How to study for source-based questions: identify what the question asks, read each source for its overall message, inspect provenance and purpose, select precise evidence, use contextual knowledge to interpret the source, compare sources directly, and evaluate what the source can and cannot support.
Source-based questions are not ordinary comprehension questions. The source is evidence to be handled, not merely text to be paraphrased. Students must decide what the source is saying, why it may have been produced, what its origin adds or limits, how it relates to other evidence, and how contextual knowledge changes the interpretation.
Cambridge examiner guidance makes this explicit: candidates are expected to handle and evaluate source material, use specific details from the sources, and ground source analysis in wider contextual knowledge. Strong preparation therefore needs repeated practice in evidence selection, provenance, comparison and judgement.
This guide treats source-based question preparation as a repeatable exam skill. The student must recognise the demand, extract or retrieve the relevant evidence, build the required reasoning, and then test the same skill on a changed question so the method becomes transferable rather than tied to one example.
This approach is especially useful when a student needs to:
- identify the overall message of a source;
- use source details as evidence rather than paraphrase;
- analyse provenance and purpose;
- compare two or more sources directly;
- use contextual knowledge appropriately;
- evaluate usefulness and reliability;
- answer source-based questions under time;
- avoid generic source-analysis templates.
Arrange a parent–student consultation with eduKate Singapore
Chat with eduKateSG on WhatsApp
Read the Question Before the Sources
The task determines what kind of source analysis is needed.
A usefulness question, comparison question and purpose question require different evidence. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Underline the command word.
- Identify named sources.
- Note mark value.
- Restate the required judgement.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
How useful is Source C? sends the student looking for both content value and limitations rather than only summary.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not analyse every possible feature of the source before knowing the task.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Identify the Overall Message
Read the source as a whole before extracting details.
Cambridge examiner guidance warns that isolated sentences can distort the source’s overall argument or point of view. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Summarise the main message in one sentence.
- Identify attitude or position.
- Separate main claim from supporting detail.
- Check whether tone changes.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A political cartoon is first summarised as criticism of a policy before individual symbols are analysed.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not build the answer from disconnected quotations.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Use Specific Source Details
Evidence should come from the source itself.
General comments about what a source says are weaker than precise references. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Quote briefly or paraphrase precisely.
- Select only relevant detail.
- Explain what the detail shows.
- Connect it to the question.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
Instead of saying the source is critical, the student identifies the specific phrase or visual element that shows criticism.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not paraphrase the whole source.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Read Provenance
Who produced the source, when, where and in what role can affect interpretation.
Provenance is useful only when connected to what the source says and the question being asked. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Identify author or creator.
- Identify date and context.
- Identify audience.
- Ask how these affect value or limitation.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A government speech may be useful for official policy and public messaging while less useful as neutral evidence of public reaction.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not write author/date/purpose mechanically without linking them to evaluation.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Analyse Purpose
Purpose goes beyond saying to inform or persuade.
The strongest answers explain what response the creator hoped to produce in a specific audience at a specific time. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Identify target audience.
- Identify desired reaction.
- Use content as evidence.
- Use context to explain why the purpose mattered.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A wartime poster is analysed for the action it sought from civilians and why that action was important in that context.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not use generic purpose labels without explanation.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Use Contextual Knowledge
Own knowledge helps interpret and evaluate the source.
Cambridge guidance expects source use to be informed by wider contextual knowledge. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Use knowledge directly relevant to the claim.
- Confirm or challenge source content.
- Explain why a statement mattered then.
- Avoid dumping unrelated facts.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
Context explains why a diplomatic statement would have been surprising or strategic at that moment.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not attach a memorised paragraph of background history to every source.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Compare Sources on the Same Issue
Direct comparison needs a shared point.
Two separate summaries are not a comparison. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Choose one issue.
- State agreement or disagreement.
- Use evidence from both.
- Explain the significance.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
Source A presents reform as necessary whereas Source B presents the same reform as dangerous, supported by a detail from each.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not discuss A completely and then B completely without linking them.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Compare Provenance When Relevant
Different creators may have different access, interests and purposes.
Provenance can explain both agreement and disagreement. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Compare roles.
- Compare timing.
- Compare audiences.
- Connect provenance to the issue.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
Two eyewitnesses may disagree because they observed different locations or had different institutional roles.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not assume one source is reliable simply because the author was present.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Evaluate Usefulness
Useful to whom and for what question?
A biased source can still be useful for studying attitudes, propaganda or official priorities. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- State what the source is useful for.
