The English Tutorial Lesson No.0052 | From Answering Questions to Asking Better Questions — Retrieval, Clarification, Diagnostic, Inferential, Evaluative, Socratic and Metacognitive Questioning

English Tutorial Lesson No.0007 · English Tutorial Lesson No.0030 · English Tutorial Lesson No.0008

Part I — Questions as cognitive tools

1. A question directs attention

“What colour was the door?” makes the reader search for a detail. “Why does the writer mention the door twice?” makes the reader search for function and recurrence. The text is the same; attention changes.

Question design therefore shapes what learners notice. Repeatedly asking only factual questions can train fact retrieval while leaving inference, evaluation and self-monitoring underdeveloped.

Every question is an attentional instruction.

2. A question creates a response space

“Did you understand?” invites yes/no and often social compliance. “Which sentence became difficult to follow?” creates a more useful response space.

The wording determines what evidence can emerge.

Diagnostic questions should make the hidden problem observable.

3. The best question depends on the learning job

A retrieval question is excellent when memory is the target and weak when evaluation is the target. A broad open question can support discussion and be inefficient for checking one factual prerequisite.

Question quality is conditional, not absolute.

Ask what thinking the learner needs next.

4. Questions can teach strategies indirectly

“What do you know already?” teaches activation of prior knowledge. “What evidence supports that?” teaches evidence checking. “What will you try if this does not work?” teaches contingency planning.

Repeated external questions can become internal self-talk.

EEF metacognition guidance explicitly uses planned questions to support planning, monitoring and evaluation until learners become more independent.

5. Questions can also create dependency

If the teacher supplies every next question forever, the student may wait passively for prompts.

Good scaffolding fades. After modelling useful questions, ask learners to generate them.

The end state is self-questioning.

6. A question can reveal knowledge without improving it

Assessment questions diagnose current performance. Learning questions should also create an opportunity to revise or extend thinking.

After “What is the main idea?” follow with feedback or comparison if the answer is weak.

Questioning should connect diagnosis to action.

Part II — Retrieval questions

7. Retrieval questions ask memory to produce

“What does corroboration mean?” “What are the three stages of the procedure?” These questions strengthen access to learned information when answered without looking.

Retrieval differs from recognition, where the learner selects from visible options.

Use retrieval deliberately for durable memory.

8. Keep retrieval questions aligned to taught knowledge

A question should not masquerade as memory practice when the content was never learned.

If the learner cannot answer because knowledge is absent, teach first.

Diagnosis should distinguish forgetting from non-learning.

9. Retrieval questions can vary cues

Definition → example → reverse definition → application. “Define metaphor.” “Which device is this?” “Give a metaphor.”

Varying cue form builds flexible access.

Memory should not depend on one worksheet wording.

10. Use delayed retrieval

Questions asked one week later test durability more than immediate repetition.

Spacing matters.

Question timing is part of design.

11. Mix retrieval with discrimination

“What is a simile?” followed by “Why is this not a simile?” requires category boundaries.

Retrieval can feed deeper reasoning.

Question sequences create progression.

12. Avoid giving the answer inside the question

“The answer is metaphor because it compares two things, right?” is not retrieval.

Leading scaffolds can be useful later, but label them as hints.

Protect the cognitive job.

Part III — Clarification questions

13. Clarification questions reduce ambiguity

“What do you mean by ‘effective’?” “Which group does ‘they’ refer to?”

These questions are useful before disagreement because the problem may be wording rather than idea.

Clarification protects shared understanding.

14. Ask for examples

“Can you give a concrete example?” forces abstract language to touch a case.

The example can reveal whether two people attach the same meaning to a term.

Examples are diagnostic.

15. Ask for rephrasing

“Can you say that another way?” checks whether the learner controls the idea or memorised one sentence.

Rephrasing can also reveal hidden uncertainty.

Flexible expression supports understanding.

16. Ask for boundaries

“Would this case count too?” “What would not count?”

Boundary questions sharpen definitions.

Non-examples create conceptual precision.

17. Clarify reference before inference

If a pronoun is ambiguous, do not build interpretation on it.

Resolve local meaning first.

Question order matters.

18. Clarification should not smuggle an answer

“When you say the writer is angry, do you really mean disappointed?” leads.

Prefer “What word in the passage makes you describe the tone as angry?”

Clarification should open rather than overwrite.

Part IV — Diagnostic questions

19. Diagnostic questions locate the weak link

“Can you read the word?” “Do you know what it means?” “Can you explain the sentence?” Each isolates a different possible source of comprehension failure.

Broad questions such as “Why are you weak in comprehension?” rarely produce useful diagnosis.

