Why is "Naming the evidence type but not evaluating it" a common mistake in Evaluating Evidence?

Why it happens: Identifying the type feels like a complete answer. How to avoid it: Naming is analysis; the marks are in judgement. After naming the type, run R-R-S and say how strongly it supports the claim. Source: 9239 Component 1 Q2 — evaluate, don't just identify

Why is "Assuming statistics are strong and anecdotes weak by default" a common mistake in Evaluating Evidence?

Why it happens: Numbers look authoritative; stories look casual. How to avoid it: Judge each piece on R-R-S, not its format. A biased or tiny-sample statistic can be weak; a well-chosen anecdote can be apt for a modest claim. Source: 9239 Component 1 Q2 — evaluate evidence on its merits

Why is "Using 'biased' as a knock-out word" a common mistake in Evaluating Evidence?

Why it happens: Spotting a vested interest feels like the end of evaluation. How to avoid it: Explain the slant, why it arises, and how it affects the weight you give the evidence — then consider corroboration. A biased source can still hold checkable data. Source: 9239 Component 1 — bias as evaluation, not labelling

Why is "Ignoring relevance — accepting impressive-sounding but off-point evidence" a common mistake in Evaluating Evidence?

Why it happens: Strong-sounding evidence is taken at face value. How to avoid it: Always check the evidence actually bears on THIS claim (e.g. 'feeling safer' does not show 'crime fell'). Relevance is the most overlooked R-R-S test. Source: 9239 Component 1 Q2 — relevance of evidence

Why is "Not matching evidence weight to the size of the claim" a common mistake in Evaluating Evidence?

Why it happens: Students rate evidence in the abstract rather than against the claim. How to avoid it: Ask: how big is the claim, and can this evidence carry that weight? A single case can't support a universal or superlative claim. Source: 9239 Component 1 Q2 — sufficiency relative to the claim

Why is "Saying evidence is weak without saying what would be stronger" a common mistake in Evaluating Evidence?

Why it happens: Identifying a weakness feels like the finish line. How to avoid it: Name the further evidence needed to substantiate or refute the claim (larger/representative sample, independent source, current data, a control). This is the top-band move. Source: 9239 Component 1 Q2 — discriminating evaluation

Why is "Arguing the topic yourself instead of evaluating the source's evidence" a common mistake in Evaluating Evidence?

Why it happens: The issues are engaging and students share their own views. How to avoid it: Keep asking: does THIS evidence support THIS source's claim? Evaluate the source's evidence; do not import outside facts or your own opinion on the topic. Source: 9239 Component 1 — engage with the source's evidence

Cambridge International A Levels Global Perspectives & Research (9239)

Evaluating Evidence

0% Complete

Detailed Study Notes

Detailed notes on Component 1: Written Exam for Cambridge International A Levels Global Perspectives & Research, covering key concepts, explanations, examples, and exam-focused revision points.

Evaluating Evidence — Cambridge AS & A Level Global Perspectives & Research 9239 (Component 1)

Once you can take an argument apart, you must judge how good its evidence really is — this is where Component 1's higher marks live. Evaluating evidence means weighing each piece on relevance, reliability and sufficiency, recognising common weaknesses (small samples, unrepresentative data, outdated sources, vested interests), and saying what further evidence would be needed to substantiate or refute a claim. These notes give you the R-R-S framework, a checklist of the weaknesses examiners reward you for spotting, and the A* finishing move — naming the better evidence the argument lacks.

At a glance

Evaluating evidence = judging how strong each piece is, not just naming it. This is where Q2's higher marks live.
Name the TYPE first (statistical, anecdotal, expert, case study, historical; primary/secondary), THEN judge it.
Use R-R-S: Relevance (does it support THIS claim?), Reliability (credible source?), Sufficiency (is there ENOUGH?).
Common weaknesses to spot: small samples, unrepresentative data, outdated sources, vested interests, no method given.
Match evidence WEIGHT to claim SIZE — a sweeping claim needs broad, representative evidence.
Strong vs weak is RELATIVE to the claim: the same statistic can be strong or weak depending on what it must prove.
The A* finish: say what FURTHER evidence would be needed to substantiate or refute the claim.
Evaluate the evidence in the SOURCE — don't argue the topic yourself or import outside facts.

What you’ll learn

Mapped to the Cambridge International A Level 9239 syllabus (2026-2028).

Distinguish strong evidence from weak evidence in source documents.
Assess the reliability and credibility of evidence, including vested interest and currency.
Identify common weaknesses: small samples, unrepresentative data, outdated sources, vested interests.
Identify what further evidence would be needed to substantiate or refute a claim.

What evaluating evidence means — and why it carries the marks

Weigh each piece of evidence for how well it supports the claim — the step up from identifying to judging.

