Identify does not mean “write everything you know about the thing you recognise.”
It means recognise, select or name the requested item accurately enough that the examiner can see you have located the right feature, process, pattern, term, source element, variable, relationship or category.
The answer is often short.
The thinking behind it may not be.
A learner may have to distinguish between two visually similar structures, separate cause from consequence, recognise a source message rather than its topic, choose the dependent rather than independent variable, identify an anomaly rather than the overall trend, or select one technical term from several near-neighbours.
The skill is recognition with discrimination.
Current official command-word guidance supports that boundary. Cambridge International defines Identify as name/select/recognise. AQA commonly defines it as name or otherwise characterise. Exact expectations still depend on the subject syllabus and the wording around the command.
This page owns one narrow examination-performance job: Identify questions across subjects—how to recognise the required object, locate it in knowledge or evidence, distinguish it from plausible alternatives, give the answer in the accepted form and stop before the response expands into Definition, Description, Explanation, Discussion or Justification.
It does not replace How to Answer State Questions in Exams. State owns the clear delivery of required information. Identify owns recognition and selection of the correct item. It also does not replace Definitions in Exams, which owns precise meaning.
The 30-Second Identify Route
- Find the target. What exactly must be identified?
- Find the evidence field. Knowledge, graph, source, diagram, table, passage, map or calculation?
- Generate the nearest candidates.
- Use the decisive feature. What distinguishes the correct candidate?
- Give the accepted name, label or characterisation.
- Check polarity and scope.
- Stop.
A compact universal frame is:
target → candidates → decisive feature → selection → stop.
Ben Recognises the Topic but Identifies the Wrong Thing
Ben is given a short passage and asked:
Identify the writer’s main concern.
The passage is about technology in schools.
Ben writes:
Technology.
He has identified the topic.
He has not identified the concern.
The stronger answer is:
The writer is concerned that constant device use is weakening students’ concentration.
The target noun controls the answer.
Identify Is Not Just Recognition
Recognition says:
I have seen this before.
Identification says:
I know which thing this is, and I can distinguish it from nearby alternatives.
That difference explains why students can feel familiar with a diagram, text feature or formula and still select the wrong answer.
The Target-Object Test
Underline what follows Identify:
- Identify the process.
- Identify one feature.
- Identify the variable.
- Identify the anomaly.
- Identify the writer’s view.
- Identify the source message.
- Identify the pattern.
- Identify the stakeholder.
- Identify the method.
- Identify the error.
The noun after the command is the object class.
A correct answer from the wrong class is still wrong.
Identify vs State
These commands often produce answers of similar length.
The difference is functional.
- Identify: recognise/select/name the correct item.
- State: express the required information clearly and briefly.
Question:
Identify the process shown in the diagram.
The learner must recognise the process.
Question:
State one condition required for the process.
The learner must give a required condition.
The final answers may both be short, but the route is different.
Identify vs Name
Name usually asks for the accepted term or title.
Identify can be broader.
It may ask the learner to:
- name a process;
- select an option;
- recognise a feature;
- characterise a pattern;
- locate a source element;
- distinguish one category from another.
If the question says “Name the structure,” one technical term may be enough.
If it says “Identify the feature that shows the structure is specialised,” the answer may need a characteristic rather than a name.
Identify vs Define
Identify gives the item.
Define gives its precise meaning.
Identify:
Diffusion.
Define:
Give the exact course definition of diffusion.
Do not spend definition-level time when identification is enough.
Identify vs Describe
Identify selects the feature.
Describe develops the feature.
Identify:
The graph contains a plateau.
Describe:
The value rises rapidly at first, then remains approximately constant from x = 20 onward.
The first labels the pattern.
The second reconstructs it.
Identify vs Explain
Identify says what it is.
Explain says why or how it occurs.
Identify:
The anomaly is the reading at 40°C.
Explain:
The anomaly may have resulted from measurement error or an uncontrolled variable…
Do not add a speculative cause to an Identify question.
Identify vs Suggest
Identify selects something supported by available evidence.
Suggest generates a plausible response when more than one valid answer may exist.
Identify:
The controlled variable is temperature.
Suggest:
One improvement would be to use a temperature-controlled water bath.
One recognises.
One proposes.
Identify vs Infer
Some subjects use inference tasks even when the word “infer” is not part of the formal command list.
Identify stays close to what can be recognised directly.
Inference moves from evidence to a conclusion not explicitly stated.
Example:
- Identify: the character repeatedly checks the door.
- Infer: the character may feel threatened.
Do not turn direct identification into unsupported inference.
The Candidate-Set Method
When identification is difficult, generate a small candidate set.
Question: identify the type of relationship shown.
Possible candidates:
- positive association;
- negative association;
- no clear association;
- direct proportionality;
- inverse proportionality.
