A source can be famous and still be weak evidence. It can be accurate on one question and irrelevant to another. It can be expert yet conflicted, recent yet methodologically poor, authoritative yet unsupported.
The broader eduKateSG guide, Vocabulary for Junior College (JC1–JC2): Adult Thinking, Professional Writing, treats vocabulary as intellectual control. Source-evaluation vocabulary gives students a way to explain why evidence deserves more or less weight.
Credibility is not the same as truth
Credibility concerns whether a source deserves belief. It may depend on expertise, transparency, track record, method and incentives. A credible source can still be mistaken. An unknown source can still be correct. The term describes warranted trust, not infallibility.
Reliability concerns consistency
A reliable measurement, process or source produces reasonably consistent results under comparable conditions. Reliability does not guarantee validity: a method can produce the same wrong answer repeatedly.
Corroboration strengthens through independence
Evidence is corroborated when independent sources point towards the same conclusion. Ten articles repeating one original claim are not ten independent confirmations. The student should ask where the information ultimately comes from.
Authority is domain-specific
An expert in economics is not automatically an expert in epidemiology. A prestigious institution may be authoritative on one dataset but not on a moral judgement. Useful words include relevant expertise, domain authority, institutional competence and scope of expertise.
Independence and incentives matter
A source may have financial, political or reputational incentives that shape what it emphasises. This does not automatically invalidate the source. It creates a reason to inspect methods and seek corroboration.
- potential conflict of interest
- financial incentive
- institutional interest
- independent review
- transparent methodology
Recency and relevance are different
The newest source is not always the best source. A recent commentary may be less useful than an older definitive dataset. Recency matters when conditions change quickly; relevance matters always.
A source-evaluation sentence
The evidence deserves moderate weight because it comes from a relevant national dataset and is independently reproducible, although its age limits how confidently it can describe current behaviour.
This sentence is useful because it contains both support and limitation.
The six-question source audit
- Who produced the information?
- What relevant expertise do they have?
- How was the evidence generated?
- What incentives or conflicts exist?
- Is it independently corroborated?
- Is it sufficiently current and relevant for this claim?
Vocabulary for calibrated trust
- authoritative
- credible
- well-substantiated
- independently corroborated
- methodologically limited
- preliminary
- uncorroborated
- context-specific
The mature writer does not divide sources into trustworthy and untrustworthy as though credibility were binary. They assign weight.
Why this matters beyond exams
Modern adults make decisions inside an environment of abundant claims. The ability to explain why one source deserves greater weight than another is a core literacy skill, not merely a comprehension technique.
Source Judgement Lab: Trace the Claim Back to Its Origin
Source evaluation becomes much stronger when students distinguish the source carrying the claim from the source that generated the evidence. A newspaper article may report a university study; ten websites may then repeat the newspaper. That is one evidential origin travelling through many channels, not eleven independent confirmations.
Provenance: Where Did This Claim Come From?
Provenance means the history and origin of information. Ask who collected the data, who analysed it, who summarised it and whether later sources changed the wording. Provenance is especially important online, where repetition can create the appearance of consensus.
Primary, Secondary and Commentary Sources
- Primary evidence: original dataset, study, document, speech or direct record.
- Secondary analysis: interpretation or synthesis of primary evidence.
- Commentary: argument or opinion built partly on evidence and partly on judgement.
None is automatically superior. The best choice depends on the claim. A primary dataset may be ideal for a numerical fact; a high-quality synthesis may be better for understanding a large research field.
Reliability, Validity and Credibility
Reliability asks whether a method or measurement is consistent. Validity asks whether it measures what it claims to measure. Credibility asks whether the source deserves belief overall. A questionnaire can be highly reliable—producing similar answers repeatedly—yet invalid if its questions measure the wrong concept.
Authority Without Deference
Expertise deserves weight but not automatic obedience. Ask whether the expertise matches the domain, whether the source provides reasons, and whether independent evidence supports the conclusion. “An expert says so” is weaker than “a relevant expert makes a claim supported by transparent evidence and corroborated methods”.
Incentives and Conflicts
A conflict of interest is a reason to inspect evidence more carefully, not an automatic proof of falsehood. A company-funded study may still be rigorous; an apparently neutral source may still have reputational or ideological incentives. Mature evaluation uses words such as potential conflict, institutional incentive, disclosed interest and independent replication.
Corroboration Must Be Independent
Two sources only corroborate one another when their evidential pathways are sufficiently independent. If both rely on the same dataset, they may provide two interpretations rather than two confirmations. Trace the chain before counting sources.
The Seven-Dimension Source Card
- Relevant expertise
- Evidence-generation method
- Validity of measurement
- Reliability/consistency
- Independence and incentives
- Recency and applicability
- Corroboration by other evidence
Worked Evaluation Sentence
“The source deserves substantial but not decisive weight: the data come from a large, relevant national sample and the method is transparent, but the findings are self-reported and therefore less reliable for measuring actual behaviour than attitudes.” Notice that the sentence assigns weight rather than declaring the source simply trustworthy or untrustworthy.
Source Comparison Drill
Take two sources making the same claim. Compare their provenance, method, domain expertise, incentives and recency. Then write a two-sentence judgement explaining which deserves greater weight and why. The task is not to choose the source you agree with. It is to explain the evidential difference.
Final Source Check
- Am I evaluating the source or merely its reputation?
- Can I trace the claim to original evidence?
- Are apparently independent sources actually repeating one origin?
- Does the evidence measure the concept being claimed?
- What limitation should reduce—not erase—the weight I give it?
Source literacy is the discipline of calibrated trust. Mature readers neither believe automatically nor reject reflexively; they ask how much confidence the evidence has earned.
Continue with the full JC1–JC2 Vocabulary: Adult Thinking, Professional Writing system.