Decompose the process.

20. Ask one variable at a time

If a learner fails a paragraph, test vocabulary, syntax, background knowledge and inference separately.

Compound diagnostic questions hide which part caused difficulty.

Good diagnosis narrows.

21. Use contrastive diagnostics

Give one easier and one harder item differing in a controlled feature.

If only long sentences fail, syntax may be involved.

Comparison creates diagnostic evidence.

22. Ask the learner to predict difficulty

“Which part do you expect to be hardest?” Compare prediction with actual failure.

This reveals metacognitive accuracy.

Self-knowledge can be trained.

23. Ask what strategy was used

A correct answer may come from a fragile strategy; a wrong answer may use a sensible method with one missing fact.

Process questions reveal learning needs beyond score.

Outcome is not the whole diagnosis.

24. Diagnose confidence as well as correctness

High-confidence wrong answers require different repair from low-confidence correct answers.

Ask for confidence before feedback.

Calibration is diagnostic data.

Part V — Inferential questions

25. Inferential questions require a constrained step beyond explicit wording

“Why did the character leave early?” may require combining action, dialogue and context.

The answer must remain anchored to textual evidence.

Inference is not free imagination.

26. Ask what evidence makes the inference possible

Every inference answer should point back to clues.

“Because I think so” is not sufficient.

Inference = evidence plus reasoning.

27. Ask for alternative interpretations

“What else could this behaviour mean?” prevents one plausible inference from becoming certainty.

Then compare evidence strength.

Inference is calibrated.

28. Ask what evidence would change the inference

This makes the claim falsifiable and teaches open-mindedness.

It is especially useful in literary and social reasoning.

Interpretation should remain evidence-responsive.

29. Ask local before global inference

Understand the sentence and paragraph before asking for theme or motive.

High-level inference depends on lower-level comprehension.

Question sequencing follows dependency.

30. Do not call unsupported speculation “higher-order thinking”

Complexity does not excuse evidence loss.

Strong inferential questions constrain the reasoning space enough for answers to be judged.

Depth needs evidence.

Part VI — Evaluative questions

31. Evaluation needs criteria

“Is this a good source?” is incomplete until the learner identifies good for what: current fact, expert explanation, lived experience, policy wording?

Ask which criteria matter.

Evaluation becomes structured.

32. Ask for evidence supporting the judgement

“Which feature of the source makes you trust it more?”

A score without reasons teaches little.

Evaluation should be auditable.

33. Ask for trade-offs

“What improves if we choose this option, and what becomes worse?”

Trade-off questions resist simplistic rankings.

Evaluation sees multiple criteria.

34. Ask who benefits and who bears cost

Stakeholder questions reveal distribution.

They do not determine the final judgement automatically.

Evaluation needs perspective.

35. Ask compared with what

“Expensive” or “effective” requires baseline.

Comparison makes evaluative language meaningful.

Criteria need reference points.

36. Ask under what conditions

An approach can be effective for one group and weak for another.

Conditional evaluation produces nuance.

Context matters.

Part VII — Causal questions

37. Why questions search for cause

But “why” can invite stories even when evidence is weak.

Ask what evidence distinguishes cause from coincidence.

Causal questioning needs method.

38. Ask what happened first

Cause must precede effect under ordinary causal logic.

Chronology can eliminate impossible explanations.

Sequence is a basic causal check.

39. Ask what else changed

Confounding variables may explain the outcome.

“What else happened at the same time?” broadens analysis.

Causation needs alternative explanations.

40. Ask what mechanism connects them

A plausible mechanism strengthens explanation but does not alone prove causation.

“How would X produce Y?”

Mechanism makes causal claims inspectable.

41. Ask what happens when X is absent

Comparison groups or natural contrasts can provide evidence.

The question encourages counterfactual thinking.

Cause implies difference under alternatives.

42. Ask whether direction could reverse

Does X cause Y, or does Y influence X?

Reverse causation is common in observational reasoning.

Causal questions should be bidirectional initially.

Part VIII — Comparative questions

43. Comparison needs a basis

“How are these texts different?” is broad. “How do they use evidence differently?” creates a dimension.

Specific comparison questions produce stronger analysis.

Criteria organise contrast.

44. Ask similarity before difference sometimes

Shared ground can make later disagreement clearer.

There is no universal order; choose based on task.

Question sequence shapes relation-building.

45. Ask whether differences matter

Not every difference is analytically important.

“So what changes because Source B uses interviews rather than a survey?”

Comparison needs consequence.

46. Ask whether the cases are comparable

Population, time and definition may differ.

Cross-text comparison should pass an alignment gate.