Then use the evidence to eliminate.
This is safer than jumping to the first familiar label.
The Decisive-Feature Test
Two candidates can look similar until one decisive feature separates them.
Examples:
- speed vs velocity → direction;
- accuracy vs precision → closeness to true value vs closeness among repeated values;
- revenue vs profit → costs accounted for;
- correlation vs causation → causal inference;
- message vs purpose → what the source says vs what it is trying to achieve;
- mean vs median → calculation structure and sensitivity to extremes;
- rotation vs reflection → orientation and transformation parameters.
The fastest identification often comes from asking:
What single feature would rule out the nearest alternative?
The Negative-Evidence Test
Identification can depend on what is absent.
A graph may not pass through the origin, so “directly proportional” is ruled out.
A source may praise a policy but never ask readers to act, so “call to action” may be the wrong purpose.
A method may vary temperature but never measure temperature, so temperature is unlikely to be the dependent variable.
What is missing can eliminate a candidate.
The Local-Evidence Rule
When the question gives a graph, source, diagram, table or passage, identify from the local evidence before importing memory.
A familiar pattern can tempt the learner to answer from expectation rather than what is actually shown.
The paper may deliberately present an exception.
Evidence first.
Memory second.
Identify From a Graph
Graph Identify questions can ask for:
- trend;
- turning point;
- anomaly;
- plateau;
- maximum/minimum;
- intercept;
- relationship type;
- outlier;
- period of fastest change;
- crossing point.
Read axes, scale and legend before identifying the pattern.
Identify an Anomaly
An anomaly is a point that does not fit the general pattern closely enough to deserve separate attention.
Do not identify the highest point automatically as anomalous.
A maximum can fit the trend perfectly.
Compare the point with neighbouring and overall behaviour.
Identify a Trend
Possible trend identities include:
- increasing;
- decreasing;
- stable;
- cyclical;
- fluctuating;
- rise then plateau;
- rise then fall;
- no clear trend.
Choose the label that captures the whole requested range.
Identify a Variable
In experimental work:
- independent variable → deliberately changed;
- dependent variable → measured response;
- control variable → kept constant to protect the comparison.
Do not identify a variable by where it appears in the table.
Identify it by its experimental role.
Identify a Control Variable
A variable can be relevant without being a control variable.
Ask:
Could this factor affect the dependent variable, and is it meant to remain constant across trials?
If yes, it is a plausible control candidate.
Identify a Scientific Observation
Observation stays close to what is seen or measured.
“A white precipitate forms” is observation.
“A new compound formed because ions reacted” is explanation.
Recognise the level the question is asking for.
Identify a Process
A process should be recognised from its defining pattern, not from one loose association.
If particles spread from a region of higher concentration toward lower concentration, diffusion is a candidate.
If water moves across a partially permeable membrane under the course definition, osmosis may be the correct process.
Near-neighbour discrimination matters.
Identify a Structure From Function
Sometimes the visible feature is not named, but its function is described.
The learner must map function back to structure.
This requires relational knowledge:
function → characteristic → structure.
Rote label recognition may fail if the picture looks unfamiliar.
Mathematics: Identify the Transformation
Ryan checks defining invariants.
- translation → same orientation and size, shifted by a vector;
- rotation → turn about a centre;
- reflection → mirror relation across a line;
- enlargement → scale factor relative to a centre.
Do not identify from appearance alone if the diagram is ambiguous.
Mathematics: Identify the Graph Type
Possible graph identities include:
- linear;
- quadratic;
- cubic;
- reciprocal;
- exponential;
- trigonometric;
- piecewise;
- step function.
Use shape plus defining features such as intercepts, asymptotes, symmetry or rate pattern.
Mathematics: Identify the Error
Error identification asks for the first incorrect move, not merely the wrong final answer.
Possible errors:
- sign change;
- distribution error;
- invalid cancellation;
- wrong inverse operation;
- domain violation;
- rounding too early;
- equating non-equivalent expressions.
Find where the solution first diverges from valid reasoning.
Mathematics: Identify the Relevant Theorem or Rule
The learner may know many theorems.
The question is which one matches the structure.
Look for trigger features:
- right triangle;
- parallel lines;
- cyclic quadrilateral;
- similar shapes;
- arithmetic sequence;
- independent events;
- normal distribution assumptions where taught.
Recognition must be conditional, not keyword-only.
English: Identify the Writer’s View
Ben distinguishes topic from proposition.
Topic:
School uniforms.
View:
The writer believes uniforms reduce visible social differences but should allow greater flexibility.
A view says what the writer thinks about the topic.
English: Identify the Tone
Tone labels can be near-neighbours.
- concerned vs alarmist;
- critical vs hostile;
- admiring vs reverential;
- cautious vs uncertain;
- humorous vs mocking;
- detached vs indifferent.
