Some exam questions are difficult because they ask for more than one intellectual job at once.
“Describe and explain.” “Compare and evaluate.” “State the trend and suggest a reason.” “Calculate the value and hence comment on the result.” “Identify the technique and explain its effect.”
The words can sit inside one sentence, but they do not form one task.
A student may answer the first command well and leave the second almost untouched. Another may blend both commands into one vague paragraph. A third may spend most of the time on the easier half because it feels more familiar.
This page owns one narrow examination-performance job: separating and completing multiple commands inside one question. How to Train Question Recognition for Exams owns broader task recognition. How to Write the Right Amount in Exams owns answer economy. This article asks: when one question contains several verbs, how does the learner keep every job visible?
The 60-Second Multiple-Command System
- Circle or identify every command.
- Translate each command into a job.
- Match evidence or working to each job.
- Notice mark allocation if given.
- Decide the order of response.
- Keep the jobs visibly separate in the answer.
- Check completion command by command.
- Stop only when every required task has been satisfied.
Aisha Describes but Does Not Explain
A Science question asks Aisha to describe the trend and explain why it occurs.
She writes three accurate sentences about the graph.
The values rise. The increase becomes slower. The curve begins to level off.
Her answer sounds substantial.
But she has not explained the mechanism.
The problem is not knowledge. It is task separation.
Her new margin note is simple:
DESCRIBE = what happened. EXPLAIN = why.
Two verbs. Two jobs.
Command Words Are Operations
Students sometimes treat command words as stylistic vocabulary.
They are better understood as operations.
- State: provide the required information directly.
- Describe: report characteristics, pattern or change.
- Explain: show the causal or logical relationship.
- Compare: make the relationship between two things explicit.
- Justify: support a claim or choice with reasons or evidence.
- Evaluate: weigh evidence or criteria and reach a judgement.
- Calculate: perform the required numerical method.
- Infer: derive a supported conclusion from evidence.
The exact meaning can vary by subject and syllabus, so official assessment language controls. But the learner should always ask: what output does this verb require from me?
One Sentence Can Contain Several Contracts
Consider:
Compare the two sources and explain which is more useful.
That contains at least two contracts:
- compare the sources;
- explain the judgement about usefulness.
A response that only says which source is more useful has skipped the comparison. A response that compares both but never reaches a judgement has skipped the evaluation of usefulness.
The question is complete only when both jobs are complete.
Separate Before You Integrate
Strong answers can eventually blend tasks smoothly.
But during training, separation comes first.
Ben can write:
- Task 1: identify the language technique;
- Task 2: explain its effect;
- Task 3: connect the effect to the writer’s purpose.
Once he reliably completes all three, the final answer can become natural prose rather than a mechanical checklist.
Mark Allocation Can Reveal Task Weight
If marks are shown, they may indicate that the question expects more than a token response to one command.
A six-mark “describe and explain” item probably deserves more work than a one-mark “state.” That does not create a universal sentence-per-mark rule, but it does help the learner budget depth.
The important question is: how much rewardable information belongs to each command?
The Hidden Second Command
Sometimes the second job is hidden in connective language.
- “and hence…”
- “and comment on…”
- “and suggest…”
- “giving reasons…”
- “support your answer with…”
- “with reference to…”
Students often notice the first verb and miss the phrase that modifies what counts as a complete answer.
Train the whole instruction, not just the first bold word.
Mathematics: Calculate and Interpret
Ryan calculates a probability correctly.
The question then asks him to comment on whether the model is reasonable.
His old answer ended with the number.
Now he treats the number as the input to a second job: interpretation.
Calculate → interpret.
The numerical method is not the whole question.
Mathematics: Show and Hence
A “show that” task may establish a result that a later “hence” instruction expects the learner to use.
Keep the established state visible. How to Handle Multi-Part Exam Questions owns the wider dependency structure. The multiple-command contribution is to notice when one instruction creates the state required by the next.
Science: Describe and Explain
Aisha uses a two-layer answer:
- Description layer: identify the observed pattern using the data or graph.
- Explanation layer: connect the pattern to the relevant scientific mechanism.
She checks that neither layer substitutes for the other.
Science: Predict and Explain
A prediction alone states what will happen.
The explanation must show why the changed condition should produce that outcome.
The strongest response may be compact:
prediction → mechanism → effect.
English: Identify and Explain Effect
Ben identifies a metaphor correctly.
That earns only the identification job if the question also asks for effect.
He then explains what the metaphor makes the reader imagine, feel or understand in this specific context.
Technique name and effect are separate information units.
English: Quote and Explain
A question may ask the learner to support an answer with textual evidence.
The quotation is not the explanation. The explanation is not a substitute for the quotation where evidence is explicitly required.
Ben’s sequence becomes:
claim → evidence → explanation.
Humanities: Compare and Evaluate
Clara first establishes the comparison.
Only then does she apply criteria and make the evaluative judgement.