Analysing evidence tells you what type it is. Evaluating evidence tells you how good it is — whether it genuinely supports the claim it is attached to. This step up from identifying to judging is exactly where Component 1's higher marks are won.

Always do two things, in order:

Name the type precisely — statistical, anecdotal, expert opinion, case study, historical (and primary or secondary).
Judge its strength — using the R-R-S framework below.

Why it matters across the paper:

Q1 (5 marks) may ask you to identify the type of evidence.
Q2 (15 marks) asks you to evaluate the evidence in one source — the core of the question.
Q3 (25 marks) asks you to compare the evidence in two sources to judge which is better supported.

The mindset: evidence is rarely simply 'good' or 'bad'. Its strength is relative to the claim it is asked to support. A single anecdote is weak support for "migrants take jobs" (a sweeping claim) but perfectly adequate to show that "an individual had this experience". So you are never just rating the evidence — you are asking can this evidence carry the weight of this claim?

Evaluating = judging strength, not just naming the type.
Always name the type (+ primary/secondary) before judging it.
Evidence strength is relative to the size of the claim it must support.
Q2 is built on this skill; Q3 compares evidence across two sources.

The R-R-S framework: Relevance, Reliability, Sufficiency

Three questions that systematically test any piece of evidence's strength.

R-R-S is your reliable framework for evaluating any piece of evidence.

Relevance — does it support this reason and conclusion?

A statistic about one city does not, by itself, support a claim about every country.
Watch for evidence that sounds impressive but doesn't actually bear on the claim (e.g. "people feel safer" used to support "crime fell").

Reliability — can you trust it?

Source/authority: a recognised, independent body, or an interested party? Vested interest weakens reliability.
Method: do we know how it was gathered? An unstated method or tiny sample is a red flag.
Currency: is it up to date? Old data may not describe the present.

Sufficiency — is there enough?

One company, one village, three farmers — too few to support a broad claim.
Sweeping words ("every", "most effective", "the future") demand broad, representative evidence.

A strong Q2 answer runs each key piece of evidence through R-R-S — and a strong Q3 answer uses R-R-S to compare the two sources' evidence directly.

Relevance: does the evidence actually bear on this claim?
Reliability: authority/vested interest, method, currency.
Sufficiency: is there enough, given the size of the claim?
Apply R-R-S to each key piece — and to compare sources in Q3.

Common weaknesses examiners reward you for spotting

Small samples, unrepresentative data, outdated sources and vested interests — name them precisely.

Cambridge mark schemes and examiner reports repeatedly credit students for naming specific weaknesses. Learn this checklist.

Weakness	What it looks like	Why it weakens the evidence
Small sample	"I asked five people…"	too few cases to generalise
Unrepresentative sample	only one group/region/age surveyed	not typical of the whole population
Outdated	a 2008 figure used today	conditions may have changed
Vested interest	industry funding research on its own product	motive to reach a particular result
No method given	"a study found…" with no detail	can't check how it was gathered
Selective/cherry-picked	only supporting data shown	ignores contrary evidence
Anecdote as proof	one story used for a general claim	a single case proves little
Vague quantifier	"many", "most", "huge"	not measured; sounds bigger than it is

Two important nuances (these earn marks):

A biased source is not automatically worthless. Note the vested interest, treat its conclusions cautiously, but recognise it may still show what a perspective argues — and its data could be checked against independent sources.
Bias ≠ reliability. An honest source can still be unreliable (small, outdated sample); an interested party can still present accurate data. Keep the two separate.

Naming the specific weakness ("a sample of five is too small to represent all students") scores; a vague "the evidence isn't very good" does not.

Spot specific weaknesses: small/unrepresentative sample, outdated, vested interest, no method, cherry-picking.
Name the weakness precisely — vague 'it's weak' does not score.
A biased source can still hold usable, checkable evidence.
Keep bias (a slant) separate from reliability (quality of the evidence).

The A* finish: what further evidence is needed

Don't just say evidence is weak — say what would substantiate or refute the claim.

The move that lifts evaluation into the top band is naming the better evidence the argument lacks — what would be needed to substantiate (support) or refute (disprove) the claim.

Don't stop at "weak" — say what would convince:

"The 12% productivity rise comes from one company during a pandemic. To support a claim about all firms in normal times, we would need multi-company data gathered over several ordinary years."

"A single anecdote about one thriving business cannot show microloans are 'the most effective' tool worldwide; that would require large-scale comparative studies across many countries and against other anti-poverty methods."

Why this scores: it proves you understand why the evidence falls short and what good evidence for this claim would look like — the precise discrimination the higher levels reward. It also naturally leads into a reasoned judgement (Q2 and Q3 both require one).

