Some children can find evidence and still build weak answers.
They underline a sentence from the passage.
They add an example to an essay.
They quote a source.
They point to a statistic.
Yet the evidence does not actually prove the claim.
It may be relevant but weak. Strong but incomplete. Accurate but answering a different question. Impressive but unrepresentative. True but insufficient.
This is the Evidence-Quality Gap: the learner knows that good English answers need support but has not yet learned to judge whether the support is relevant, credible, representative and sufficient for the claim being made.
Evidence is not a decoration added after the point
Weak evidence use often follows a formula:
Make a point. Add an example. Move on.
But evidence has a job.
It should increase our reason to believe the claim.
That means the learner must ask not only, “Do I have evidence?” but also:
- Does this evidence actually relate to my claim?
- How directly does it support it?
- Is one example enough?
- Could there be another explanation?
- Does the evidence represent the wider situation?
- How certain should my conclusion be?
SETC’s The Evidence Read owns the explicit reading procedure for proving interpretations. This article owns the parent-facing diagnosis: why a child can locate support yet still use it badly.
Failure pattern 1: any relevant detail is treated as proof
Question:
How do we know the character was frightened?
Child answers:
He was in the forest at night.
The detail is relevant to a frightening setting.
It does not prove the character was frightened.
Better evidence might be trembling, hesitation, avoidance or explicit thought.
Relevance and evidential force are different.
Failure pattern 2: one example becomes a general rule
A learner writes:
Online learning is ineffective because my cousin stopped paying attention during one lesson.
The anecdote may be true.
It cannot support such a broad conclusion by itself.
The scale of the evidence must match the scale of the claim.
SETC’s Generalisation Control in Timed Essays owns the exam-facing version of this reasoning problem.
Failure pattern 3: the child confuses vividness with strength
A dramatic story feels persuasive.
A dry dataset feels less memorable.
But emotional force and evidential force are not the same.
Strong reasoning asks what the evidence establishes, not how memorable it is.
Failure pattern 4: quotation is treated as authority
A quote sounds impressive, so the learner assumes it is strong evidence.
But who said it?
What did that person know?
Were they reporting data, giving opinion or defending an interest?
This connects to SETC’s Perspective-Tracking Gap: evidence quality depends partly on understanding the source behind the statement.
Failure pattern 5: the child does not distinguish claim from evidence
“Experts say this policy is harmful.”
That is still a claim unless we know which experts, what evidence they used and how the conclusion was reached.
Repeating someone else’s conclusion does not automatically supply evidence for it.
Failure pattern 6: evidence is not connected back to the point
The child inserts a statistic, quotation or example and assumes its significance is obvious.
But evidence often needs interpretation.
The reader needs the bridge:
This matters because…
Without that bridge, the paragraph becomes evidence dumping.
This is one reason the Idea Development Gap and evidence quality often appear together.
Failure pattern 7: contradictory evidence is ignored
The learner finds one detail supporting the preferred answer and stops searching.
Later evidence points another way.
Strong reasoning updates.
Weak evidence judgement treats support as a treasure hunt: once one supportive item is found, the job feels complete.
Failure pattern 8: quantity is confused with sufficiency
Three weak examples do not necessarily equal one strong argument.
Evidence should be judged by quality and fit, not merely count.
Sometimes one decisive textual clue is enough.
Sometimes many examples are still inadequate because the claim is too broad.
Failure pattern 9: the child cannot calibrate certainty
Evidence rarely supports every claim with equal force.
Strong language reflects this:
- shows;
- strongly suggests;
- indicates;
- is consistent with;
- may imply;
A child who writes proves when the evidence only suggests has an evidence-quality problem and a language-calibration problem.
SETC’s Uncertainty Calibration in Timed Essays owns the examination-facing precision layer.
Failure pattern 10: evidence selection happens before the claim is clear
The child searches for “good quotes” before deciding what the paragraph needs to establish.
Evidence then drives the argument randomly.
A stronger process starts with the claim, then asks what kind of evidence would genuinely test or support it.
A parent diagnostic: rank the evidence
Give one simple claim and three pieces of possible evidence.
Ask the child to rank them from strongest to weakest and explain why.
Then change the claim slightly.
Does the ranking change?
This reveals whether the learner understands that evidence quality is relational: evidence is strong for a particular claim, not in the abstract.
Repair principle 1: teach relevance before strength
First ask whether the evidence speaks to the claim at all.
Only then ask how strong it is.
Repair principle 2: make claim scale explicit
One person’s experience may support:
This can happen.
It usually cannot support:
This always happens to everyone.
Teach children to shrink or qualify claims when evidence cannot carry the original scale.
Repair principle 3: compare evidence types
- direct quotation;
- observed behaviour;
- statistic;
- example;
- expert interpretation;
- personal anecdote;
- comparison;
Ask what each can and cannot establish.
Repair principle 4: require the reasoning bridge
After evidence, ask:
How does this make the claim more believable?
If the child cannot explain the bridge, the evidence may be decorative.
Repair principle 5: search for disconfirming evidence
After finding support, ask:
What in the text or source would make us less certain?
This prevents confirmation-only reading.
Repair principle 6: calibrate conclusion language
Match verbs to evidential strength.
Evidence may prove, support, suggest, indicate, weaken or merely be consistent with a claim.
Language becomes a measurement instrument for reasoning.
Repair principle 7: debate evidence, not only conclusions
IES Reading for Understanding work used discussion and debate to develop reasoning and comprehension.
A useful discussion question is not merely:
Which side do you agree with?
but:
Which side currently has the stronger evidence, and what additional evidence would change your mind?
The evidence-quality ladder
- Locate possible evidence.
- Check relevance.
- Identify source and evidence type.
- Judge directness and strength.
- Explain the reasoning bridge.
- Check sufficiency.
- Look for contradiction or limitation.
- Calibrate the certainty of the conclusion.
- Revise the claim if the evidence cannot support its original scale.
What not to confuse with the Evidence-Quality Gap
- If the child cannot locate evidence at all, reading comprehension may be upstream.
- If the evidence is strong but the child cannot explain it, idea development may be the main issue.
- If the child confuses who made a claim, perspective tracking may be involved.
- If the learner lacks relevant knowledge to judge plausibility, background knowledge may limit evaluation.
What improvement should look like
- Evidence is selected for fit, not impressiveness.
- Examples are less likely to become sweeping generalisations.
- Quotations are attributed and interpreted.
- Contradictory evidence is considered.
- Claims become more appropriately qualified.
- The child explains why evidence supports the point.
- Answers distinguish relevance, strength and sufficiency.
The deeper lesson
Finding evidence is only the first half of reasoning.
The second half is deciding what the evidence deserves to let us say.
That is where many apparently strong English answers fail.
The useful question is:
Does my child know not only where the evidence is, but how much weight that evidence can carry?
When the answer becomes yes, comprehension becomes reasoning and writing becomes more trustworthy.
Research spine
- IES — Catalyzing Comprehension Through Discussion and Debate
- What Works Clearinghouse — Improving Adolescent Literacy
- What Works Clearinghouse — Teaching Secondary Students to Write Effectively
Continue with the Why English Can Feel Difficult guides or browse The Library.