Secondary 3–4 learners increasingly need vocabulary that lets them judge not only what evidence says, but how much confidence it deserves. Words such as reliable, credible, relevant, sufficient and representative describe different strengths and limits.
This guide develops those distinctions for comprehension, argumentative writing, source evaluation and analytical explanation. It is an original eduKateSG learning guide, not an official examination paper.
Why Evidence Vocabulary Matters
“There is evidence” is not enough. The next question is what kind of evidence it is and whether it supports the exact claim being made.
A source can be reliable but irrelevant. Evidence can be relevant but insufficient. A survey can be large but unrepresentative.
Strong vocabulary helps the learner keep these jobs separate.
The Hidden Learning Problem: One Positive Word for Everything
Students often use good evidence, strong evidence or trustworthy evidence without identifying why.
The repair is to ask one discriminating question at a time:
- Can the information be depended on?
- Is the source believable for this claim?
- Does the evidence actually address the question?
- Is there enough evidence?
- Does the sample reflect the wider group?
Reliable
Reliable describes information, a method or a source that performs consistently or can be depended on within the relevant conditions.
Example: repeated measurements from a well-maintained instrument may be reliable if the method is consistent and appropriate.
Reliability does not automatically prove relevance to a different question.
Credible
Credible means believable or worthy of confidence, often because the source has relevant expertise, transparency, evidence or accountability.
A credible source can still make a mistake. Credibility is not infallibility.
Relevant
Relevant means directly connected to the claim or question being evaluated.
Evidence about student enjoyment may be relevant to engagement but not sufficient to prove academic improvement.
Sufficient
Sufficient means enough for the stated purpose.
One anecdote may be relevant but insufficient to support a broad population claim.
Sufficiency always depends on the claim.
Representative
Representative describes a sample or set of observations that reasonably reflects the wider population or range it is intended to stand for.
A survey of only one highly specialised group may not represent all students.
A First-Principles Evidence Check
- What exact claim is being made?
- What evidence is offered?
- Is it relevant to that claim?
- Is there enough of it?
- How dependable is the source or method?
- Does the sample reflect the group being discussed?
Worked Example
Claim: “All Secondary students prefer digital worksheets.”
Evidence: eight students from one class said they preferred them.
The evidence may be relevant to preference. It may even be credible if their responses were recorded accurately. But it is not automatically sufficient for the claim “all Secondary students,” and the sample may not be representative of the wider population.
The vocabulary reveals exactly where the claim becomes too broad.
Practice A — Choose the Evidence Word
A1. A source directly discusses the question being asked. → relevant.
A2. A measurement method gives consistent results across repeated trials. → reliable.
A3. A small sample cannot support a broad national claim. → insufficient / not sufficiently representative.
A4. A qualified expert explains the method and provides supporting data. → potentially credible, subject to evaluation.
Practice B — Separate Relevant From Sufficient
One student reports that a new timetable reduced stress.
This may be relevant evidence for that student’s experience. It is not sufficient by itself to establish the effect on every student.
Write one sentence preserving both points.
Model: “The student’s report is relevant to individual experience, but one response is insufficient to establish a school-wide effect.”
Practice C — Representative Samples
A school surveys only members of the debate team about confidence in public speaking.
Ask:
- Is the survey relevant to public-speaking confidence?
- Is the sample likely to represent all students?
- What additional sampling would improve the claim?
The key distinction is that relevance and representativeness are different.
Practice D — Credibility Without Blind Trust
“An expert said it, so it must be true” is too strong.
A more disciplined response asks whether the expertise is relevant, whether evidence is provided, whether the claim falls within that expertise, and whether other evidence is consistent.
Credibility increases confidence; it does not remove the need for reasoning.
Practice E — Build an Evaluation Paragraph
Fictional evidence: a school trials a new revision schedule in one class for two weeks. Most students report liking it, but no test results are compared.
Model: “The student feedback is relevant to perceived usefulness, but the evidence is insufficient to establish an improvement in academic performance. Because the trial involved one class for a short period, the findings may not be representative of the wider cohort.”
The paragraph uses different evaluative words for different limits.
Common Secondary 3–4 Errors
- calling any source reliable without evaluating the method;
- confusing credibility with proof;
- treating relevant evidence as automatically sufficient;
- ignoring sample representativeness;
- using “biased” as a catch-all criticism without identifying the mechanism;
- judging a source by appearance rather than evidence quality.
How We Teach Evidence Evaluation in Small Groups
Warm-up retrieval
Students classify short evidence descriptions.
Single-criterion fence
Relevant vs irrelevant comes before sufficiency and representativeness.
Worked claim-evidence pairs
Students state exactly what the evidence supports.
Think-aloud evaluation
Students explain which evidence word they chose and why.
Mixed source work
Several strengths and weaknesses appear together.
Error review
The tutor distinguishes vocabulary confusion from reasoning or comprehension gaps.
Three Learner Pathways
Repair
The student calls evidence simply good or bad. We build one evaluation dimension at a time.
Stabilisation
The student knows the terms but applies them inconsistently. We compare near-neighbour cases.
Extension
The student moves into source triangulation, bias, uncertainty, causal claims and evidence weighting.
What Progress Should Look Like
- more precise source evaluation;
- fewer blanket judgements;
- better distinction between relevance and sufficiency;
- more careful population claims;
- stronger argumentative writing;
- better calibration of evidence and conclusion.
Frequently Asked Questions
Does reliable mean true?
No. Reliable generally concerns dependability or consistency. A method can be consistent yet systematically wrong if it is badly calibrated.
Does credible mean unbiased?
No. Credibility and bias are separate dimensions.
Can one example ever be sufficient?
Yes for some narrow claims, but usually not for broad population claims. Sufficiency depends on the exact claim.
Why teach representative separately?
Because a large amount of evidence can still mislead if it comes from a sample that does not reflect the group being discussed.
Helpful Reading
- Evidence Verbs and Claim Strength
- Quantifiers and Scope
- Evaluation Language
- Secondary 3 Vocabulary Owner
- Secondary 4 Vocabulary Owner
Next Step
The goal is a learner who does not merely ask whether evidence exists, but whether it is the right evidence, enough evidence, and evidence that deserves the confidence being placed in it.
Continue through the Vocabulary Learning Hub or Secondary English Tuition.