A simple template:

State the weakness (which R-R-S test fails, named specifically).
Say what better evidence would look like (larger/representative sample, independent source, current data, comparison group).
Conclude how this affects how far the claim is supported.

Get into the habit of finishing every evidence evaluation with "to be convincing, this would need…". It turns a criticism into a judgement.

Name the further evidence needed to substantiate or refute the claim.
Specify what 'better' looks like: larger/representative sample, independent source, current data, a comparison.
This shows WHY the evidence falls short — the top-band discrimination.
Finish every evaluation with 'to be convincing, this would need…'.

Evaluating evidence in the exam

Apply name-type → R-R-S → weaknesses → further-evidence to each key piece, and compare across sources in Q3.

Put it together with a repeatable routine you can run on any piece of evidence under time.

The evaluation routine (per key piece of evidence):

Name the type (statistical / anecdotal / expert / case study / historical; primary or secondary).
Run R-R-S — Relevance, Reliability, Sufficiency.
Name the specific weakness (small sample, vested interest, outdated, etc.).
State the further evidence needed to substantiate or refute the claim.
Conclude how far the evidence supports the claim.

In Q2 (15 marks): apply this to the key evidence in one source, then reach a judgement on how convincing the source is overall.

In Q3 (25 marks): use it comparatively — "Source A's evidence (national survey, 10,000 people) is far more representative than Source B's (two neighbours), so A's claim about 'the public' is much better supported." Comparing evidence quality directly is central to the Q3 judgement.

Time discipline: you cannot evaluate every sentence. Pick the key pieces of evidence — the ones the conclusion most depends on — and evaluate those well. Depth on the load-bearing evidence beats a shallow sweep of everything.

Routine: name type → R-R-S → specific weakness → further evidence → conclude.
Q2: evaluate the key evidence in one source, then judge it overall.
Q3: compare the two sources' evidence directly to judge which is better supported.
Evaluate the load-bearing evidence in depth, not every sentence shallowly.

Quick recap

Evaluating evidence means judging strength, not just naming the type — this is where Component 1's higher marks live.
Name the type (+ primary/secondary), then apply R-R-S: Relevance, Reliability, Sufficiency.
Spot specific weaknesses: small/unrepresentative samples, outdated data, vested interests, no method, cherry-picking.
Strength is relative to the claim — match evidence weight to claim size; don't judge by format.
Finish by naming the further evidence needed to substantiate or refute the claim (the A* move).
Evaluate the load-bearing evidence in the source; in Q3 compare the two sources' evidence directly — never import outside facts.

Memorise this

Verbatim phrases and definitions Cambridge mark schemes credit.

Evaluate evidence — judge strength (not just name the type); where Q2's higher marks live.
R-R-S — Relevance, Reliability, Sufficiency.
Reliability checks — authority/vested interest, method, currency.
Weaknesses — small sample, unrepresentative, outdated, vested interest, no method, cherry-picked, vague quantifier.
Match weight to claim size — sweeping claims need broad, representative evidence.
Bias ≠ reliability — an honest source can be unreliable; an interested source can be accurate.
A* finish — state the further evidence needed to substantiate or refute the claim.
Golden rule — evaluate the source's evidence; never argue the topic or import outside facts.

How it’s examined

Evaluating evidence is the heart of Component 1 Question 2 (15 marks), where you must analyse AND evaluate the argument and evidence in one source and reach a supported judgement, and it is central to Question 3 (25 marks), where you compare the evidence in two sources to decide which makes the stronger case. Question 1 (5 marks) tests the prior step — identifying the type of evidence. All of this is AO1 within the 1h 30m, 45-mark paper. The mark schemes reward precise evaluation: naming the evidence type, applying relevance, reliability and sufficiency, identifying specific weaknesses (small or unrepresentative samples, outdated data, vested interests, missing method), and saying what further evidence would substantiate or refute the claim. Examiner reports repeatedly flag three failings: stopping at naming the evidence without judging it, dismissing sources with the label 'biased' instead of evaluating them, and drifting into the candidate's own opinion on the topic. The strongest answers judge each load-bearing piece of evidence against the size of the claim it must carry, and finish by stating the better evidence the argument lacks.

Sources: Cambridge International AS & A Level Global Perspectives & Research 9239 syllabus (for examination in 2026, 2027 and 2028); 9239 Component 1 specimen paper and mark scheme (Cambridge International); 9239 Examiner Reports — Component 1 (evaluating evidence). Last reviewed 2026-06-30.

Take this whole topic with you

Step-by-step worked examples — Evaluating Evidence

Step-by-step solutions to past-paper-style questions on evaluating evidence, written exactly the way a tutor would explain them at the board.

