How to Learn by Teaching | Turn Explaining Into a Study Method

How to learn by teaching is to prepare an explanation for another person, generate that explanation from your own knowledge, respond to questions, notice gaps and then revise what you understand.

Teaching can improve the teacher’s learning because it changes the task. The student must organise ideas, anticipate confusion, select examples and produce an explanation that another mind can follow.

However, the benefits are not automatic. A learner can simply read notes aloud, create polished slides without understanding or divide a group project so narrowly that little learning occurs. This guide shows how to make teaching generative.

This guide is designed for students, parents and educators who want a practical answer to how to learn by teaching someone else. The aim is not to add another revision slogan. It is to turn the strategy into a visible sequence that can be taught, practised, checked and transferred.

This approach is especially useful for learners who need to:

students who understand better when they explain aloud;
students preparing presentations;
study partners who want structured peer teaching;
students learning concept-heavy subjects;
students who need to discover gaps in understanding;
students who want to practise answering questions;
tutors using peer explanation in small groups;
students creating revision videos or teaching materials.

A More Important Learning Move Than It First Appears

Preparing to teach changes attention. The learner begins asking what the main ideas are, what sequence makes sense, which examples will clarify them and which questions a listener might ask.

During teaching, explanation must be generated. That production can expose missing links that were hidden during rereading.

Questions from the learner being taught can make the process even more diagnostic, because they force the student-teacher to distinguish actual understanding from a rehearsed script.

The Hidden Problem: Teaching Can Become Performance Without Learning

A polished presentation can be created by copying definitions, dividing slide work and reading from notes. The audience may receive information while the student-teacher learns little.

The learning value appears when the teacher must organise, retrieve, explain, respond and adapt. Source material should support preparation, but not carry the whole delivery.

The goal is not to become a professional teacher. It is to use the demands of teaching as a structured way to generate and test understanding.

What Learning by Teaching Actually Means

Working definition: Learning by teaching is a generative learning approach in which a learner prepares and communicates instructional material for others, using explanation, organisation and response to questions as opportunities to deepen their own learning.

Teaching can be real or simulated. A student can teach a classmate, sibling, study group, camera or imagined audience. A real audience adds the possibility of questions and feedback.

The strongest version requires the learner to produce from memory or limited notes, explain relationships and check accuracy afterwards. Reading a source aloud is not the same process.

Our First-Principles Teaching Method for Learning by Teaching

1. Choose a bounded teaching target

Select one concept, method or relationship that can be explained coherently.

Avoid teaching an entire subject in one session.

2. Prepare the structure

Decide the learning goal, sequence, examples and likely misconceptions.

Preparation itself should require understanding, not slide decoration.

3. Retrieve before writing the script

List the main points from memory first, then check sources.

This reveals what the learner already owns and what still depends on notes.

4. Teach in your own words

Explain the idea clearly without reading line by line.

Use technical language where needed, but unpack it.

5. Use examples and contrasts

Give a valid example, a contrasting case and a question for the learner.

This forces the teacher to clarify boundaries.

6. Invite questions

Let the listener interrupt or ask for clarification.

Mark any question that cannot be answered confidently rather than bluffing.

7. Repair after teaching

Return to sources and correct unclear or inaccurate parts.

The teaching session generates the repair list.

8. Reteach or retrieve later

Explain the same idea again after a delay with fewer notes.

The second performance tests whether the learning became more independent.

What Happens During a 90-Minute Learning Session

10 minutes — choose and retrieve

Each student selects one small concept and lists the main points without notes.

15 minutes — prepare the teaching structure

Students plan sequence, examples and one misconception.

15 minutes — first teach-back

One student explains while two peers listen.

15 minutes — questions and challenge

Peers ask for clarification, examples or boundary cases.

20 minutes — repair and fresh explanation

The student checks gaps and reteaches more precisely.

15 minutes — independent application

All students answer a new question using the taught concept.

The exact timing changes with age, subject and current readiness. The stable structure is more important: expose the decision, require an attempt, compare with a reliable model, repair the first meaningful weakness and return later with less support. A learner should not spend the entire lesson watching somebody else perform the thinking.

Why Three Students Can Be a Useful Tutorial Size

In a three-student tutorial, learning by teaching can be made visible quickly. One learner can attempt the task, another can explain the principle, and the third can test a boundary case or alternative. The tutor can then compare the quality of reasoning rather than relying only on final answers.

Small-group teaching is useful when it increases the density of meaningful feedback. The goal is not constant discussion. Short explanation is followed by individual work so every student must eventually perform without the group carrying the decision.

