blazar & Hogan, The Effect of Teacher Coaching on Instruction and Achievement.
11. Coaching Has to Reach the Classroom
Professional learning can become intellectually rich but operationally distant. Teachers discuss principles, inspect research and exchange resources but never test one changed move in a real lesson.
Instructional coaching keeps the return path short:
idea → classroom move → student response → evidence → revised move
12. Student Evidence Is the Final Test
A teacher can execute a target move exactly as planned and still fail to improve learning. The move is not the outcome.
Coaching should therefore ask what changed for students: participation, quality of explanation, retrieval success, misconceptions revealed, independence, accuracy, transfer or another relevant signal.
This prevents fidelity from becoming ritual.
13. Coaching Should Develop Teacher Judgement
The weakest coaching creates dependency: the teacher waits for an expert to identify every problem and prescribe every move.
The strongest coaching gradually transfers diagnosis. Teachers learn to notice the same evidence, name the same bottlenecks and choose increasingly good next actions themselves.
Eventually the teacher becomes capable of running much of the coaching loop internally.
14. Coaching and Teacher Noticing
Teacher noticing is a natural coaching target because many instructional improvements begin with seeing better.
A coach may help a teacher notice who is not participating, where misconceptions first appear, when students are copying rather than reasoning or which part of an explanation overloads working memory.
Once perception improves, action can improve.
15. Coaching and Implementation Fidelity
New programmes often fail because the visible activity is copied while the active ingredient disappears.
Coaching can protect fidelity by identifying the mechanism that must survive adaptation. A questioning routine may vary in wording while preserving retrieval, universal participation and evidence of thinking.
Good coaching preserves function while allowing professional judgement.
16. Cross-Domain Comparison: Sports Coaching and Surgical Training
Sports coaches observe performance, isolate one controllable weakness, demonstrate, rehearse, adjust and retest under competition-like conditions. Surgical training similarly combines expert observation with simulation, deliberate rehearsal and debrief.
Teaching is not sport or surgery, but the performance-development architecture is recognisable: expertise improves when evidence from real execution is converted into focused practice.
17. Missing-Node Scan: Where Coaching Commonly Breaks
- Goals are too broad to observe.
- Coaches give many comments but no priority.
- Observation is evaluative rather than diagnostic.
- Teachers understand the advice but never rehearse it.
- Follow-up occurs months later, after the implementation window has closed.
- Coaching focuses on teacher behaviour without checking student response.
- The coach prescribes a preferred style rather than a mechanism tied to learning.
- Trust is weak because coaching and appraisal are blurred.
- Programmes scale faster than coach quality can be maintained.
- The same coaching script is applied regardless of teacher experience or classroom context.
18. A Practical Coaching Cycle
- Select one student-learning problem worth solving.
- Translate it into an observable teacher move.
- Collect baseline evidence.
- Agree on one high-leverage next action.
- Model or rehearse the action.
- Implement it in a real lesson.
- Collect student and teacher evidence.
- Debrief quickly.
- Keep, adapt or replace the move.
- Repeat until the teacher can diagnose and adjust independently.
19. Evidence and Limits
The coaching literature is promising, but the average effect does not mean every coaching programme works. Content, coach skill, teacher participation, dose, trust, organisational support and the quality of the target practice all matter.
There is also a measurement problem. Instructional quality is difficult to capture, student outcomes are influenced by many variables and short studies can miss whether change survives after coaching ends.
Coaching should therefore be treated as a disciplined improvement process rather than a branded intervention. Its value lies in the quality of the loop.
20. The Return Path
The teacher leaves the workshop knowing that wait time matters.
The coach helps turn that knowledge into a specific classroom move. They observe. Rehearse. Try. Watch the students. Adjust.
A week later, the teacher no longer needs the reminder.
Months later, the teacher can diagnose a similar problem alone.
Instructional coaching works when feedback does not end as advice. It returns to the classroom as a better decision, a better move and eventually better professional judgement.