Use the textual evidence to discriminate rather than choosing the first familiar adjective.
English: Identify a Language Feature
Possible features include:
- metaphor;
- simile;
- personification;
- rhetorical question;
- repetition;
- contrast;
- imperative;
- modal verb;
- semantic field.
Identify from the defining structure, not because a sentence “sounds descriptive.”
English: Identify a Structural Feature
- flashback;
- cyclical structure;
- shift in viewpoint;
- delayed revelation;
- contrast between sections;
- chronological progression;
- refrain or repeated motif.
Do not confuse a structural feature with its effect.
The feature is identified first.
Analysis explains what it does.
Humanities: Identify a Source Message
A source message is what the source communicates.
It is usually a proposition.
The source argues that citizens should support the government’s recovery programme.
“Recovery” is only a topic.
Humanities: Identify Source Purpose
Purpose asks what the source is trying to achieve.
- persuade;
- reassure;
- mobilise;
- justify;
- criticise;
- inform;
- discourage;
- legitimise.
Identify purpose from audience, context, message and form.
Do not assume every public source exists “to inform.”
History: Identify a Cause, Trigger or Consequence
Clara separates causal roles.
- cause → contributes to the event;
- trigger → helps determine immediate timing;
- consequence → follows from the event.
One fact can be historically important and still belong to the wrong causal category.
Geography: Identify the Distribution Pattern
Possible identities include:
- clustered;
- dispersed;
- linear;
- coastal;
- central;
- peripheral;
- random;
- concentrated around nodes.
Use location evidence to select the pattern.
Geography: Identify a Landform or Process
Recognise from defining characteristics, not memorised picture shape alone.
If an unfamiliar diagram rotates or simplifies the landform, relational knowledge still works.
Economics: Identify the Market Change
A graph can show:
- movement along demand;
- shift of demand;
- movement along supply;
- shift of supply;
- new equilibrium;
- shortage;
- surplus.
Do not identify every rightward movement as “higher demand.”
Check whether the curve moved or the point moved along the curve.
Economics: Identify the Type of Inflation or Unemployment
Use the mechanism/evidence provided.
Do not classify merely from the outcome label.
Several mechanisms can produce rising prices or unemployment.
Business: Identify the Stakeholder
Identify who is affected or who holds the interest described.
“Wants secure employment and higher pay” points toward employees.
“Wants reliable supply contracts and prompt payment” points toward suppliers.
Map interest to stakeholder.
Business: Identify the Objective
Possible objectives include:
- profit;
- survival;
- growth;
- market share;
- customer satisfaction;
- social/environmental objectives.
Use case evidence rather than assuming every business prioritises profit at every stage.
Computer Science: Identify the Data Type
Map the stored information to the type.
- true/false → Boolean;
- whole count → integer;
- single symbol → character;
- text sequence → string;
- fractional numeric value → suitable real/float representation depending on syllabus.
Use the terminology expected by the course.
Computer Science: Identify the Security Threat
Recognise the threat from the attack mechanism.
- fake credential request → phishing;
- malicious software encrypting files for payment → ransomware;
- guessing many passwords automatically → brute force;
- intercepting communications → interception/man-in-the-middle where relevant.
Do not identify only from one buzzword if the scenario points elsewhere.
Identify a Study Weakness
Aisha looks at the evidence.
If a learner recognises answers with notes but cannot retrieve them unaided, the weakness is not necessarily understanding.
It may be retrieval stability.
If the learner recalls formulas but selects the wrong one for unfamiliar questions, the weakness may be method recognition or transfer.
Identify the first failing operation rather than applying a generic label.
Identify an Exam Error
Exam errors can be classified:
- knowledge;
- retrieval;
- question recognition;
- method selection;
- execution;
- notation;
- time allocation;
- checking;
- command-word mismatch;
- answer-transfer error.
Do not call every lost mark “careless.”
Identification gives repair a target.
The First-Divergence Principle
When identifying an error in a chain, find the first step that became invalid.
A later wrong answer may merely inherit an earlier mistake.
Repair the first divergence.
Do Not Identify the Topic When Asked for the Message
Topic:
Public transport.
Message:
The city should invest more in public transport to reduce congestion.
The second says something about the first.
Do Not Identify the Evidence When Asked for the Conclusion
Evidence:
Y rises from 10 to 30 as X increases.
Conclusion:
Y is positively associated with X over the observed range.
Know the object class.
Do Not Identify the Cause When Asked for the Effect
Question:
Identify one effect of higher interest rates.
Answer:
Borrowing may fall.
“Interest rates rise” merely repeats the cause.
Do Not Identify a Label From One Superficial Cue
A text containing “like” is not automatically a simile.
A graph that rises is not automatically directly proportional.
A question using “because” is not automatically an Explain question if it asks the student to identify a reason from the source.