How Learning by Teaching Changes Across Subjects

English

Teach a grammar distinction, vocabulary nuance, comprehension reasoning move or writing structure. Listeners should challenge examples and ask why a particular language choice works.

The disciplinary standard still matters. A general learning strategy can organise attention and practice, but it cannot replace knowing what counts as a valid method, precise explanation, relevant evidence or complete answer in this subject.

Mathematics

Teach why a method works, not merely the steps. Peers should ask when the method applies, what changes under a different condition and how to check the result.

Science

Teach a mechanism or experimental relationship with diagrams and causal explanation. Questions should probe variables, evidence and boundaries.

Humanities

Teach a concept, event sequence or argument using evidence and qualifications. Peers can ask for alternative explanations or missing causes.

What Strong Performance Looks Like

The teaching target is bounded.
Preparation begins with retrieval.
Explanations are in the learner’s own words.
Examples clarify the concept.
Questions are invited.
Uncertainty is admitted.
Sources are used for repair.
Reteaching occurs after correction.
Listeners also complete a task.
Teaching leads into independent performance.

These behaviours are observable. They can therefore be modelled, practised and retested. The student does not need a new personality; the student needs a more reliable sequence of decisions.

Common Failure Modes

1. Reading slides aloud

The student presents text supplied by the screen instead of generating an explanation.

Reduce on-screen wording and require teaching from a small cue card.

2. Presentation design overload

Time goes into fonts, animation and images rather than understanding.

Separate learning preparation from visual production.

3. No audience interaction

The student delivers a memorised monologue and never has to adapt.

Add questions, examples and misconception checks.

4. Teaching what is not understood

The student prepares a confident explanation on top of a weak foundation.

Require source checking and a pre-teach retrieval test.

5. No repair

Unanswered questions are forgotten after the presentation ends.

Keep a gap list and revisit sources immediately.

6. Group fragmentation

Each member teaches a tiny isolated slide and never understands the whole topic.

Require every student to retrieve the overall structure.

7. Teaching without later retrieval

The explanation happens once and is never revisited.

Reteach or self-test after delay.

8. Audience passivity

Listeners watch but do not process.

Give them a question, comparison or summary task.

Primary School, Secondary School and Examination Years

Primary school

Use very short teach-backs with toys, drawings or mini-whiteboards. A child can explain one rule or process to an adult, sibling or classmate and answer one simple question.

Keep the routine short, concrete and repeatable. Adults can provide structure initially, but should reduce prompts as soon as the child can perform the decision independently.

Secondary school

Students can prepare peer explanations, audio notes, revision videos or whiteboard teaching. Limit notes and require response to questions so the activity remains generative.

Secondary students need greater independence because subjects become denser and assessments require more transfer. They should increasingly be able to explain why a strategy fits, not merely follow a worksheet or teacher cue.

Before major examinations

Use learning by teaching earlier in revision to deepen understanding and expose gaps. Closer to exams, convert the repaired knowledge into timed retrieval and authentic questions.

Near examinations, decrease novelty in the study system and increase authenticity in the task. Use the same established method on mixed, timed and exam-like work rather than inventing a complicated new routine during the final days.

A Practical Diagnostic Checklist

Can the learner state the teaching goal?
Can they retrieve the structure before opening notes?
Can they explain in their own words?
Can they provide examples?
Can they answer questions?
Do they admit uncertainty?
Do they repair gaps afterwards?
Can they reteach with fewer notes?
Do listeners learn actively too?
Does later application improve?

Use the checklist to choose the next teaching move, not to label the learner. If several items are weak, repair the earliest breakdown or the one that creates the largest downstream cost.

Practice Laboratory

Practice 1: Two-minute teach-back

Explain one concept for two minutes with no notes.

Mark where the explanation breaks or becomes vague.

Practice 2: Cue-card limit

Teach using only five short prompts.

This prevents the script from carrying the whole explanation.

Practice 3: Question storm

After teaching, listeners ask five why, how or boundary questions.

Unanswered items become the study list.

Practice 4: Example challenge

The teacher gives one example; the audience gives a near non-example.

The teacher must explain the decisive difference.

Practice 5: Teach the error

Explain a common wrong method and why it fails.

Then demonstrate the correct principle.

Practice 6: Role rotation

In a three-student group, rotate teacher, questioner and checker.

All three later complete an independent task.

Practice 7: Video without script

Record a short explanation from a cue list, then review for missing relationships.

Rerecord only after targeted repair.

Practice 8: Diagram teaching

Teach a Science process using a blank diagram built live.

The audience checks whether every arrow has a causal meaning.