Comparison can fail before analysis begins.

Part IX — Definitional questions

47. Ask what category the concept belongs to

“A metaphor is a kind of what?”

Category gives structural placement.

Definitions become more than synonym lists.

48. Ask what distinguishes it

“How is a metaphor different from a simile?”

Contrast sharpens boundaries.

Definitions need distinguishing features.

49. Ask for examples and non-examples

Concept mastery requires knowing membership and exclusion.

Students can test definitions.

Categories become operational.

50. Ask where the boundary is uncertain

Borderline cases reveal whether the definition is too broad or narrow.

Advanced learning tolerates ambiguity.

Definitions can be revised.

Part X — Socratic questioning

51. Socratic questioning is not endless “why?”

Useful Socratic dialogue probes definitions, assumptions, evidence, implications and alternatives.

The purpose is examination of reasoning, not teacher performance or learner humiliation.

Questions should follow the learner’s actual answer.

52. Probe assumptions

“What must be true for that argument to work?”

Hidden premises become visible.

Reasoning can then be tested.

53. Probe evidence

“What evidence supports that?” “How reliable is it?”

Claims become connected to sources.

Socratic dialogue can strengthen epistemic discipline.

54. Probe alternatives

“What else could explain this?”

Alternative hypotheses reduce tunnel vision.

Open inquiry becomes structured.

55. Probe implications

“If that is true, what follows?”

Consequences expose inconsistency or hidden cost.

Reasoning extends beyond first claims.

56. Probe consistency

“Does that rule apply to the earlier example too?”

Consistency questions reveal ad hoc reasoning.

Principles should survive comparable cases.

57. Probe criteria

“What makes one option fairer?”

Evaluation becomes explicit.

Values are named rather than assumed.

58. Stop when the learner needs teaching, not another question

If knowledge is missing, endless probing can feel punitive.

Explain, model or provide evidence, then resume questioning.

Socratic method is not withholding help.

Part XI — Metacognitive questions

59. Planning questions come before action

EEF questioning resources recommend prompts that help learners plan, monitor and evaluate their learning.

Planning examples: What is the task? What do you know already? Which strategy might fit? What could go wrong?

Questions externalise future control.

60. Monitoring questions occur during work

“Is the strategy working?” “What evidence have you gathered?” “Are you still answering the question?”

Monitoring catches drift while revision remains cheap.

Self-regulation becomes active.

61. Evaluation questions come after work

“What worked?” “What failed?” “What would you change next time?”

Reflection should produce future strategy, not generic feelings.

Evaluation feeds the next plan.

62. Questions should be subject-specific

EEF guidance emphasises embedding metacognition in curriculum content. “What strategy will you use?” becomes stronger as “How will you decide whether this paragraph contains evidence or inference?”

Generic thinking skills need domain anchoring.

Questions should touch real knowledge.

63. Least-help-first questioning supports independence

Begin with a broad prompt, then clue, then model only if needed.

This preserves learner effort while preventing unproductive struggle.

Question intensity should adapt.

64. Fade external questions

After repeated modelling, ask “What question should you ask yourself now?”

The learner begins generating the scaffold internally.

Questioning becomes self-regulation.

Part XII — Open and closed questions

65. Closed questions are efficient for checks

“Is the source dated 2026?” can confirm a fact quickly.

Closed does not mean low quality.

Match format to job.

66. Open questions create wider response spaces

“How does the source’s date affect your confidence?” invites explanation.

Open questions are useful when reasoning or perspective matters.

But they can be too broad without scaffolding.

67. Funnel from broad to specific

Start: “What do you notice?” Then “Which sentence creates that effect?”

Funnelling combines autonomy with evidence.

Question sequence can narrow productively.

68. Reverse funnels can build confidence

Begin with a concrete fact, then move to inference and evaluation.

Useful for novices who need secure entry.

Cognitive demand can be staged.

69. Too many closed questions create guessing games

Learners may infer what teacher wants from binary cues.

Follow important answers with “How do you know?”

Reasoning should surface.

70. Too many open questions can overwhelm

“Tell me everything about this text” provides little direction.

Constrain by purpose.

Openness is not the same as clarity.

Part XIII — Leading and loaded questions

71. Leading questions imply preferred answers

“Don’t you think the writer is unfair?” makes disagreement socially harder.

Use neutral wording where genuine judgement is desired.

Question framing influences evidence.

72. Loaded questions contain assumptions

“Why did the policy fail?” assumes failure.

Ask first whether the outcome qualifies as failure under defined criteria.

Premises should be inspectable.

73. Double-barrelled questions ask too much at once

“Was the source reliable and persuasive, and did it use enough evidence?” contains several judgements.