Use defining criteria, not trigger words alone.
Do Not Give Three Candidates “Just in Case”
If one item is requested, multiple incompatible answers create uncertainty.
Choose.
If the evidence genuinely does not distinguish, state that only when the task permits it.
Do Not Add a Definition After a Correct Identification
Question:
Identify the type of sampling used.
Answer:
Stratified sampling.
A full definition may be unnecessary and can introduce error.
Do Not Explain After a Correct Identification
Question:
Identify the anomaly.
Answer:
The point at x = 7.
Why it is anomalous may be useful only if the command asks for explanation.
Do Not Confuse Similarity With Identity
Two things can look similar and belong to different categories.
Identification requires the defining feature.
This is why near-neighbour practice is more valuable than repeatedly identifying obvious examples.
The Identify Error Taxonomy
- Wrong-object failure: topic given when message requested, cause given when effect requested.
- Near-neighbour failure: similar technical term selected.
- Superficial-cue failure: one keyword triggers the wrong label.
- Evidence-ignore failure: memory overrides local evidence.
- Polarity failure: wrong item selected because NOT/LEAST/EXCEPT is missed.
- Multiplicity failure: one item requested, several incompatible answers given.
- Over-answer failure: definition or explanation added unnecessarily.
- Under-specification failure: answer too broad to identify uniquely.
- First-divergence failure: symptom identified instead of originating error.
- Category-level failure: answer given at wrong abstraction level.
The Near-Neighbour Drill
Pairs and triplets should be trained together:
- accuracy / precision / reliability;
- mean / median / mode;
- revenue / profit / cash flow;
- message / purpose / audience;
- cause / trigger / consequence;
- association / proportionality / causation;
- translation / rotation / reflection;
- tone / mood / viewpoint;
- independent / dependent / control variable.
Ask what decisive feature separates them.
The Elimination Drill
Give four candidates.
The learner must eliminate three using evidence.
This trains identification as discrimination rather than guesswork.
The Local-Evidence Drill
Use unfamiliar diagrams, altered contexts and non-standard examples.
The learner may not answer from a memorised picture.
They must point to the decisive evidence in the stimulus.
The Wrong-Object Drill
Mix prompts asking for:
- topic;
- message;
- purpose;
- audience;
- tone;
- cause;
- effect;
- variable;
- unit;
- condition.
The learner names the requested object class before answering.
The Identify–State Switch Drill
Round 1:
Identify the process shown.
Round 2:
State one condition required for the process.
The first recognises.
The second retrieves a fact.
The Identify–Describe Switch Drill
Identify: The pattern is clustered.
Describe: Most sites are concentrated around the river and transport junctions, with few in the western region.
This trains resolution control.
The Identify–Explain Switch Drill
Identify: The anomaly is the reading at 50°C.
Explain: One possible reason is…
The switch teaches where identification stops.
The Timed Identify Drill
target → candidates → decisive feature → answer → stop.
Identification should become fast after the discriminating features are learned.
The Final-Quarter Identify Drill
Late-paper fatigue increases near-neighbour confusion.
Place mixed Identify questions near the end of timed practice and score:
- target object correct;
- candidate set appropriate;
- decisive evidence noticed;
- near-neighbour rejected;
- polarity correct;
- answer form accepted;
- no unnecessary explanation.
The First-Divergence Review
- Did I identify the target object correctly?
- Did I look at the right evidence?
- Did I generate the right candidate class?
- Did I use the decisive feature?
- Did I eliminate the nearest alternative?
- Did I respect polarity and scope?
- Did I stop after identification?
Primary Learners: Identify Means “Which One Is It?”
Young learners can begin with visual and concrete discrimination.
- identify the shape;
- identify the animal;
- identify the verb;
- identify the larger number;
- identify the object that floats.
The teaching goal is to notice the feature that makes the choice correct.
Lower Secondary: Add Near-Neighbour Discrimination
Students should become reliable at distinguishing:
- variable roles;
- graph relationships;
- text features;
- source message/purpose;
- cause/effect;
- technical vocabulary.
Upper Secondary: Add Evidence and Boundary Control
Identification should increasingly rely on:
- defining criteria;
- local evidence;
- scope;
- domain;
- negative evidence;
- near-neighbour elimination;
- claim strength.
JC, IB, IP and Advanced Learners: Identify Becomes Model Selection
Advanced learners may need to identify:
- the relevant theorem;
- the correct model;
- the first invalid assumption;
- the dominant mechanism;
- the appropriate evidence type;
- the key constraint;
- the strongest interpretation supported by data.
The answer can still be short even when the discrimination is sophisticated.
Parents: Ask “What Feature Makes You Sure?”
A useful home prompt is:
What feature tells you it is that answer rather than the next most likely one?