Practice 9: Math method lesson

Teach when a method applies, why it works and how to verify the result.

Then solve one fresh problem together.

Practice 10: Delayed reteach

Explain the concept again several days later without the original presentation.

Compare which details and relationships survived.

Three Student Cases

Case 1: Maren makes beautiful slides

Maren spends hours designing presentation slides and then reads them aloud.

Her tutor limits each slide to one visual or five words and requires the explanation to come from memory.

Preparation shifts from decoration to organisation and retrieval.

Case 2: Iona avoids questions

Iona teaches fluently until a peer asks something unexpected, then changes the subject.

The group adds a rule that unanswered questions are valuable diagnostic data, not embarrassment.

She begins marking gaps openly and repairs them after the session.

Case 3: Leonie teaches only one section

In group projects, Leonie learns her assigned part but not how it fits the whole topic.

Every student must now give a one-minute overview of the complete structure before teaching their section.

The local expertise becomes connected to global understanding.

How Parents Can Help Without Taking Over

Parents can support learning by teaching by asking for the learner’s current decision before supplying an answer. Useful prompts include: “What do you think first?”, “What makes that step valid?”, “Where did the explanation become unclear?”, “What would you check?”, “What changes in a new example?”, and “What will you try without looking?”

When the student can perform a step independently, remove the prompt. When the student cannot, restore the smallest amount of support needed to restart thinking, then fade it again. Support should transfer control rather than permanently replace it.

How to Measure Progress

Students organise knowledge more clearly.
Unanswered questions reveal gaps earlier.
Explanations rely less on scripts.
Examples improve.
Teaching becomes more adaptive.
Reteaching after repair becomes easier.
Listeners participate actively.
Later retrieval and application improve.

Marks remain important, but they are a lagging indicator. Earlier progress may appear as better starts, more accurate strategy choice, clearer explanations, less dependence on models or more reliable transfer. Track those behaviours and then confirm that they translate into stronger performance.

A Four-Week Implementation Plan

Week 1 — Short teach-backs

Use one two-minute explanation per study session.

Focus on structure, not performance polish.

Week 2 — Add audience questions

Use a peer or family member and record unanswered questions.

Repair each gap before the next session.

Week 3 — Reduce source dependence

Teach from cue cards or memory and add examples plus non-examples.

Check accuracy after delivery.

Week 4 — Connect teaching to performance

Follow every teach-back with fresh questions or writing.

Keep the strategy where explanation predicts stronger application.

Evidence and Responsible Use

A 2022 meta-analysis of learning by teaching found a small overall advantage when students generated teaching materials compared with control activities, while also showing that outcomes depend on implementation. The evidence suggests that generative production matters: visual or audio-visual teaching materials sometimes outperformed text-only products, and full access to source materials during generation was not always beneficial. Learning by teaching should therefore be structured to require explanation, organisation and retrieval rather than mere content reproduction.

No learning strategy guarantees a grade. Outcomes also depend on prior knowledge, teaching quality, task design, sleep, practice, attendance and the learner’s wider context. Use the method as part of a complete learning system rather than as a replacement for subject teaching.

Frequently Asked Questions

Do I need a real person to teach?

No. An imaginary audience, voice recording or whiteboard explanation can still require generation, though a real listener adds useful questions.

Is learning by teaching the same as the Feynman Technique?

They overlap strongly. The Feynman Technique is a popular structured routine centred on simple explanation and gap repair; learning by teaching is a broader research area.

Should I write a full script?

Usually not. A full script can become another source to read. Use a brief structure and generate the explanation.

What if I teach something wrong?

Check afterwards and correct it. Accuracy verification is essential.

Can this work in Mathematics?

Yes, especially when teaching includes why a method works and when it applies, followed by real problem solving.

Does making a presentation automatically improve learning?

No. The benefit depends on whether the learner has to organise, retrieve and explain rather than simply copy.

How long should a teach-back be?

Short enough to stay focused. Two to ten minutes per bounded concept is often sufficient for diagnostic use.

When should I use it before exams?

Use it to deepen and repair understanding, then shift toward authentic retrieval and timed practice as the exam approaches.

Helpful Reading on eduKateSingapore

Teaching Helps Most When the Learner Has to Generate

The educational value is not the audience, the slide deck or the performance.

Retrieve the structure. Explain it. Invite a question. Mark the gap. Repair it. Teach again. Then solve or write independently. That is how teaching becomes a study method.

Properly taught kids shine a bright light into the future.

Extended Diagnostic Workshops

Workshop 1: Choose a bounded teaching target × Presentation design overload

Set up one fresh task in which the learner must practise choose a bounded teaching target while watching specifically for presentation design overload. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Select one concept, method or relationship that can be explained coherently.