If she jumps straight to “Source A is more useful,” she may never prove the comparison on which that judgement depends.
Humanities: Explain and Support
An explanation without supporting evidence can remain generic.
Evidence without explanation can remain disconnected.
Treat them as two linked commands even when the question phrases one as a condition rather than a separate verb.
Writing Prompts Can Hide Several Jobs
Mira reads: “Discuss whether schools should limit students’ use of artificial intelligence, considering both benefits and risks.”
The prompt contains several jobs:
- discuss;
- address benefits;
- address risks;
- reach a position on limits.
If she writes only about benefits, she has not merely produced an unbalanced essay. She has failed part of the task contract.
The Command Split
For training, rewrite complex questions into separate lines.
1. Describe the trend.
2. Explain the trend.
3. Support the explanation with data.
Then answer.
Over time, the learner should become able to perform the split mentally without rewriting the question.
Use a Task Ledger
For high-value questions, build a miniature task ledger in rough work:
- D = describe;
- E = explain;
- J = justify;
- 2 examples;
- final judgement.
Cross each item mentally as it is completed.
How to Use Rough Work in Exams owns the wider scratch-space system. The task ledger is one narrow use of that space.
Order the Commands Deliberately
Often the natural order is the order printed.
But some integrated answers work better when evidence appears before judgement, or when calculation appears before interpretation.
The important thing is that every command remains traceable in the final response.
Do Not Give Equal Space Automatically
Two commands do not always deserve equal word count.
“State and explain” may need one short state plus a longer causal explanation. “Calculate and comment” may require a substantial calculation and a concise interpretation.
Allocate according to the intellectual work and mark scheme, not grammatical symmetry.
The Missing-Command Check
Before leaving the question, reread only the command words and conditions.
- Did I describe?
- Did I explain?
- Did I compare both?
- Did I justify the choice?
- Did I use the required source or data?
- Did I reach the requested judgement?
This is faster than rereading every sentence and is targeted at task omission.
The Last-Verb Problem
Under time pressure, the last verb in a long question is especially vulnerable.
The learner begins answering after recognising the first task and never returns to the rest of the instruction.
Train the habit of finishing the reading before beginning the response.
Multiple Commands in Multi-Part Questions
A multi-part question can contain multiple commands inside each subpart.
For example, part (c)(ii) may ask the learner to “calculate and hence explain.”
Do not assume that reaching (c)(ii) means only one job remains. Separate locally inside each subpart.
Multiple Commands Under Time Pressure
When time is short, students often sacrifice the later command.
A better emergency strategy is to preserve minimum completion for every required job before polishing any one of them.
If a six-mark question requires comparison and evaluation, a concise valid comparison plus a concise judgement may protect more marks than a long comparison with no evaluation at all.
The Two-Command Drill
Give ten questions containing two commands.
Before answering, the learner must write the two jobs in plain language.
Example:
COMPARE = say how they are alike/different. JUSTIFY = give evidence for my choice.
Then answer within a time limit.
The Hidden-Condition Drill
Use questions where the second job appears in a phrase such as “using evidence from the graph” or “with reference to the source.”
The learner must identify both the visible verb and the completion condition.
The Command-Weight Drill
Show several multiple-command questions with different mark allocations.
Ask how the likely depth of each job changes with the marks and subject convention.
The aim is to prevent both under-answering and automatic equal splitting.
The Multiple-Command Audit
- Does the learner identify every command before writing?
- Can each command be translated into a concrete job?
- Are hidden conditions such as evidence or source use noticed?
- Is mark weight considered?
- Does each command receive the evidence or working it requires?
- Can tasks be integrated without one disappearing?
- Does the learner avoid equal word allocation by habit?
- Are later commands protected under time pressure?
- Can completion be checked command by command?
- Does the learner stop only when the full task contract is satisfied?
Red, Amber and Green Command Control
Red: the learner sees only the topic or first command; later jobs disappear; evidence and judgement requirements are missed.
Amber: commands are recognised but sometimes blended badly or allocated the wrong depth under pressure.
Green: every command becomes a visible job, each receives the right evidence or working, and the final answer remains integrated without losing any part of the task.
Aisha Reads the Verbs Again
Describe the trend.
Explain why.
Use the data.
Three jobs.
She writes one sentence for the pattern, two for the mechanism, and one short data reference that proves the pattern she described.
The answer is not longer because she wrote more.
It is more complete because she answered everything.
The Canonical Boundary
This page owns multiple commands inside a single examination question: command separation, task mapping, evidence assignment, answer integration and completion checking.
It does not replace broad question recognition, subject-specific command-word teaching or answer economy. Its narrow job is to stop one sentence with several verbs from becoming an answer to only the first verb.
The Return Path
Read every verb.
Turn every verb into a job. Give every job the evidence, working or judgement it needs. Then check the instruction again.
One question can contain several tasks. A complete answer knows the difference.