Question type:

1Judging one piece of evidence with R-R-S
Getting startedIdentify & classify• AO1, evidence, R-R-S
Question
Source: "Our diet plan works. In a one-week trial, six of my friends said they felt slimmer." Evaluate the strength of this evidence.
Step-by-step solution
1. Step 1
  Name the type. Six friends' self-reports = anecdotal, primary evidence.
2. Step 2
  Relevance. 'Felt slimmer' is subjective and only loosely linked to whether the plan actually 'works'.
3. Step 3
  Reliability. Friends may want to please the author; one week is short; nothing was objectively measured.
4. Step 4
  Sufficiency. Six people is far too few to show the plan works in general.
Answer
Weak evidence: anecdotal self-reports from six friends over one week. Doubtful relevance ('feeling slimmer' ≠ measured results), low reliability (friendship bias, no objective measure, very short trial) and clearly insufficient (tiny, unrepresentative sample). Objective measurements from a large, varied group over a longer period would be far stronger.
Examiner tip
A clean R-R-S judgement that names the type and ends with the better evidence needed — the finishing move that lifts evaluation marks.
2Naming a specific weakness
Getting startedIdentify & classify• AO1, evidence, weakness
Question
Source: "A 2007 report concluded that the city's air quality was excellent, so there is no need for new pollution controls." Identify the main weakness of using this evidence today.
Step-by-step solution
1. Step 1
  Name the type. A research report (secondary).
2. Step 2
  Spot the key flaw. It is from 2007 — likely outdated for a present-day claim about whether controls are needed.
3. Step 3
  Explain the effect. Air quality and traffic may have changed greatly since; old data may no longer describe the present.
Answer
The main weakness is that the evidence is outdated (2007). Air quality, traffic and population may have changed substantially since, so a 17-year-old report cannot reliably show that no new controls are needed today. Current data would be required.
Examiner tip
Naming the specific weakness (currency/outdated) and explaining its effect scores; a vague 'the evidence isn't good' does not.
3Evaluating reliability: vested interest and method
Building confidenceIdentify & classify• AO1, reliability, vested interest
Question
Source: "A study funded by a fast-food chain found that its meals are perfectly healthy. There is therefore no need for warning labels." Evaluate the evidence.
Step-by-step solution
1. Step 1
  Type. A research study (secondary).
2. Step 2
  Vested interest. Funded by the fast-food chain, which profits from the conclusion — a clear conflict of interest.
3. Step 3
  Method. No sample size, method or independent replication is given.
4. Step 4
  Weight + further evidence. Treat cautiously; independent, peer-reviewed studies would be needed before accepting 'no warning labels'.
Answer
Unreliable: the study is funded by the company that profits if its meals are seen as healthy (vested interest), and no method, sample or independent replication is given. One such study is insufficient to justify the strong policy claim. Independent, peer-reviewed, replicated research would be needed — though the chain's data could still be checked against neutral sources rather than dismissed outright.
Examiner tip
Strong reliability evaluation: vested interest + missing method + insufficiency, ending with the better evidence required — and the mature note that a biased source's data can still be checked.
4Spotting a relevance gap
Building confidenceMulti-step problem• AO1, relevance
Question
Source: "Our new policing strategy is working — a survey shows 70% of residents now feel safer in the town centre." Evaluate how well the evidence supports the claim that the strategy reduced crime.
Step-by-step solution
1. Step 1
  Type. Survey/statistical evidence (feelings of safety).
2. Step 2
  Identify the gap. The claim is that crime fell; the evidence measures how safe people feel — these are different things.
3. Step 3
  Explain. People can feel safer while crime is unchanged (e.g. due to publicity), so the evidence is not directly relevant to a fall in crime.
4. Step 4
  Further evidence. Actual crime statistics before and after the strategy would be needed.
Answer
The evidence has a relevance gap: it measures how safe residents feel (70% feel safer), but the claim is that the strategy reduced crime. Feelings of safety can change without crime changing, so this survey does not directly support the claim. Recorded crime statistics before and after the strategy would be the relevant evidence.
Examiner tip
Relevance is the most overlooked R-R-S test. Spotting that 'feeling safer' ≠ 'crime fell', and naming the relevant evidence needed, is high-quality evaluation.
5Matching evidence weight to claim size
StretchMulti-step problem• AO1, sufficiency
Question
Source: "A village in Kenya cut child illness after installing one clean-water well. Clearly, well-building is the most effective way to improve health across the developing world." Evaluate the evidence against the claim.
Step-by-step solution
1. Step 1
  Type. A single case study (one village), primary.
2. Step 2
  Size the claim. 'The most effective way… across the developing world' is a sweeping, superlative, global claim.
3. Step 3
  Weigh. One village cannot show what is 'most effective' across many countries — a hasty generalisation; no comparison with other health measures is offered.
4. Step 4
  Further evidence. Large-scale comparative studies across many regions, comparing wells with other interventions, would be needed.
Answer
The evidence (one village case study) is far too narrow for the claim it must carry ('most effective… across the developing world'). It is insufficient and unrepresentative — a hasty generalisation — and offers no comparison with alternative health measures. To support such a superlative global claim, large-scale comparative studies across many regions, weighing wells against other interventions, would be required.
Examiner tip
The A* discrimination: judging evidence against the size of the claim, naming the hasty generalisation, and stating the broad comparative evidence a credible version would need.
6Naming evidence that would substantiate or refute
StretchMulti-step problem• AO1, further evidence
Question
Source: "Productivity at our firm rose 12% after we went fully remote, so remote work boosts productivity everywhere." State what evidence would substantiate the claim and what would refute it.
Step-by-step solution
1. Step 1
  Identify the claim type. A causal, generalised claim (remote work → higher productivity, everywhere).
2. Step 2
  Note the weaknesses. One firm; possible other causes; the period may be unusual.
3. Step 3
  Substantiating evidence. Multi-firm data across many sectors over several normal years, ideally comparing remote vs non-remote firms.
4. Step 4
  Refuting evidence. Firms that went remote and saw productivity fall, or evidence the rise was caused by something else.
Answer
To substantiate: broad, multi-firm data across sectors over several normal years, comparing firms that went remote with those that did not (a control), to establish a genuine causal effect. To refute: evidence of firms whose productivity fell after going remote, or evidence that the 12% rise was caused by another factor. Naming both shows the single-firm figure is currently insufficient to support a universal causal claim.
Examiner tip
Top-band thinking: distinguishing evidence that would substantiate from evidence that would refute, and recognising a control/comparison is needed for a causal claim. This is exactly the discrimination the higher levels reward.