- Use content value.
- Use provenance value.
- State limitations relative to the enquiry.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A partisan newspaper may be limited for neutral factual balance but useful for studying partisan rhetoric.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not equate bias with uselessness.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Evaluate Reliability Carefully
Reliability is not a moral label.
Students should test claims against provenance, context and corroboration. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Identify claim.
- Consider motive and access.
- Cross-check with another source.
- Use contextual knowledge.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A retrospective memoir is checked against contemporary evidence before a judgement is made.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not call every official source reliable or every propaganda source unreliable.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Use Corroboration
Sources can support or challenge one another.
Agreement matters more when the sources are meaningfully independent. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Find the common claim.
- Check whether sources share origin.
- Use specific matching detail.
- Note meaningful contradiction.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
Two independent sources agree on the timing of an event, strengthening confidence in that element.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not treat copied or derivative sources as independent corroboration.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Handle Visual Sources
Cartoons, photographs and posters combine content with symbolism and audience.
Visual details must be interpreted in context. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Identify literal elements.
- Identify symbols or exaggeration.
- Infer message.
- Link purpose to context.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A cartoon’s oversized figure is treated as a clue to the artist’s criticism, not merely described visually.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not describe every object without explaining its meaning.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Handle Statistics and Maps as Sources
Numerical and spatial sources also have provenance and limitations.
Data can look objective while still depending on collection choices and categories. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Read units and scale.
- Check source origin.
- Identify what is missing.
- Use values selectively.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A migration table is useful for trend size but may not explain motives, so another source is needed.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not assume numbers are self-explanatory.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Build a Source-Based Paragraph
A paragraph should answer the question, use source evidence and explain the reasoning.
Structure keeps analysis from turning into paraphrase. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Judgement sentence.
- Source detail.
- Context or provenance.
- Explanation linked to the question.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
The paragraph states why Source C is useful, cites one specific detail and explains how the author’s role increases value for that enquiry.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not begin every paragraph with Source A says.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
Practise Full Sets Under Time
Source papers often contain several question types.
Students need to switch between purpose, comparison, usefulness and synthesis. The important test is whether the student can make the correct decision without a model answer supplying the structure. A useful study method therefore produces an answer, judgement, comparison or application that can be checked.
What to do
- Use real source sets.
- Allocate time by marks.
- Plan briefly.
- Review question-type errors afterwards.
Keep the task small enough that the reasoning is visible. One source comparison, one provenance judgement, one application example or one short paragraph often gives clearer diagnostic evidence than a long passive review session.
Worked example
A timed set reveals that the student spends too long on the first comparison and rushes the final synthesis question.
After the attempt, compare with a reliable source, mark scheme or teacher explanation only after committing. Correct the smallest missing feature, close the model and produce the reasoning again independently.
Common mistake
Do not practise only one source skill at a time near the exam.
If the same weakness returns, change the intervention. Use a contrasting example, revisit prior knowledge, change the representation, ask a precise question or practise with a new context. Repetition helps only when it targets the actual bottleneck.
How the Method Changes by Subject
English
For English source-like comprehension, use precise textual evidence and distinguish literal meaning from inference.
English questions often require precise use of evidence and scope. Practice should end in actual responses rather than recognition of model wording.
Mathematics
Source-based work is less common, but data, diagrams and written scenarios can still require evidence-based interpretation.
Mathematics application requires selecting a method from structure rather than chapter labels. Explain why the method fits before calculating.
Science
Experimental descriptions and datasets can be treated as evidence: inspect method, variables, results and limitations.
Science transfer requires using mechanisms, variables, diagrams and data in unfamiliar contexts.
Humanities
This is the main application: use content, provenance, contextual knowledge, corroboration and explicit judgement.
Humanities source work requires evidence, provenance, contextual knowledge and judgement without simply paraphrasing the source.
Three Student Patterns
Maren
Maren writes everything she knows about the historical topic and underuses the sources. She needs source-evidence-first paragraphs.
Iona
Iona paraphrases sources accurately but rarely evaluates provenance. She needs creator-purpose-audience practice.
Leonie
Leonie knows the source skills but spends too long annotating. She needs a fixed reading routine and mark-based time limits.
Different students can produce the same weak answer for different reasons. One may know too much and lose scope, another may recognise but not retrieve, and another may rush the analysis under time.
A Seven-Day Implementation Cycle
- Day 1: identify the question type and complete one diagnostic attempt.
- Day 2: repair the highest-value reasoning gap.
- Day 3: use a new example or source.