Separate them.

One question should usually have one main job.

74. Negative questions can confuse

“Don’t you disagree?” creates polarity problems.

Prefer direct wording.

Question grammar affects response accuracy.

75. False choices restrict reasoning

“Is the author angry or sad?” may exclude fear, frustration or mixed tone.

Use alternatives only when the task genuinely limits possibilities.

Question options shape thought.

76. Presupposition can teach bias

A question can embed a contested claim as background. Teach advanced learners to identify presuppositions.

They may need to challenge the question before answering.

Critical literacy includes question critique.

Part XIV — Wait time and response quality

77. Thinking needs silence

Teachers often answer their own questions too quickly. Wait long enough for retrieval or reasoning.

Question difficulty should influence wait time.

Silence can be productive.

78. Do not punish slower processing

Fast responders are not always deeper thinkers.

Use think-write-pair formats where useful.

Participation design affects whose thinking becomes visible.

79. Follow-up after partial answers

“You have identified the claim. What evidence supports it?”

Build from what is correct rather than replacing the answer wholesale.

Follow-up should be contingent.

80. Ask the whole class before naming a respondent

This can encourage everyone to think rather than only the selected student.

Use according to classroom context and safeguarding.

Question timing shapes participation.

Part XV — Follow-up questions

81. Follow-up should depend on the answer

A prepared sequence that ignores the learner’s response is an interrogation script, not diagnostic questioning.

Listen for uncertainty, unsupported claims and interesting connections.

Questions should adapt.

82. Probe evidence

“Where in the text?” “Which data point?”

Evidence follow-ups ground reasoning.

They are high-value across domains.

83. Probe mechanism

“How would that produce the result?”

Mechanism follow-ups deepen causal explanation.

They expose vague links.

84. Probe scope

“Does that apply to every case?”

Scope follow-ups reduce overgeneralisation.

Qualification becomes interactive.

85. Probe comparison

“How is this different from the earlier example?”

Comparison follow-ups build relational knowledge.

They support transfer.

86. Probe uncertainty

“Which part are you least sure about?”

Learners practise confidence calibration.

Uncertainty becomes discussable.

Part XVI — Questions for reading

87. Pre-reading questions activate knowledge

“What do you already know about this topic?” “Which terms might appear?”

Do not give away the text’s conclusion.

Activation prepares comprehension.

88. During-reading questions monitor understanding

“What changed in this paragraph?” “Who does ‘they’ refer to?”

Keep questions close to the reading process.

Monitoring prevents accumulated confusion.

89. Post-reading questions integrate

“What is the author’s central claim?” “Which evidence matters most?”

Global questions require a complete model.

They should follow adequate local comprehension.

90. Source questions expand critical reading

“Who produced this?” “What evidence can this source access?”

Source identity becomes part of comprehension.

Reading extends beyond content.

91. Cross-text questions create synthesis

“Where do the sources agree, and is their evidence independent?”

Use SETC-ETL-0049 for full development.

Questions can bridge multiple documents.

92. Reader-response questions need textual return

“How did this make you feel?” can open engagement. Follow with “What in the text produced that response?”

Personal response and analysis can connect.

Evidence keeps interpretation grounded.

Part XVII — Questions for writing

93. Planning questions define purpose

Who is the reader? What must they understand or do? What is the central message?

Questioning creates writing architecture.

Blank pages become decisions.

94. Idea-development questions add depth

Why? How? Compared with what? Under what conditions? Who is affected?

Use selectively rather than mechanically.

Questions extend thinking.

95. Evidence questions protect arguments

What evidence supports this? Is it representative? Is the source verified?

Writing and research become connected.

Claims gain accountability.

96. Revision questions re-see the draft

Which paragraph has no distinct job? Where does the reader need more explanation? What could be cut?

Questions create distance from familiar prose.

Revision becomes inquiry.

97. Editing questions target patterns

Are sentence boundaries complete? Are pronoun referents clear? Which tense controls this section?

Focused questions outperform “check grammar”.

Editing becomes procedural.

98. Audience questions test communication

What background does this reader lack? Which word may be unfamiliar? Is the call to action visible?

Questions shift the writer out of private knowledge.

Writing becomes user-centred.

Part XVIII — Questions for vocabulary and grammar

99. Vocabulary questions should move beyond definition

What part of speech? Which words commonly occur with it? What register? What is a near-synonym and difference?

Word knowledge is multidimensional.

Questions should reflect usable vocabulary.

100. Morphology questions reveal structure

What happens if we add this suffix? Which family member fits the sentence?

Questions can support induction.

Patterns reduce isolated memorisation.