Model Answers — Evaluating Evidence

High-scoring sample answers for evaluating evidence on the Cambridge International AS & A Level 9239 paper, with examiner-style notes mapping each response to the mark scheme and assessment objectives.

Question 1
9239 Component 1 Q1 style (5 marks)5 marks
[Q1 — 5 marks] Source: "Air pollution is a global health crisis. The World Health Organization estimates it causes seven million premature deaths a year, and in my own city I have watched the winter smog thicken with my own eyes." (a) Identify the type of evidence in "the WHO estimates it causes seven million premature deaths a year" and state whether it is primary or secondary. [2] (b) Identify the type of evidence in "in my own city I have watched the winter smog thicken" and state whether it is primary or secondary. [2] (c) Identify the conclusion the evidence is used to support. [1]
Model answer
(a) Statistical data (a quantified estimate); secondary — the figure is produced by the WHO, not gathered by the author.

(b) Anecdotal / observational evidence (a personal account); primary — the author reports it first-hand ("in my own city… with my own eyes").

(c) Conclusion: "Air pollution is a global health crisis."
Why this scores
Point-marked: in (a) and (b), one mark for the precise type and one for primary/secondary; one mark for the conclusion in (c). 'Data' alone is too vague for the type. Work only from the source.
Question 2
9239 Component 1 Q1 style (5 marks)5 marks
[Q1 — 5 marks] Source: "Our new fitness app works. In a survey we ran with 5,000 users, 82% reported exercising more, and three professional athletes have endorsed it. It is the best fitness app on the market." (a) Identify the type of evidence in "a survey we ran with 5,000 users, 82% reported exercising more" and state whether it is primary or secondary. [2] (b) Identify the type of evidence in "three professional athletes have endorsed it". [1] (c) Identify the conclusion. [1] (d) Identify one reason given to support the conclusion. [1]
Model answer
(a) Statistical / survey data; primary — the company ran the survey itself ("a survey we ran").

(b) Expert / celebrity endorsement (a form of expert opinion).

(c) Conclusion: "It is the best fitness app on the market."

(d) One reason: 82% of surveyed users reported exercising more (also acceptable: it has professional-athlete endorsements).
Why this scores
One mark per point, two for (a) (type + primary/secondary). Identifying types precisely — and noting the survey is primary because the company ran it — is the deconstruction skill; evaluation of these types comes in Q2.
Question 3
9239 Component 1 Q2 style (15 marks, single source)15 marks
[Q2 — 15 marks] Analyse and evaluate the argument and evidence in the source below. (Do not refer to any other source.) Source: "A 2012 report funded by an airline found that aviation's environmental impact is minimal. We should therefore allow unlimited airport expansion."
Model answer
Deconstruction. The conclusion is a policy claim — "we should allow unlimited airport expansion" — supported by one piece of evidence: a 2012 airline-funded report finding aviation's impact "minimal" (research/expert evidence, secondary).

Evaluating the evidence (R-R-S).

