Nature and meaning of science

Incorrect predictions indicate that the model's assumptions or limits need review. Revise models according to evidence.

Without sample selection details it is difficult to know whether the group was representative. Check sampling transparency.

Improvement based on new evidence reflects the self-correcting nature of science. Treat revision as progress.

Scientific conclusions must rest on observations, experiments, measurements, and testable evidence. A claim may have been accepted for a long time and still be wrong or incomplete, so tradition alone does not establish its truth. Option B is incorrect because old ideas can also be tested using experiments and new evidence. Exam tip: Treat statements such as “it has been believed for years” as claims needing evidence, not as evidence themselves.

The scale division indicates the smallest difference that can be read reliably. Identify instrument limitations.

A result that is repeated consistently but differs from the true value suggests a systematic error. Such an error may arise from incorrect calibration, a zero error, or an instrument fault, so checking instrument calibration is most useful. Taking an average can reduce random error, but it does not remove a systematic error in the same direction. Exam tip: for consistent but inaccurate readings, first consider systematic error and calibration.

A scientific conclusion may be limited to particular samples and conditions. Understand limits of applicability.

If the difference is smaller than the uncertainty its reality is unclear. Consider measurement limits.

A falsifiable claim allows a test whose result could oppose it. Link falsifiability with testability.

A prior plan reduces bias from changing analysis after seeing the results. Link it with transparency.

Contamination of the control group weakens comparison of the true treatment effect. Check clear separation of groups.

A study generally supports a hypothesis rather than proving it as final truth. Identify cautious scientific wording.

Repetition increases reliability but generalisation requires testing under new conditions. Distinguish reliability from external validity.

Personal preference cannot be universally settled by evidence. Distinguish factual questions from value questions.

Constructive criticism examines claims and methods rather than attacking individuals. Choose evidence-based review.

Direct answer: Option A, how much the values varied. An average gives one central value, but it does not show whether the individual measurements were close together or widely separated. For example, two groups can both have an average of 10: one may contain 9, 10, and 11, while another may contain 1, 10, and 19. Their averages are equal, but their spread is very different. Step by step: collect all values; calculate the average; examine how far values lie from that average; report a measure or description of spread. Option A is correct because missing spread hides variation, consistency, and possible unusual values. Option B is wrong because the independent variable can usually be named separately in the method or table. Option C is wrong because the experiment date is not what spread describes. Option D is wrong because scientist experience is unrelated to numerical variation. Memory cue: average tells the centre; spread tells how scattered the data are.

When other influencing factors are controlled the cause of the observed effect is clearer. Link internal validity with control.

A broad explanation accounts for many related observations within one framework. Identify explanatory power.

The direct answer is A: the test is not fair or falsifiable. A scientific claim must be tested in a way that could show either supporting or opposing evidence. Here, the investigator has planned only for results that agree with the claim. Step by step: first state the claim; next decide what observation would support it and what observation would disagree with it; then collect evidence without ignoring inconvenient results. If opposing results were never possible, the claim cannot be properly checked and the test is biased. Option A is correct because it identifies both the unfair selection of evidence and the lack of falsifiability. Option B is wrong because a diverse sample is not automatically a weakness and may improve representation. Option C is wrong because a precise instrument usually improves measurement. Option D is wrong because clear results are helpful, not a defect. Memory cue: a good scientific test must allow the claim to fail.

Confidentiality protects participants' private information. Identify principles of research ethics.

Ethical research must protect participants from unreasonable harm and must consider whether the expected benefits justify the risks. If the possible risks are greater than the benefits, the study should not simply continue as planned. Researchers should pause it, review the design, reduce the risks where possible, and stop it if safety cannot be protected. Participants should receive clear information and should not be exposed to hidden or avoidable danger. Publishing results first does not make an unsafe study ethical.

A review may show that the sample, procedure, equipment, or consent process needs to change. An ethics committee or responsible research body can examine these issues before work resumes. Therefore, option A is correct: the study requires review or stopping. Hiding risks and withholding information violate responsible research practice. The central principle is that participant welfare has priority over convenience, speed, or a desired result.

Confidence increases when different independent lines of evidence point to the same conclusion. Identify converging evidence.

Scientific knowledge is strengthened through collective checking and communication along with individual work. Understand its collaborative nature.

New evidence can clarify the conditions under which a claim applies. Treat revision as refinement.

Science is not only a body of knowledge but also a process of testing and improvement. Understand it as both knowledge and process.