101. Grammar questions should connect form and meaning

Why did the writer choose past perfect here? What time relationship does it express?

Grammar becomes communicative.

Questions avoid form-only drills.

102. Error questions should diagnose the decision

What is the subject? Which time frame? What does the pronoun refer to?

Questions expose underlying mechanism.

Correction becomes learning.

Part XIX — Questions for listening and speaking

103. Listening questions should distinguish gist and detail

First: what is the conversation mainly about? Later: what time did they agree on?

Question sequence mirrors listening purpose.

Do not demand every detail at once.

104. Segmentation questions target sound-to-word mapping

Which words did you hear in this phrase? Where might the boundaries be?

Use transcript after initial listening.

Questions diagnose connected speech problems.

105. Speaking questions can support rehearsal

What is your main point? What example will you use? What question might the audience ask?

Planning improves spontaneous control.

Do not script every answer.

106. Interaction questions create dialogue

What do you want to ask the previous speaker? What needs clarification?

Questions keep discussion responsive.

Conversation becomes collaborative thinking.

Part XX — Questions for research

107. Research questions need answerability

Too broad: “How does technology affect humanity?” More workable: “How has automated feedback changed revision behaviour among secondary students under specific conditions?”

Question scope shapes research quality.

Narrow enough to investigate, broad enough to matter.

108. Research questions need evidence routes

What sources or data could answer this? If no plausible evidence exists, the question may need reframing.

Research design begins in question design.

Questions should imply methods.

109. Distinguish descriptive, causal and evaluative questions

What happened? Why did it happen? Was it desirable? These need different evidence.

Do not answer one type with evidence suited only to another.

Question type controls source type.

110. Subquestions create a research map

Definitions, mechanism, evidence, counterargument, limitation.

Subquestions should support the central question rather than become unrelated topics.

Research architecture is question architecture.

Part XXI — Questions in interviews

111. Interview questions are live information tools

SETC-ETL-0030 owns full interview design. The question-design principle here is that live follow-ups should arise from answers.

Prepare themes, not a rigid interrogation.

Listening changes the next question.

112. Open with accessible questions

Beginning with the hardest or most sensitive question can reduce rapport.

Sequence according to purpose and ethics.

Question order affects data quality.

113. Clarify vague nouns

“Support”, “pressure”, “success”. Ask what they meant in concrete terms.

Interview data improve through definition.

Abstract language needs unpacking.

114. Ask for examples and chronology

What happened? What happened next? Can you describe a specific time?

Concrete questions support richer records.

Memory and interpretation remain distinct.

Part XXII — Questions in discussion and seminars

115. Questions can connect speakers

“How does your point relate to Maren’s example?”

Connection questions prevent serial mini-speeches.

Discussion becomes a shared model.

116. Questions can slow premature consensus

“Which criterion have we not considered?”

Good questions keep thinking open until evidence is sufficient.

Decision quality improves.

117. Questions can invite quieter participation

“Iona, would you like to add anything?” can create space, but avoid public pressure.

Participation design requires sensitivity.

Questions have social effects.

118. Questions can repair disagreement

“What do you think the other position values?” “Where is the smallest point of agreement?”

Clarification can reduce heat.

Questioning supports collaborative reasoning.

Part XXIII — Primary progression

119. Primary 1–2: retrieval and concrete clarification

Who? What? Where? Which word? Show me.

Questions should match oral language and reading level.

Concrete success builds confidence.

120. Primary 1–2: teach children to ask for help specifically

Replace “I don’t get it” with “What does this word mean?” or “Can you show me the first step?”

Help-seeking becomes precise.

Questioning supports independence.

121. Primary 3–4: why and how with evidence

Ask children to explain reasons and point to clues.

Do not reward unsupported imagination as inference.

Evidence enters questioning.

122. Primary 3–4: learner-generated quiz questions

Students create one fact question, one why question and one vocabulary question after reading.

Writing questions deepens comprehension.

They see question types.

123. Primary 5–6: metacognitive prompts

What is the task? Which strategy will you use? How will you check?

Model first, then fade.

Questions become internal scaffolds.

124. Primary 5–6: evaluation begins with criteria

Which source is more useful for this question, and why?

Criteria prevent “I like this one”.

Judgement becomes explicit.

Part XXIV — Secondary progression

125. Secondary 1–2: question ladders

Move from retrieve → clarify → infer → evaluate on one passage.

Students identify what cognitive operation changed.

Question literacy becomes explicit.

126. Secondary 1–2: diagnose own errors

After wrong answer, ask: Did I misunderstand word, sentence, question scope or evidence?

Questioning turns errors into categories.

