Hypothesis and experimentation

Changing two factors together makes the cause of germination differences unclear. In exams identify mixed causes.

If the effect is similar to measurement uncertainty a clear conclusion cannot be drawn. In exams compare the difference with the error range.

An instrument with finer divisions is needed to measure five millilitres reliably. In exams check both range and resolution.

A negative control provides a baseline without active treatment and helps detect unwanted effects. In exams view it as the no-effect comparison group.

Random allocation distributes initial sample characteristics more fairly between groups. In exams link it with reducing selection bias.

Different acid volumes change the total amount of reactant and may confound the effect of concentration. In exams keep volume constant.

A similar change in the control group suggests a cause other than the treatment. In exams understand the importance of control responses.

Using a different reference point each time makes the measurements incomparable. In exams use a fixed starting point.

Repeated opposing results under controlled conditions reduce the likelihood of chance. In exams look for repetition and control as strong evidence.

Without units the physical meaning of numbers is unclear. In exams include units with every measurement heading.

A bar graph is more suitable for comparing separate categories. In exams choose the graph according to the type of data.

When other influential factors are controlled the result can be linked to the independent variable. In exams connect validity with clarity of cause.

Different final gas volumes mean the trials are not comparing the same extent of reaction. In exams ensure a common endpoint.

An explanation formed after viewing data is a new hypothesis requiring an independent test. In exams distinguish prior prediction from post-data explanation.

Results obtained under highly specific conditions may not apply in general settings. In exams identify limits of generalisation.

Comparison with a known standard can reveal systematic calibration error. In exams use a standard to check instruments.

One trial may be influenced by chance or error so repetition is necessary. In exams do not treat one result as absolute proof.

Results from different procedures are not directly comparable. In exams keep the method constant during repetitions.

Knowledge of treatment can influence the observer's judgement. In exams link blind assessment with reducing bias.

Correlation alone does not prove causation because other factors may affect both variables. In exams distinguish association from cause.

More repetition can reduce random error but cannot remove a constant systematic error. In exams recognise the need to correct the apparatus.

The large mass of the pot and soil may obscure the actual change in plant mass. In exams make measurements sensitive to the target variable.

Without clear steps another person cannot repeat the same experiment. In exams include details such as quantities time and apparatus.

Predefined intervals keep the procedure objective and consistent. In exams understand the value of planning the method beforehand.

Analysing all groups reduces selection bias and represents the full evidence. In exams base conclusions on the complete data set.