Reliability: the most serious flaw is a clear vested interest — an airline profits from expansion, so a report it funds has a motive to find "minimal" impact; it is not independent, and no method is given.

Currency: the report is from 2012; on a fast-moving environmental question, more recent (and possibly contrary) evidence should be preferred.

Relevance and sufficiency: even if relevant, a single industry-funded study cannot outweigh the wider body of independent climate research — it is insufficient for so strong a claim.

Evaluating the reasoning. The argument assumes the funded report is objective and that one favourable study settles the matter. The conclusion is also extreme ("unlimited"): even reliable evidence of low per-flight impact would not justify unlimited expansion, because impacts are cumulative.

Judgement and A* finish. The evidence is weak and unsuitable for the conclusion — compromised by vested interest, dated, and insufficient. To justify expansion, the argument would need recent, independent, peer-reviewed evidence on aviation's true (and cumulative) environmental impact; evidence that emissions rise substantially with expansion would refute it.
Why this scores
Single-source Q2: deconstruct, then evaluate the evidence on reliability (vested interest, method), currency and sufficiency, expose the 'unlimited' over-reach and the cumulative-impact point, and finish with the better/refuting evidence needed. Stay on this source only.
Question 4
9239 Component 1 Q2 style (15 marks, single source)15 marks
[Q2 — 15 marks] Analyse and evaluate the argument and evidence in the source below. (Do not refer to any other source.) Source: "Crime has fallen since we installed CCTV. In the first month after installation, reported thefts in the town centre dropped by 30%. CCTV obviously works, and cameras should be installed in every public space in the country."
Model answer
Deconstruction. The conclusion is a broad policy claim: "cameras should be installed in every public space in the country." Its support is one statistic — "reported thefts in the town centre dropped by 30% in the first month after installation" (statistical evidence).

Evaluating the evidence (R-R-S).

Relevance: the figure covers reported thefts, in one town centre, over one month — relevant to a small, local, short-term effect, but the conclusion is national: a serious scope mismatch.

Reliability: one month is far too short and may reflect normal or seasonal fluctuation; "reported" thefts may differ from actual crime.

Sufficiency: a single town's one-month figure is wholly insufficient for a nationwide policy, and "obviously works" overstates it.

Evaluating the reasoning. The argument assumes the CCTV caused the drop, ignoring other factors (more police, publicity, or crime simply displaced elsewhere) — a correlation-not-causation flaw. It also leaps from a local result to a universal recommendation (a hasty generalisation).

Judgement and A* finish. The evidence supports, at most, a modest, local, short-term association — nowhere near enough for cameras "in every public space." To substantiate the claim it would need data from many areas over a sustained period, ideally comparing areas with and without CCTV (a control) and checking whether crime merely moved elsewhere; evidence that crime did not fall (or rose nearby) in comparable areas would refute it.
Why this scores
Single-source Q2: R-R-S exposes the scope mismatch, short window and insufficiency; the reasoning section names the correlation-causation flaw and the generalisation; the A* finish specifies the controlled, multi-area evidence needed (and what would refute it). Committed judgement; no comparison.
Question 5
9239 Component 1 Q3 style (25 marks, comparison + judgement)25 marks
[Q3 — 25 marks] Compare and evaluate the arguments, evidence and reasoning in Source A and Source B, and reach a reasoned judgement on which presents the more convincing case. Source A: "A sugar tax works. After our country introduced one, national sales data showed sugary-drink sales fell 21% in two years, and dentists report fewer child cavities." Source B: "Sugar taxes are unfair and ineffective. They hit the poorest families hardest, and a study funded by the beverage industry found no measurable fall in obesity. People simply switch to cheaper sugary products."
Model answer
Introduction and thesis. Both sources judge sugar taxes but reach opposite conclusions — A that they "work", B that they are "unfair and ineffective". This answer argues that Source A's evidence is the stronger, though B raises a genuine fairness concern A ignores.

Comparing the evidence. A relies on national sales data (a 21% fall) plus professional reports from dentists; B relies on an industry-funded study plus assertions about regressiveness and substitution.

Reliability: A's national sales data is broad and comparatively independent, whereas B's central study is funded by the beverage industry — a clear vested interest that undermines its "no measurable fall" finding.

Relevance: A's sales fall is directly relevant to whether the tax changes behaviour, though "sales fell" is not the same as "health improved"; the dentist reports add a health signal but are anecdotal/professional. B's regressiveness claim is relevant to fairness, not effectiveness.

Sufficiency: A offers two converging strands; B's effectiveness case rests on one compromised study, and its "people switch to cheaper sugary products" is a plausible mechanism but unevidenced.