Self-correction improves.

127. Secondary 3–4: source interrogation

Who produced this? What evidence? What is omitted? Which claim needs checking?

Questions support critical reading.

Media literacy becomes procedural.

128. Secondary 3–4: timed self-questioning

Compress classroom question routines into a short exam checklist.

Deep practice creates fast internal prompts.

Speed follows knowledge.

Part XXV — JC progression

129. JC: conceptual questioning

What definition is being assumed? Which criterion matters? What mechanism connects cause and effect?

Questions become abstract but remain evidence-linked.

Advanced reasoning is structured.

130. JC: argument stress-testing

What would a strong counterexample look like? Which evidence would change the thesis? Under what condition would the claim fail?

Questions improve qualification.

Arguments become robust.

131. JC: source synthesis questions

Why do credible sources disagree? Are they measuring the same construct? Is agreement independent?

Use SETC-ETL-0049 for full text-set work.

Questions coordinate evidence.

132. JC: self-regulated study

Which strategy produced the error? What should be practised next? How will success be tested after delay?

EEF’s 16–19 metacognition guidance supports explicit planning, monitoring and evaluating toward independence.

Questions guide study allocation.

Part XXVI — Question sequencing

133. Begin where the learner can enter

A question slightly below the target can activate knowledge and reduce cognitive overload.

Do not mistake accessibility for low expectations.

Sequence builds toward challenge.

134. Increase one dimension of difficulty at a time

Move from fact to inference while keeping vocabulary accessible.

Controlled progression reveals the weak link.

Question ladders can be diagnostic.

135. Return to evidence after abstraction

After theme or evaluation, ask which text detail supports the idea.

High-level thinking should remain anchored.

Sequence can loop.

136. Use questions to prepare later questions

A definition question can make an evaluation question answerable.

Knowledge prerequisites matter.

Good questioning has architecture.

137. End with transfer

“Would the same rule apply to this new case?”

The final question tests whether learning survives the example.

Transfer should be designed, not assumed.

Part XXVII — Question wording

138. Use concrete nouns

“What does this show?” may be unclear when several visuals are present.

Name the chart or sentence.

Reference clarity improves answers.

139. Avoid unnecessary vocabulary difficulty

If inference is the target, obscure wording in the question can become a competing barrier.

Questions should be simpler than the intended thinking unless vocabulary itself is being tested.

Language load matters.

140. Put the main operation early

“Compare the two sources…” tells learners what to do before details.

Long question stems can hide the task.

Frontloading supports execution.

141. State response constraints clearly

One reason, two examples, 50 words, choose all that apply.

Answer scope is part of question design.

Constraints should serve the learning goal.

142. Avoid unnecessary negatives

“Which option is not…” increases processing and error risk. Use negative forms only when exclusion is the intended skill and highlight them appropriately.

Question wording should not create accidental traps.

Difficulty should be meaningful.

Part XXVIII — Question quality failures

143. Failure: rhetorical question presented as inquiry

The teacher already has the only acceptable answer and punishes alternatives.

Repair: either state the teaching point directly or create genuine criteria for evaluation.

Authenticity matters.

144. Failure: question too broad

“What do you think?” produces random scope.

Repair: specify object and operation.

Openness needs structure.

145. Failure: question too narrow for target

A literature lesson asks only dates and names.

Repair: include inference and evaluation after prerequisites.

Question sets should represent the learning goal.

146. Failure: answer hidden in grammar

Leading tags and obvious cues convert retrieval to recognition.

Repair: remove answer signals.

Protect cognitive demand.

147. Failure: double-barrelled question

Student answers one part and appears incomplete.

Repair: separate operations.

One job per question improves diagnosis.

148. Failure: no wait time

Teacher rephrases or answers immediately.

Repair: allow thinking according to difficulty.

Silence can be instructional.

149. Failure: follow-up ignores answer

Question sequence runs like a script.

Repair: listen for evidence, uncertainty and misconception.

Dialogue needs contingency.

150. Failure: endless probing without teaching

Learner lacks prerequisite knowledge but receives more questions.

Repair: explain or model, then resume.

Questioning is not a substitute for instruction.

151. Failure: only confident students answer

Questioning appears successful while most learners remain invisible.

Use think time, writing, pairs or sampling strategies.

Participation affects diagnostic validity.

152. Failure: metacognitive questions become generic rituals

“What strategy will you use?” is repeated regardless of subject.

Repair: tie prompts to actual content decisions.

Metacognition belongs inside domain learning.

Part XXIX — Resident learners

153. Maren asks broad questions

She likes open inquiry and asks “What do you think?” before learners have a clear object.