Time goes into fonts, animation and images rather than understanding. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: After teaching, listeners ask five why, how or boundary questions. Unanswered items become the study list. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 2: Prepare the structure × Teaching what is not understood

Set up one fresh task in which the learner must practise prepare the structure while watching specifically for teaching what is not understood. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Decide the learning goal, sequence, examples and likely misconceptions.

The student prepares a confident explanation on top of a weak foundation. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: In a three-student group, rotate teacher, questioner and checker. All three later complete an independent task. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 3: Retrieve before writing the script × Group fragmentation

Set up one fresh task in which the learner must practise retrieve before writing the script while watching specifically for group fragmentation. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. List the main points from memory first, then check sources.

Each member teaches a tiny isolated slide and never understands the whole topic. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Teach when a method applies, why it works and how to verify the result. Then solve one fresh problem together. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 4: Teach in your own words × Audience passivity

Set up one fresh task in which the learner must practise teach in your own words while watching specifically for audience passivity. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Explain the idea clearly without reading line by line.

Listeners watch but do not process. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Teach using only five short prompts. This prevents the script from carrying the whole explanation. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 5: Use examples and contrasts × Presentation design overload

Set up one fresh task in which the learner must practise use examples and contrasts while watching specifically for presentation design overload. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Give a valid example, a contrasting case and a question for the learner.

Time goes into fonts, animation and images rather than understanding. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Explain a common wrong method and why it fails. Then demonstrate the correct principle. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 6: Invite questions × Teaching what is not understood

Set up one fresh task in which the learner must practise invite questions while watching specifically for teaching what is not understood. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Let the listener interrupt or ask for clarification.

The student prepares a confident explanation on top of a weak foundation. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Teach a Science process using a blank diagram built live. The audience checks whether every arrow has a causal meaning. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 7: Repair after teaching × Group fragmentation

Set up one fresh task in which the learner must practise repair after teaching while watching specifically for group fragmentation. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Return to sources and correct unclear or inaccurate parts.

Each member teaches a tiny isolated slide and never understands the whole topic. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Explain one concept for two minutes with no notes. Mark where the explanation breaks or becomes vague. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 8: Reteach or retrieve later × Audience passivity

Set up one fresh task in which the learner must practise reteach or retrieve later while watching specifically for audience passivity. Begin without the answer visible. Ask the learner to state the goal, attempt the task and mark the first uncertainty before checking. Explain the same idea again after a delay with fewer notes.

Listeners watch but do not process. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: The teacher gives one example; the audience gives a near non-example. The teacher must explain the decisive difference. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 9: Choose a bounded teaching target × Presentation design overload

Time goes into fonts, animation and images rather than understanding. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Record a short explanation from a cue list, then review for missing relationships. Rerecord only after targeted repair. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 10: Prepare the structure × Teaching what is not understood

The student prepares a confident explanation on top of a weak foundation. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Explain the concept again several days later without the original presentation. Compare which details and relationships survived. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 11: Retrieve before writing the script × Group fragmentation

Each member teaches a tiny isolated slide and never understands the whole topic. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: After teaching, listeners ask five why, how or boundary questions. Unanswered items become the study list. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 12: Teach in your own words × Audience passivity

Listeners watch but do not process. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: In a three-student group, rotate teacher, questioner and checker. All three later complete an independent task. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 13: Use examples and contrasts × Presentation design overload

Time goes into fonts, animation and images rather than understanding. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Teach when a method applies, why it works and how to verify the result. Then solve one fresh problem together. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 14: Invite questions × Teaching what is not understood

The student prepares a confident explanation on top of a weak foundation. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Teach using only five short prompts. This prevents the script from carrying the whole explanation. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 15: Repair after teaching × Group fragmentation

Each member teaches a tiny isolated slide and never understands the whole topic. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Explain a common wrong method and why it fails. Then demonstrate the correct principle. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 16: Reteach or retrieve later × Audience passivity

Listeners watch but do not process. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Teach a Science process using a blank diagram built live. The audience checks whether every arrow has a causal meaning. Schedule a delayed return so the next success cannot be explained only by immediate imitation.

Workshop 17: Choose a bounded teaching target × Presentation design overload

Time goes into fonts, animation and images rather than understanding. After correction, change one consequential feature: wording, values, representation, context, evidence, order or time pressure. Then use this related exercise: Explain one concept for two minutes with no notes. Mark where the explanation breaks or becomes vague. Schedule a delayed return so the next success cannot be explained only by immediate imitation.