Science and health discussions become unreliable very quickly when words such as “prove”, “risk”, “safe”, “natural” and “effective” are used loosely. Secondary 3 students need vocabulary that separates observation from inference and possibility from certainty.
This guide complements the Top 100 Vocabulary List Secondary 3 Grade A1, especially empirical, hypothesis, inductive, deductive, inference, evaluate and systematic.
Observation
An observation records what is measured or noticed. It does not automatically explain why the pattern exists.
Hypothesis
A hypothesis is a proposed explanation that can be tested against evidence. A good hypothesis is specific enough that observations could support or weaken it.
Empirical evidence
Empirical evidence comes from observation or experiment. Its strength depends on how well the evidence was collected, measured and interpreted.
Correlation and causation
A correlation is an association between variables. Causation means one factor contributes to producing another. Confounding variables, reverse causation and chance can complicate interpretation.
Risk
Risk combines probability and consequence. A rare event can still deserve attention if the potential harm is severe; a common event may be less serious if consequences are small.
Relative and absolute risk
A percentage increase can sound dramatic without showing the starting level. If risk rises from one in ten thousand to two in ten thousand, it has doubled relatively while remaining small absolutely.
Prevention and treatment
Prevention reduces the chance that a problem occurs; treatment responds after it develops. Public-health discussions often require both.
Efficacy and effectiveness
Efficacy often concerns whether an intervention works under controlled conditions. Effectiveness concerns how well it works in real-world settings. The distinction helps explain why results may change outside trials.
Benefit and harm
A responsible health discussion compares expected benefits, possible harms and uncertainty. “Safe” rarely means zero risk; it usually means risks are acceptably low relative to benefits under defined conditions.
Population and individual
Population-level evidence describes patterns across groups; individual outcomes can differ. A treatment that helps most people may not help every person.
Bias
Bias can enter through sampling, measurement, analysis or reporting. It is a systematic influence, not simply personal prejudice.
Replication
Replication tests whether findings can be observed again. One study can be informative without being the final word.
Consensus
Scientific consensus is broad agreement among relevant experts based on accumulated evidence. It is not unanimity and can change when strong new evidence emerges.
Uncertainty
Uncertainty is not ignorance. It can often be estimated and described. Honest uncertainty strengthens science by showing where conclusions are robust and where evidence remains limited.
Anecdotes
Personal experience can reveal possible effects and human impact but usually cannot establish how often an outcome occurs or what caused it.
Worked discussion: health claims
“A viral story about one person’s recovery is not enough to establish treatment effectiveness. Stronger evaluation requires empirical evidence from appropriately designed studies, comparison groups where relevant, and attention to harms as well as benefits.”
Worked discussion: science communication
“Good science communication should be succinct without becoming misleading. Simplifying jargon is useful, but removing uncertainty or limitations can create false confidence.”
A topic vocabulary bank
- hypothesis
- empirical
- correlation
- causation
- risk
- efficacy
- effectiveness
- bias
- replication
- consensus
- uncertainty
- prevention
Practice routine
Choose one health or science claim. Identify the evidence type, population, comparison, possible confounders, effect size, uncertainty and whether the claim describes correlation or causation.
Return to the hero
Use the Secondary 3 Grade A1 hero to reinforce science reasoning through empirical evidence, hypotheses, inference, systematic analysis and careful conclusions.
Checkpoint
- Can you distinguish observation from explanation?
- Can you distinguish correlation from causation?
- Can you interpret risk more carefully?
- Can you distinguish efficacy from effectiveness?
- Can you discuss uncertainty without treating it as failure?
- Can you evaluate anecdotes against broader evidence?
If so, science vocabulary becomes a discipline for saying exactly what the evidence allows—and no more.
Deep Science Lab: Claim Strength, Effect Size, Risk and Uncertainty
Science and health vocabulary becomes more reliable when students separate four questions: what was observed, how large the effect appears to be, how certain the evidence is, and how the result should change a decision. A dramatic-sounding percentage can still describe a small absolute change, while a modest-looking change can matter greatly when the consequence is serious.
The Claim-to-Evidence Ladder
| Evidence state | Safer language | What remains open |
|---|---|---|
| single observation | may indicate | chance, measurement error, alternative explanations |
| repeated consistent observations | supports the hypothesis | scope and causation may still be limited |
| multiple strong independent studies | provides substantial evidence | new contexts and subgroups may differ |
| mixed findings | evidence remains inconclusive or context-dependent | which conditions explain the difference |
Effect Size Matters
Students sometimes treat “a difference exists” as equivalent to “the difference is important”. Ask how large the change is and whether it matters in practice. A tiny measurable effect may be real but practically limited. Conversely, even a small change may be important when applied across many people or when the outcome is serious. Vocabulary such as modest, substantial, negligible and clinically or practically meaningful should be tied to context rather than used automatically.
Relative Risk and Absolute Risk
Relative changes can sound dramatic because they compare one rate with another. Absolute risk keeps the starting level visible. If an event changes from one case in a large group to two cases in the same-sized group, the relative increase is large while the absolute change remains small. Good interpretation reports enough context for the reader to understand both.
Population Evidence and Individual Decisions
Research can describe what tends to happen across groups without predicting one person’s outcome perfectly. A population average is useful evidence, but individual response can differ because of age, circumstances, biology, behaviour or other variables. Vocabulary should preserve this distinction rather than turning “on average” into “for everyone”.
Replication and Convergence
One study can raise a serious possibility. Confidence grows when findings are reproduced and when different methods point toward the same explanation. This is why replication, corroboration and converging evidence are more useful than treating one striking result as final proof.
Uncertainty Has Structure
Uncertainty may come from small samples, noisy measurements, incomplete mechanisms, disagreement across studies or limited transfer to new populations. Naming the source of uncertainty is stronger than simply saying “we are not sure”.
The Health-Claim Audit
- What exactly is the claim?
- What population or condition does it apply to?
- What kind of evidence supports it?
- Is the claim about correlation, causation, efficacy or effectiveness?
- How large is the effect?
- What harms, costs or uncertainty remain?
- What conclusion is proportionate to the evidence?
Deep Practice Set
- Take one general science claim and rewrite it at three certainty levels.
- Create one example where relative change sounds more dramatic than absolute change.
- Explain why one anecdote cannot establish prevalence.
- Identify one possible confounder in a correlation.
- Use science-reasoning vocabulary from the Secondary 3 Grade A1 hero to write a conclusion that is precise without becoming timid.
Scientific vocabulary reaches maturity when it lets students say exactly how much the evidence supports, how large the effect is, and what uncertainty still belongs in the conclusion.