Her intervention is operation-plus-object: infer what? evaluate by which criterion?

Openness becomes purposeful.

154. Iona asks excellent questions but too many

Learners become dependent on her prompts.

Her intervention is fade-and-transfer: after three guided questions, student must generate the fourth.

Scaffolding begins to disappear.

155. Leonie answers her own questions

She dislikes silence and rephrases after one second.

Her intervention is planned wait time and written think moments.

More learner thinking becomes visible.

Part XXX — A 120-minute questioning tutorial

156. Core studio

1. 10 minutes: identify the learning job.
2. 15 minutes: write retrieval, clarification and diagnostic questions.
3. 15 minutes: write inferential and evaluative questions with evidence criteria.
4. 15 minutes: build a question sequence from entry to transfer.
5. 15 minutes: role-play responses and contingent follow-ups.
6. 15 minutes: design planning, monitoring and evaluation prompts.
7. 15 minutes: remove leading assumptions and double barrels.
8. 10 minutes: fade the sequence so learners generate questions.
9. 10 minutes: test on a fresh topic.

The final goal is not a long list of clever questions. It is a compact sequence that makes the desired thinking happen.

Part XXXI — Practice labs

157. Practice Lab: classify the question

Students sort twenty questions into retrieval, clarification, inference, evaluation, metacognitive and diagnostic categories.

Ambiguous cases are discussed.

Question functions become explicit.

158. Practice Lab: same topic, different thinking

Take one paragraph and write five question types about it.

Compare how the required mental operation changes.

The text stays constant; cognition changes.

159. Practice Lab: leading-question repair

Rewrite ten leading or loaded questions neutrally.

Identify the hidden assumption in each.

Question bias becomes visible.

160. Practice Lab: double-barrel split

Break compound questions into ordered subquestions.

Choose which should come first.

Sequence and diagnosis improve.

161. Practice Lab: wait-time observation

One student asks, another answers, a third times silence and records interruptions.

Repeat with longer wait.

Participants compare answer quality.

162. Practice Lab: follow-up tree

Create three possible learner answers—correct, partial, misconception—and design a different follow-up for each.

Questioning becomes adaptive.

Contingency is planned.

163. Practice Lab: least-help-first

Design prompt → clue → model → solution for one task.

Test where different learners become successful.

Support level becomes diagnostic.

164. Practice Lab: evidence ladder

Start with “What do you think?” and revise into “What do you think, and which two details support it?”

Then add “What detail might challenge it?”

Questions progressively discipline evidence.

165. Practice Lab: metacognitive trio

For one task, write one planning, one monitoring and one evaluation question.

Run the task and see whether each prompt was useful at its stage.

Timing matters.

166. Practice Lab: student-generated questions

After reading, each learner writes a retrieval, inference and evaluation question for peers.

Peers rate answerability and evidence.

Creating questions deepens task understanding.

167. Practice Lab: question-to-strategy map

Match question types to likely strategies: scan, close read, compare, verify source, brainstorm, calculate.

Students see that questions imply methods.

Task decoding improves.

168. Practice Lab: diagnostic interview

One student gives a wrong comprehension answer. Tutor may ask only five questions to locate the failure.

No explanations until diagnosis.

Question efficiency is tested.

169. Practice Lab: ambiguity test

Give vague questions with pronouns and unclear scope. Students list plausible interpretations.

Rewrite so one intended task remains.

Question clarity becomes user-tested.

170. Practice Lab: answerability test

Some questions cannot be answered from the supplied text. Students identify whether outside research is required.

Question-source fit becomes explicit.

Good questions respect evidence boundaries.

171. Practice Lab: research question narrowing

Start with “How does social media affect people?” Narrow by population, outcome, setting and time.

Students explain which evidence each version would require.

Scope and method connect.

172. Practice Lab: Socratic stress test

Take one claim and ask definition, assumption, evidence, alternative, implication and consistency questions.

Stop if prerequisite knowledge is missing.

Socratic questioning remains humane.

173. Practice Lab: question bias

Write two prompts on the same issue, one leading and one neutral. Compare generated answers.

Students see framing effects.

Question wording influences data.

174. Practice Lab: AI-generated questions audit

Ask an AI system for practice questions, then classify cognitive level, ambiguity and answerability.

Reject weak questions.

Automation still needs instructional judgement.

175. Practice Lab: quiz balance

Audit a ten-question quiz. How many test retrieval versus application or reasoning?

Adjust to match learning goals.

Assessment samples cognition.

176. Practice Lab: self-questioning card

Learners design five internal prompts for one recurring task.

Use for three weeks, then remove the card.

External scaffolding becomes internal.