This develops discrimination rather than guessing confidence.
Tutors: Train the Wrong Alternatives
Do not show only obvious positive examples.
Show:
- close non-examples;
- ambiguous-looking diagrams;
- similar technical terms;
- sources with different message and purpose;
- graphs that rise but are not proportional;
- variables whose roles change between experiments.
Identification improves when the learner understands why alternatives fail.
The Three-Student Identify Comparison
Ben recognises the topic but answers the wrong object.
Aisha selects the right category but adds two incompatible alternatives.
Clara notices the decisive feature, selects one answer and stops.
Compare:
- Who read the target correctly?
- Who used evidence?
- Who eliminated near-neighbours?
- Who over-answered?
- Who used the shortest complete response?
The Identify Dashboard
- target object recognised;
- evidence field located;
- candidate class correct;
- decisive feature found;
- near-neighbour eliminated;
- polarity respected;
- scope respected;
- accepted term/form used;
- one clear selection;
- no unnecessary explanation.
The Independence Test
Identify performance is independent when the learner can:
- parse the target object instantly;
- locate decisive evidence;
- generate plausible alternatives;
- distinguish near-neighbours;
- use local evidence rather than expectation;
- respect polarity and scope;
- give the accepted name/characterisation;
- stop without adding unsupported inference.
The Red–Amber–Green Audit
Red: learner answers the wrong object, confuses near-neighbours, guesses from superficial cues, ignores local evidence or lists several incompatible candidates.
Amber: recognition is usually correct, but difficult distractors, unfamiliar representations, polarity or closely related terms create inconsistency.
Green: the learner identifies the target class, notices the decisive feature, rejects the nearest alternative, gives the accepted answer and stops.
The Eleven-Question Audit
- What exactly must I identify?
- Where is the relevant evidence?
- What answer class am I choosing from?
- What are the nearest alternatives?
- What feature distinguishes the correct answer?
- Is there evidence against my first choice?
- Is the wording negative or comparative?
- Am I answering at the correct abstraction level?
- Does the evidence support this identification?
- Have I given one clear accepted answer?
- Can I stop now?
What Mastery Looks Like
Clara sees the question.
She does not ask what topic it reminds her of.
She asks what object the examiner wants identified.
She looks at the local evidence.
Two candidates come to mind.
She finds the feature that separates them.
She selects one.
She does not add a definition nobody requested.
She does not explain a mechanism nobody requested.
She moves on.
Deep Layer: Identification Is a Classification Decision
At a deeper level, every Identify question is a classification problem.
The learner receives an input—a graph, phrase, diagram, data pattern, scenario, source, method, mathematical object, error or behaviour—and must assign it to the correct class.
That class might be “positive association,” “control variable,” “metaphor,” “rotation,” “cost-push inflation,” “supplier,” “phishing,” “retrieval failure” or “timing error.”
The visible answer can be one word. The hidden operation is:
features observed → candidate classes → decision boundary → selected label.
Decision Boundaries
A decision boundary is the feature that separates two nearby classes.
- direct proportion vs general positive association → constant ratio / graph through origin where relevant;
- profit vs revenue → whether costs have been deducted;
- message vs purpose → what is communicated vs what the source aims to achieve;
- rotation vs reflection → turning around a centre vs mirror relation;
- independent vs dependent variable → changed by researcher vs measured response;
- tone vs mood → writer/speaker voice vs atmosphere/effect where the course uses that distinction.
Students become much faster when they learn the boundary rather than memorising isolated definitions.
Prototype Recognition vs Rule-Based Recognition
Some learners identify by prototype. They remember a familiar example and ask whether the new item “looks like it.”
This works on easy questions and becomes fragile when the representation changes.
Rule-based recognition asks instead:
What defining property must be present?
A rotated diagram, unusual context or unfamiliar wording then becomes less threatening because the learner is matching structure rather than pictures.
False Positives and False Negatives
- False positive: identify a feature that is not actually present.
- False negative: fail to identify a feature that is present.
A learner sees a rising line and labels it “directly proportional.” That can be a false positive if the graph does not pass through the origin or the ratio is not constant.
Another learner sees an outlier but ignores it because the overall trend is strong. That can be a false negative.
Training should target both.
Sensitivity and Specificity as a Learning Analogy
- high sensitivity → notice most real examples;
- high specificity → reject most lookalikes.
A learner who calls every figurative phrase a metaphor may have high sensitivity for figurative language and poor specificity for the exact device.
A learner who refuses to call anything an anomaly unless it is wildly distant may have high specificity but poor sensitivity.
Expert identification balances detection and discrimination.
Confidence Is Not Evidence
Recognition can feel fluent. That feeling is not proof.
A familiar-looking graph, phrase or diagram can activate the wrong label very quickly.