Comparing the reasoning and assumptions. A assumes that falling sales equate to a health benefit — a gap, since it does not directly measure obesity. B assumes its industry study is reliable and that substitution fully cancels the effect, neither of which it supports. So both have gaps, but A's gap (sales vs health) is partly bridged by the dentist evidence, whereas B's core evidence is fatally compromised by vested interest.

Comparing criteria and perspectives. The sources partly use different criteria: A judges by public-health effectiveness, B by effectiveness and fairness/equity. B's equity point — that flat taxes hit the poorest hardest — is a legitimate consideration that A simply does not address.

Reasoned judgement. On the evidence, Source A is clearly more convincing: independent national data plus a health signal, whereas B's case rests on an industry-funded study and unevidenced substitution. However, B contributes a valid criterion A ignores — fairness. The most defensible verdict is that A better shows sugar taxes change behaviour and plausibly improve health, while B's real strength is not its evidence but its reminder that effectiveness must be weighed against equity. Had A provided direct health-outcome data (e.g. obesity or cavity rates over time) and addressed distributional effects, its case would be close to conclusive.
Why this scores
Top-band Q3: integrated comparison of evidence (independent national data vs industry-funded study — vested interest), reasoning/assumptions (sales-vs-health gap; substitution unevidenced), and criteria (effectiveness vs equity), ending in a committed verdict that concedes B's genuine fairness point. Comparative connectives throughout; sequential summaries cap the mark.
Question 6
9239 Component 1 Q3 style (25 marks, comparison + judgement)25 marks
[Q3 — 25 marks] Compare and evaluate the evidence, reasoning and perspectives in Source A and Source B, and reach a reasoned judgement on which presents the more convincing case. Source A: "The new treatment should be adopted nationwide. A peer-reviewed study of 200,000 patients across 30 hospitals found it cut average recovery time by 25%." Source B: "The new treatment is a miracle cure that should replace all others. My cousin tried it and recovered in days, and a wellness blogger with millions of followers swears by it."
Model answer
Introduction and thesis. Both sources praise the new treatment, but they differ enormously in the quality of evidence and the size of the claim they make. This answer argues that Source A is overwhelmingly the more convincing, and that B, despite reaching a bolder conclusion, is the weaker case.

Comparing the evidence. A cites a peer-reviewed study of 200,000 patients across 30 hospitals — large-scale, independently reviewed statistical evidence. B cites a single anecdote (a cousin) and a social-media endorsement (a blogger).

Reliability: A's evidence is independently peer-reviewed with a method implied by scale; B's anecdote is uncontrolled (recovery could be coincidence or natural healing) and the blogger has no medical authority and possible vested interest (followers/sponsorship).

Representativeness and sufficiency: A's huge multi-hospital sample is representative and sufficient, whereas B's single case and one influencer are wholly insufficient to support any general claim.

Comparing the reasoning and claim size. This is decisive: A makes a bounded, proportionate claim (a 25% average reduction → adopt it) matched to broad evidence; B makes a sweeping, superlative claim ("miracle cure… replace all others") on the back of one anecdote — a textbook hasty generalisation where the claim vastly exceeds the evidence. B also relies on an appeal to popularity (the blogger's "millions of followers").

Comparing perspectives and criteria. A reasons from an evidence-based, scientific perspective (what controlled data shows); B from an anecdotal/testimonial perspective (personal experience and popularity). The criterion that should govern a medical decision — reliable, representative evidence — clearly favours A.

Reasoned judgement. Source A is far more convincing. Its evidence is independent, large-scale and representative, and its claim is proportionate to it; whereas B rests on an unrepresentative anecdote and an appeal to popularity stretched into an extreme claim. The irony is that B's conclusion is the bolder, yet its evidence is the weakest — exactly the mismatch that should make a careful reader trust it least. For B's claim to carry any weight it would need precisely what A already has: large, controlled, independent data. On any reasonable criterion for a medical decision, A prevails.
Why this scores
Top band: the integrated comparison turns on evidence quality (peer-reviewed scale vs anecdote + appeal to popularity) AND the mismatch between claim size and evidence, plus the criteria that should govern the decision. Committed verdict with comparative language throughout. The point that B's bolder claim has the weaker evidence is exactly the discrimination examiners reward.

Key Definitions and Keywords — Evaluating Evidence

Definitions to memorise and the exact keywords mark schemes credit for evaluating evidence answers — sharpened from recent examiner reports for the 2026 Cambridge International AS & A Level 9239 sitting.