- Day 4: mix the skill with another question type.
- Day 5: complete a short timed set.
- Day 6: revisit only recurring errors.
- Day 7: retest after a gap with reduced support.
The cycle can be compressed when the assessment is near, but it should preserve independent attempt, feedback, correction, transfer and delayed confirmation.
Frequently Asked Questions
Should I quote sources?
Use brief precise quotations or accurate paraphrase; the important part is explaining what the detail proves.
What is provenance?
Information about a source’s origin, such as creator, date, context, role and audience.
Does bias make a source useless?
No. A biased source may be highly useful for studying attitudes, propaganda or official priorities.
Should I use outside knowledge?
Yes when the exam expects it, but keep it directly relevant to interpreting or evaluating the source.
How do I compare sources?
Compare both on the same issue and support the comparison with specific evidence from each.
How do I evaluate usefulness?
Ask useful for what enquiry, then weigh content, provenance and limitations.
What about cartoons?
Identify the overall message, symbols, exaggeration, audience and purpose in context.
How do I handle statistics?
Check units, source origin, categories and what the data cannot show.
How do I get faster?
Practise a fixed routine: question, message, evidence, provenance, context, judgement.
How do I know I am ready?
You can analyse unfamiliar sources, compare them directly and make evidence-based judgements under time.
One-Page Operating Manual
Question: Identify the required source skill.
Message: Read the source holistically.
Evidence: Use precise details.
Provenance: Analyse origin, audience and purpose.
Context: Apply relevant knowledge.
Compare: Use a shared issue.
Evaluate: Judge usefulness or reliability for the enquiry.
Time: Practise full source sets.
Why Evidence Must Do a Job
Evidence earns value when it supports a claim, comparison, evaluation or application. Listing a quotation, number or fact without explaining its relevance leaves the reasoning incomplete. Practise adding the sentence that tells the reader what the evidence proves or limits.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Use Contrast
Many difficult exam questions are discrimination tasks. Compare two sources, two concepts or two possible methods directly and identify the decisive difference. Contrast reduces the chance that a student knows both ideas separately but still confuses them under pressure.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Test Transfer
After correcting one answer, change the source, scenario, numbers, representation or wording. If the student can still recognise the underlying demand and construct a valid response, the skill is becoming transferable.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Review Under Time
Once accuracy is stable, use short timed sets. The student should practise deciding what matters quickly, producing enough evidence and moving on before one question absorbs a disproportionate amount of the paper.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Use Feedback Without Copying
Read the feedback after committing to an answer. Identify the missing function—evidence, provenance, mechanism, application, comparison or judgement—then hide the model and produce a new response. Copying changes the page without necessarily changing the process.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Close the Loop
Define what stable means. A skill is not stable merely because the original corrected answer can be reproduced. Require success after a delay and on at least one changed question, then move the skill to lighter maintenance.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
Why Evidence Must Do a Job
Evidence earns value when it supports a claim, comparison, evaluation or application. Listing a quotation, number or fact without explaining its relevance leaves the reasoning incomplete. Practise adding the sentence that tells the reader what the evidence proves or limits.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Use Contrast
Many difficult exam questions are discrimination tasks. Compare two sources, two concepts or two possible methods directly and identify the decisive difference. Contrast reduces the chance that a student knows both ideas separately but still confuses them under pressure.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Test Transfer
After correcting one answer, change the source, scenario, numbers, representation or wording. If the student can still recognise the underlying demand and construct a valid response, the skill is becoming transferable.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
How to Review Under Time
Once accuracy is stable, use short timed sets. The student should practise deciding what matters quickly, producing enough evidence and moving on before one question absorbs a disproportionate amount of the paper.
After practising one source set, use a new source type or creator so the student must apply the same evidence-handling process without relying on familiar content.
Helpful Reading
- Cambridge IGCSE History | Specimen Source-Based Paper
- Cambridge IGCSE History | Examiner Report on Source Use
- Cambridge Pre-U History | Source-Based Mark Scheme
Source-Based Questions Reward Evidence Handling, Not Source Paraphrase
The source is evidence to be interpreted and evaluated in relation to a question.
Read the message, use precise details, analyse provenance, bring in context and make the judgement explicit.
The strongest source answer shows what the evidence can support—and where its limits begin.
Arrange a Parent–Student Consultation
eduKateSG
8 Fourth Avenue
Singapore 268674
Near Sixth Avenue MRT
Premium 3-pax small-group tuition
By appointment
Properly taught kids shine a bright light into the future.