Before committing, ask for one decisive feature:
What makes this answer true rather than merely familiar?
Identification Under Uncertainty
Not every stimulus is perfectly clean. A graph may be noisy. A source may contain mixed tone. A business case may support more than one objective. A historical source may have more than one plausible purpose.
- identify the strongest-supported option;
- avoid absolute wording if the evidence is mixed;
- use the requested scope to narrow the decision;
- do not manufacture certainty by adding extra labels.
The Evidence-Threshold Test
Ask how much evidence is enough to assign the label.
One strong defining feature can be enough. One weak surface cue may not be.
- a single “like” does not automatically make a simile;
- a line rising does not prove direct proportionality;
- one angry word does not necessarily make the whole passage hostile;
- a company making money does not identify profit growth unless costs and period are clear;
- one unusual data point is not automatically an anomaly if the wider model predicts it.
Identification Can Be Multi-Cue
Some labels require several cues together.
A source purpose may be inferred from audience, message, context, language, medium and requested action.
A scientific process may require direction of movement, membrane condition, particle type and concentration relation.
A mathematical transformation may require size, orientation and centre/line/vector relations.
When one cue is insufficient, combine cues.
Identification Can Be Hierarchical
- broad class → function;
- subclass → quadratic function;
- specific form → upward-opening quadratic with two real roots.
The question decides the level.
The Abstraction-Level Test
Too broad: “Business” when asked to identify a stakeholder.
Too narrow: “the supplier of corrugated cardboard boxes” when the expected category is simply “supplier.”
Match the abstraction level to the assessment language.
Identification From Partial Evidence
Some questions deliberately provide only part of the usual pattern.
The learner must identify what is justified by the available cues without inventing the rest.
If a source shows criticism but no requested action, identify a critical viewpoint rather than assuming the purpose is to persuade readers to protest.
If a graph shows an association but the origin is outside the plotted range, do not identify direct proportionality from a partial line alone.
Identification From Noisy Evidence
Noise means irrelevant variation around the signal.
In data, small fluctuations may sit around an overall trend. In text, one humorous line may appear inside an otherwise serious passage. In a business case, one month may behave differently from the longer pattern.
Identify the dominant structure unless the question specifically targets the exception.
Signal vs Noise
signal changes the classification; noise does not.
If removing one small fluctuation leaves the same label, it may be noise. If removing one feature changes the category, it is likely signal.
Identification From Contradictory Cues
Sometimes cues point in different directions.
A source uses polite language but attacks an opponent’s competence. A graph mostly rises but contains a large downturn. A company reports revenue growth but falling cash flow.
Do not force a simplistic label if the object itself is mixed. Use the scope in the question to choose the relevant evidence window.
Identification and Context
The same feature can mean different things in different contexts.
A negative number can mean debt, temperature below a reference point, direction, loss or a coordinate. A repeated word can be emphasis, motif, lexical cohesion or simply unavoidable terminology.
Context determines category.
Identification and Function
When appearance is unfamiliar, function can reveal identity.
What does this thing do in the system?
A component that stores temporary working data may be identified from function even if the diagram symbol is unfamiliar.
A paragraph that acknowledges an opposing position before responding may function as a counterargument section even if it lacks an obvious heading.
Identification and Exclusion
Sometimes the quickest route is to identify what the item cannot be.
- not reflection because orientation is preserved and no mirror line fits;
- not direct proportion because the line does not pass through the origin;
- not the dependent variable because it is deliberately changed;
- not profit because costs have not been deducted;
- not purpose because the answer states content rather than intended effect.
Elimination is legitimate reasoning, especially under pressure.
Identification and Counterexamples
A single counterexample can destroy an over-broad identification.
If a proposed “always increasing” function falls anywhere in the domain, the label fails.
If a proposed “formal tone” passage contains mostly conversational slang and direct banter, the label may need revision.
Counterexample testing is a powerful final check.
Worked Identification 1: Positive Association vs Direct Proportion
A graph rises as x increases but does not pass through the origin.
Positive association.
Do not identify direct proportion without the stronger defining condition.
Worked Identification 2: Anomaly vs Maximum
A data series rises smoothly to its highest point and then falls smoothly.
The highest point is a maximum. It is not automatically an anomaly because it fits the pattern.
Worked Identification 3: Independent vs Dependent Variable
Temperature is set to 20, 30, 40 and 50°C while reaction time is recorded.
- independent variable → temperature;
- dependent variable → reaction time.
The roles come from the method, not table position.
Worked Identification 4: Accuracy vs Precision
Five measurements lie very close together but far from the accepted value.
High precision, low accuracy.
The decisive feature is clustering versus closeness to the accepted value.
Worked Identification 5: Rotation vs Reflection
The shape turns 90° around a fixed point.
Rotation.
The fixed centre and turn rule separate it from reflection.