Part XXXII — Transfer tests

177. Transfer Test: new subject

Use the same diagnostic-question method in science or mathematics explanation.

Does the learner identify task, evidence and strategy?

Question architecture should transfer.

178. Transfer Test: no teacher prompts

Give a difficult task and ask the learner to write the questions they need before starting.

Compare with previous teacher prompts.

Independence becomes observable.

179. Transfer Test: time pressure

Learner has thirty seconds to choose three self-questions before answering an exam item.

Deep routines must compress.

Speed grows from practice.

180. Transfer Test: misleading question

Give a question containing a false presupposition. Learner should challenge the premise rather than accept it.

Critical question literacy includes refusing bad questions.

Answering is not always compliance.

181. Transfer Test: question without enough evidence

Learner must say what additional source is needed.

Unanswerability can be correct.

Good inquiry respects evidence limits.

182. Transfer Test: unfamiliar wording

Paraphrase a known question type without familiar command words.

Can the learner identify the cognitive job?

Concept survives surface change.

183. Transfer Test: peer questioning

Learner uses follow-up questions in live discussion without teacher prompts.

Questions should respond to actual answers.

Adaptive questioning transfers socially.

184. Transfer Test: delayed self-questioning

One week later, remove the prompt card. Does the learner still plan, monitor and evaluate?

EEF’s goal of internal scaffolding is tested.

Durability matters.

Part XXXIII — Personal toolkit

185. The question-purpose card

What thinking job do I need: retrieve, clarify, diagnose, infer, evaluate, compare, explain, plan, monitor or reflect?

Choose type before wording.

Purpose controls design.

186. The clarity card

One main job? concrete referent? necessary context? no hidden assumption? response constraint clear?

Use before asking.

Question grammar should serve thinking.

187. The evidence card

Can the answer be supported? What source or text could justify it?

Do not ask pseudo-evidence questions.

Answerability matters.

188. The follow-up card

What if answer is correct, partial, wrong, uncertain or surprising?

Prepare branching possibilities for high-value questions.

Dialogue becomes adaptive.

189. The metacognitive card

Before: What is the task? During: Is this working? After: What will I change next time?

Keep prompts subject-specific.

Planning-monitoring-evaluation becomes routine.

190. The fade card

Which question can the learner now ask themselves? Remove one external prompt.

Scaffolding should shrink.

Independence is designed.

191. The bias card

Am I implying a preferred answer? Assuming disputed facts? Restricting choices artificially?

Neutralise where genuine inquiry is desired.

Questions carry frames.

192. The wait card

How much processing does this question require? Am I allowing it?

Silence is part of questioning.

Timing shapes evidence.

193. The transfer card

Can the same thinking job be asked in a new topic or genre?

Use fresh contexts.

Question sequences should build portable cognition.

194. What progress looks like

The learner increasingly recognises that different questions perform different cognitive jobs, writes clearer and less leading questions, sequences them according to prerequisites, waits for genuine responses, uses evidence-sensitive follow-ups, chooses metacognitive prompts that support planning, monitoring and evaluation and gradually generates the next useful question without teacher prompting.

The deepest shift is from being a person who waits to be questioned to a person who can interrogate a text, a claim, a strategy and their own learning deliberately.

Where this tutorial connects inside SETC

• English Tutorial Lesson No.0007 — Curiosity to Research
• English Tutorial Lesson No.0030 — Question to Interview
• English Tutorial Lesson No.0008 — Practice to Independence
• English Tutorial Lesson No.0046 — Criteria to Self-Assessment
• English Tutorial Lesson No.0049 — One Text to a Text Set

External research and teaching references

• Education Endowment Foundation — Supporting Independence Through Questioning
• EEF — Promoting Metacognitive Talk in the Classroom
• EEF — Metacognition and Self-Regulated Learning
• EEF — Developing Independent Learners: Metacognition and Self-Regulation (2026)

Final tutorial challenge

Give the learner one unfamiliar 600-word text, one small dataset and one short source description. They have ten minutes to design a sequence of eight questions that would teach another learner to understand and evaluate the material: at least one retrieval question, one clarification question, one inference question, one evidence question, one evaluative question, one diagnostic branch, one metacognitive monitoring question and one transfer question. They must then role-play two different answers and change the follow-up accordingly.

The final judgement is not based on how sophisticated the questions sound. It is based on whether each question creates a useful next act of thinking, whether the sequence responds to evidence and whether the learner can eventually generate these prompts for themselves.

English Learning Library · Primary English · Secondary English · Advanced English / General Paper

Return to practice, feedback and tutorial design: English Practice, Model Answers and Feedback.