Hypothesis and experimentation

A ruler divided only into centimetres cannot accurately resolve small millimetre differences. Choose an instrument with finer divisions.

Not knowing group identity can reduce expectation based observation bias. Apply the same measurement rules to both groups.

A small inconsistent difference may indicate a weak effect or natural variation. Use more samples and repetitions to strengthen the test.

Scientific hypotheses can change according to evidence. Test the revised hypothesis with a new experiment.

Different instruments may produce different readings. Use the same calibrated instrument or verified equivalent instruments.

Repetition by the same student shows repeatability but not independent reproducibility. Testing by another person or group is needed.

Terms such as ‘good’ and ‘poor’ can mean different things to different observers, so they do not provide objective data. The dependent variable should be recorded using clear measurable criteria, such as plant height in centimetres, percentage germination, or reaction time. Changing ‘good’ to ‘very good’ is still subjective. Exam tip: Record experimental outcomes with numbers, units, or pre-decided observation criteria.

The control group should remain free from the treatment. Exposure contaminates the baseline for comparison.

The first statement can be contradicted by experimental results so it is falsifiable. A scientific hypothesis must be testable.

With different initial values percentage change can show relative growth. Keep both original values and calculations.

A procedural change may affect the result and must be recorded. Scientific records should be complete and honest.

Direct answer: Option A, testing different water amounts on seed germination, most needs a large sample size. Individual seeds naturally differ in viability, age, genetics, and health. If only a few seeds are used, one unusually weak or strong seed can strongly affect the result. Using many seeds in each water group reduces the influence of individual differences, makes averages more stable, and helps reveal the general effect of water amount. Option A is correct because germination is a biological process with natural variation. Option B is wrong: calibrating one thermometer concerns one instrument, not a variable population of living subjects. Option C is wrong for the same reason; checking one ruler does not require many specimens. Option D is wrong because measuring one container is a single-object measurement, not a sample-based biological comparison. The groups should also be similar, randomly assigned, and given equal light and temperature. Memory cue: biological variation calls for replication and larger samples.

The direct answer is A, the light condition also changed. In a fair fertiliser experiment, fertiliser should be the only important difference between treatment and control groups. Light affects photosynthesis and plant growth, so placing one group near a window and the other in darkness introduces a second cause. Step by step: the treatment is fertiliser; the measured result may be plant growth; both groups must receive comparable water, soil, temperature, pot size and light; here light differs; if the groups grow differently, we cannot know whether fertiliser or light caused it. A is correct. B is wrong because fertiliser is the independent variable being tested, not a controlled variable. C is wrong because height is usually a measured result or starting condition, not the deliberately changed factor here. D is wrong because a dark control group is not automatically more reliable; unequal light makes comparison unfair. Memory cue: one experiment should change one main variable at a time.

A difference smaller than instrument capability cannot be distinguished accurately. Measure again with a finer instrument.

Prior expectation can influence vague classification. Blind testing and clear criteria reduce bias.

Several supportive trials increase confidence but do not establish permanent final truth. Science remains open to new evidence.

Clear columns for both variables help comparison and trend identification. Include headings and units.

Direct answer: Option A, repeatability, will decrease first. Repeatability means obtaining comparable results when the same method is performed again under the same conditions. If the method changes slightly between trials, then a difference in results may come from the changed method rather than from the factor being studied. The trials are no longer directly comparable, so confidence in repeated performance falls. Option A is correct. Option B is irrelevant because colour purity is not a general quality of experimental procedure. Option C is wrong because changing the method does not automatically change how many samples were used. Option D is wrong because the length of a hypothesis is unrelated to procedural consistency. To protect repeatability, write a precise method, use the same equipment and quantities, keep conditions constant, and follow the same sequence each time. Memory cue: same method, repeatable result.

A negative control shows the normal outcome without active treatment. It provides comparison for changes caused by the treatment.

A new hypothesis formed from existing data is not independently supported by the same data. It requires a new test.

A one plant control group may be strongly affected by an unusual individual. Use adequate and comparable sample sizes in both groups.

A result from one species and environment may not apply elsewhere. Test under different conditions before making broad conclusions.

A null hypothesis usually states that there is no effect or difference. It is tested using data.

A hypothesis offers a possible explanation while an experiment gathers evidence for or against it. The conclusion should use all data.

With large variation a difference in averages alone may not be sufficient evidence. More samples and repetitions help clarify the reliability of the result.