Worked Identification 6: Writer’s Topic vs Writer’s View
The passage concerns homework and argues that excessive repetitive homework reduces motivation.
- topic → homework;
- view → excessive repetitive homework can reduce motivation.
Read the target noun.
Worked Identification 7: Message vs Purpose
A poster says “Join the clean-up this Saturday—our river needs you.”
- message → the community should help clean the river;
- purpose → persuade or recruit residents to join the clean-up.
Content and intended effect are related but distinct.
Worked Identification 8: Cause vs Trigger
Economic hardship persists for years; a sudden arrest sparks demonstrations.
- long-term cause → economic hardship;
- immediate trigger → arrest.
Timing helps separate causal roles.
Worked Identification 9: Movement Along Demand vs Shift of Demand
Price changes while the demand curve itself remains fixed.
Movement along the demand curve.
A shift requires the curve itself to move.
Worked Identification 10: Revenue vs Profit
A case gives total sales income before costs.
Revenue.
Profit requires costs to be deducted.
Worked Identification 11: Stakeholder From Interest
The case says a group wants long-term contracts and prompt payment.
Suppliers.
The interest pattern distinguishes the stakeholder.
Worked Identification 12: Phishing vs Malware
An attacker sends a fake login page to steal credentials.
Phishing.
The defining mechanism is deceptive credential capture.
Worked Identification 13: Retrieval Failure vs Understanding Failure
A learner explains a concept accurately with prompts but cannot produce it unaided the next day.
Retrieval instability.
The evidence does not automatically show conceptual misunderstanding.
Worked Identification 14: Method-Selection Error
A learner recalls several formulas correctly but repeatedly chooses the wrong one for unfamiliar questions.
Question-recognition or method-selection weakness.
More formula memorisation alone may not repair the problem.
Worked Identification 15: Timing Error vs Knowledge Error
A learner scores accurately on attempted final questions but leaves three blank because time expires.
Timing/completion failure.
Blank answers are not automatically knowledge gaps.
Worked Identification 16: Checking Error
A correct answer is changed to an incorrect one without new evidence.
Unstable checking / unsupported answer change.
Worked Identification 17: Command-Word Error
The question asks “Describe,” but the student writes only reasons.
Command-word mismatch.
The content may be true and still answer the wrong job.
Worked Identification 18: Over-Answering Risk
The student correctly identifies “rotation,” then adds an incorrect definition of reflection.
The original identification was secure. The extra sentence creates avoidable ambiguity.
Worked Identification 19: Scope Error
The question asks for the tone of the final paragraph, but the student labels the tone of the whole article.
Wrong scope.
Correct object class, wrong evidence window.
Worked Identification 20: Negative-Polarity Error
The question asks for the factor that does not change the outcome. The student selects the strongest factor that does change it.
Polarity reversal.
The Identification Ladder
- Notice: detect relevant evidence.
- Classify: recognise the answer family.
- Discriminate: separate near-neighbours.
- Localise: respect scope and evidence window.
- Calibrate: avoid stronger labels than evidence supports.
- Select: commit to the best-supported answer.
- Inhibit: suppress unnecessary additions.
- Transfer: identify correctly in unfamiliar representations.
The Identification Strength Matrix
- Correct target + decisive feature: strong identification.
- Correct topic + wrong target: superficial recognition.
- Correct class + weak evidence: fragile guess.
- Right answer + incompatible extra answers: self-created ambiguity.
- Wrong answer + strong explanation: wrong classification.
- Correct answer + wrong scope: local/global mismatch.
- Correct category + wrong abstraction level: imprecise identification.
The Two-Candidate Rule
For difficult items, force yourself to name the nearest alternative.
Then ask what separates them.
This adds only seconds and prevents many familiar-label mistakes.
The One-Cue Rule
Before committing, produce one cue that supports the answer.
Rotation—fixed centre and turn.
Phishing—credentials requested through deception.
Supplier—concern is contracts and payment.
The cue is mostly for internal verification. The final answer can stay short.
The Counter-Cue Rule
If uncertain, look for one cue that would rule the answer out.
If the proposed direct-proportion graph misses the origin, reject the label. If the proposed dependent variable is deliberately set by the experimenter, reject the label.
Counter-cues reduce confirmation bias.
The Rapid Recognition Drill
Show 30 short examples from one category family and allow three to five seconds each.
label + one decisive feature.
Speed develops after accuracy.
The Mixed-Family Drill
Mix graph types, source tasks, experimental variables, language features and business categories.
This prevents the learner from predicting the answer family from the worksheet heading.
The Unfamiliar-Representation Drill
- rotate the diagram;
- change variable letters;
- rewrite the context;
- use a new graph scale;
- replace familiar vocabulary with equivalent wording;
- move the feature to a different position in the passage.