Evaluating evidence
Examiner keyword
Judging how strongly a piece of evidence supports the claim it is attached to — its relevance, reliability and sufficiency — rather than merely naming its type.
Relevance, Reliability, Sufficiency (R-R-S)
Examiner keyword
The three tests of evidence strength: is it relevant to this claim, is the source reliable/credible, and is there enough of it to support the size of the claim?
Relevance
Examiner keyword
Whether the evidence actually bears on the claim made. Watch for evidence that sounds impressive but addresses something slightly different (e.g. 'feeling safer' vs 'crime fell').
Reliability
Examiner keyword
How far evidence can be trusted, judged by authority/independence (watch for vested interest), the method used, and currency (how up to date it is).
Related: def ee vested interest
Sufficiency
Examiner keyword
Whether there is enough evidence for the size of the claim. Sweeping or universal claims need broad, representative evidence; a single case is rarely enough.
Vested interest
Examiner keyword
A personal or financial stake that may bias a source (e.g. an industry funding research into its own product). A key reliability weakness — but a reason to calibrate, not automatically dismiss.
Representativeness
Examiner keyword
Whether a sample or example is typical of the wider group it is used to describe. Small or one-sided samples are unrepresentative and cannot support general claims.
Hasty generalisation
Examiner keyword
Drawing a broad conclusion from too few or unrepresentative cases — e.g. inferring a 'miracle cure' from a single recovery.
Currency (of evidence)
Examiner keyword
How up to date evidence is. Outdated data may no longer describe present conditions and is a common, easily-spotted weakness.
Further evidence
Examiner keyword
The better evidence an argument would need to substantiate (support) or refute (disprove) its claim — e.g. a larger representative sample, an independent source, current data, or a comparison/control group. Naming it is a top-band move.

Common Mistakes and Misconceptions — Evaluating Evidence

The traps other students keep falling into on evaluating evidence questions — taken from recent Cambridge International AS & A Level 9239 examiner reports and mark schemes — and how to avoid them.

✕Naming the evidence type but not evaluating it
9239 Component 1 Q2 — evaluate, don't just identify
Why it happens
Identifying the type feels like a complete answer.
How to avoid it
Naming is analysis; the marks are in judgement. After naming the type, run R-R-S and say how strongly it supports the claim.
✕Assuming statistics are strong and anecdotes weak by default
9239 Component 1 Q2 — evaluate evidence on its merits
Why it happens
Numbers look authoritative; stories look casual.
How to avoid it
Judge each piece on R-R-S, not its format. A biased or tiny-sample statistic can be weak; a well-chosen anecdote can be apt for a modest claim.
✕Using 'biased' as a knock-out word
9239 Component 1 — bias as evaluation, not labelling
Why it happens
Spotting a vested interest feels like the end of evaluation.
How to avoid it
Explain the slant, why it arises, and how it affects the weight you give the evidence — then consider corroboration. A biased source can still hold checkable data.
✕Ignoring relevance — accepting impressive-sounding but off-point evidence
9239 Component 1 Q2 — relevance of evidence
Why it happens
Strong-sounding evidence is taken at face value.
How to avoid it
Always check the evidence actually bears on THIS claim (e.g. 'feeling safer' does not show 'crime fell'). Relevance is the most overlooked R-R-S test.
✕Not matching evidence weight to the size of the claim
9239 Component 1 Q2 — sufficiency relative to the claim
Why it happens
Students rate evidence in the abstract rather than against the claim.
How to avoid it
Ask: how big is the claim, and can this evidence carry that weight? A single case can't support a universal or superlative claim.
✕Saying evidence is weak without saying what would be stronger
9239 Component 1 Q2 — discriminating evaluation
Why it happens
Identifying a weakness feels like the finish line.
How to avoid it
Name the further evidence needed to substantiate or refute the claim (larger/representative sample, independent source, current data, a control). This is the top-band move.
✕Arguing the topic yourself instead of evaluating the source's evidence
9239 Component 1 — engage with the source's evidence
Why it happens
The issues are engaging and students share their own views.
How to avoid it
Keep asking: does THIS evidence support THIS source's claim? Evaluate the source's evidence; do not import outside facts or your own opinion on the topic.

Evaluating Evidence

Detailed Study Notes

At a glance

What you’ll learn

Quick recap

Memorise this

How it’s examined

Take this whole topic with you

1Judging one piece of evidence with R-R-S

2Naming a specific weakness

3Evaluating reliability: vested interest and method

4Spotting a relevance gap

5Matching evidence weight to claim size

6Naming evidence that would substantiate or refute

Evaluating evidence

Relevance, Reliability, Sufficiency (R-R-S)

Relevance

Reliability

Sufficiency

Vested interest

Representativeness

Hasty generalisation

Currency (of evidence)

Further evidence

✕Naming the evidence type but not evaluating it

✕Assuming statistics are strong and anecdotes weak by default

✕Using 'biased' as a knock-out word

✕Ignoring relevance — accepting impressive-sounding but off-point evidence

✕Not matching evidence weight to the size of the claim

✕Saying evidence is weak without saying what would be stronger

✕Arguing the topic yourself instead of evaluating the source's evidence