If performance collapses, the learner recognised the prototype, not the concept.
The Adversarial Example Drill
- a rising graph that is not proportional;
- a sentence containing “like” that is not functioning as a simile;
- a company with rising revenue but falling profit;
- a source that informs while primarily trying to persuade;
- a high point that is not an anomaly;
- a repeated measurement set that is precise but inaccurate.
Adversarial practice hardens decision boundaries.
The Error-Classification Drill
Give ten wrong exam responses and identify the failure type:
- knowledge;
- recognition;
- method selection;
- execution;
- notation;
- timing;
- checking;
- command-word control.
This teaches identification as diagnosis.
The Two-Minute Identify Clinic
- Circle Identify.
- Underline the target noun.
- Mark the evidence window.
- Name two plausible candidates.
- Find one decisive cue.
- Select one answer.
- Delete any unnecessary explanation.
Thirty Advanced Identify Prompts
- Identify whether a graph shows association or proportionality.
- Identify the anomaly without confusing it with the maximum.
- Identify the experimental variable role.
- Identify whether repeated measurements show precision or accuracy.
- Identify the transformation from invariant properties.
- Identify the first invalid algebraic step.
- Identify the theorem suggested by the geometry.
- Identify the writer’s view rather than topic.
- Identify message rather than purpose.
- Identify purpose rather than audience.
- Identify tone from mixed cues.
- Identify a language feature from structure rather than keyword.
- Identify a structural feature in a passage.
- Identify a long-term cause versus immediate trigger.
- Identify a consequence versus a cause.
- Identify a distribution pattern on an unfamiliar map.
- Identify a landform from structural evidence.
- Identify movement along a curve versus curve shift.
- Identify revenue versus profit versus cash flow.
- Identify a stakeholder from its interests.
- Identify a business objective from context.
- Identify a data type from stored information.
- Identify a security threat from attack mechanism.
- Identify a retrieval failure from study evidence.
- Identify a method-selection failure from exam work.
- Identify a timing failure from an incomplete script.
- Identify a checking failure from answer changes.
- Identify a command-word mismatch.
- Identify the strongest defensible conclusion from data.
- Identify when evidence is insufficient to distinguish two candidates.
The Seven-Day Identify Repair
- Day 1: target-object recognition.
- Day 2: near-neighbour discrimination.
- Day 3: graph, diagram and variable identification.
- Day 4: text, source and language-feature identification.
- Day 5: business/economics/computing classification.
- Day 6: error diagnosis and unfamiliar representations.
- Day 7: rapid mixed identification under fatigue.
The Twelve-Week Identify Arc
- Weeks 1–2: obvious positive examples and target nouns.
- Weeks 3–4: close alternatives and decision boundaries.
- Weeks 5–6: negative evidence and adversarial examples.
- Weeks 7–8: unfamiliar representations and cross-subject transfer.
- Weeks 9–10: mixed command words and scope/polarity control.
- Weeks 11–12: full-paper speed, final-quarter recognition and independent verification.
The Final-Week Identify Card
- what object?
- what evidence window?
- what candidate family?
- nearest alternative?
- decisive cue?
- polarity?
- select one;
- stop.
Why Identification Matters Beyond Examinations
Real expertise often begins with correct identification.
A doctor identifies the relevant sign before diagnosis. An engineer identifies the failure mode before repair. A programmer identifies the bug class before patching. A lawyer identifies the issue before arguing the rule. A manager identifies the bottleneck before allocating resources. A teacher identifies the first weak link before choosing an intervention.
Misidentification creates expensive downstream work.
If the wrong problem is named, the right solution may never be attempted.
Clara eventually stops treating Identify as a low-level command word.
She sees it as the first act of intelligent control:
recognise what is actually in front of you before deciding what to do with it.
The Canonical Boundary
This page owns Identify questions as an examination-performance structure: target recognition, candidate generation, near-neighbour discrimination, decisive-feature selection, local-evidence use, polarity/scope control and concise accepted naming or characterisation.
It does not replace State, Name, Define, Describe, Explain or Suggest. State expresses required information clearly and briefly. Name supplies an accepted term. Define gives precise meaning. Describe develops characteristics. Explain gives reasons/mechanisms. Suggest generates plausible possibilities. Identify asks: which thing is this?
Official Command-Word References
Command-word definitions vary by specification. Cambridge International currently defines Identify as name/select/recognise. AQA resources commonly define it as name or otherwise characterise. Learners should always use the glossary in the syllabus for the examination they are sitting. See the Cambridge International command-word guide and relevant AQA command-word guidance.
The Return Path
Find the target.
Find the evidence.
Generate the nearest candidates.
Find the decisive feature.
Select one.
Stop.
A strong Identify answer is not the first familiar label. It is the label that survives comparison with the nearest plausible alternative.
