Scientific and technological skills: experimentation, modeling, and communication
This domain tests how you design an investigation, control variables, interpret data, use models, and communicate results. Questions often present a procedure or graph and ask what conclusion is supported, what variable is controlled, or how to improve reliability.
Experimentation
A hypothesis is a testable prediction. The independent variable is changed by the experimenter; the dependent variable is measured; controlled variables are kept constant. A control group provides a baseline without the treatment. Repeat trials and larger samples improve reliability. A fair test changes only the independent variable. Safety, ethical treatment, and accurate measurement are part of a valid protocol.
Data and graphs
Tables organize raw data with units. Graphs should have labelled axes with units, a suitable scale, and a clear title. A line graph suits continuous change; a bar graph compares categories. Accuracy is closeness to the true value; precision is consistency among repeated measurements. Anomalous points should be checked, not automatically deleted. Correlation does not prove causation.
Modeling and communication
A model represents a system to explain or predict behaviour, but it has limits. Particle diagrams, food webs, circuit schematics, and scale drawings are models. Good communication includes a clear purpose, a reproducible method, results with units, and a conclusion tied to evidence. Technological reports also state constraints, tests, and design revisions.
Worked example
A student tests whether fertilizer affects plant growth. Three plants receive different fertilizer amounts, and height after four weeks is measured. The independent variable is fertilizer amount; the dependent variable is height; controlled variables include plant species, light, water, soil, and pot size. If all plants also receive different amounts of water, the experiment is confounded: water could explain the height difference. A conclusion that fertilizer caused growth would be unsupported until water is controlled.
Traps
- Calling any prediction a hypothesis without a test.
- Changing more than one variable at a time.
- Confusing accuracy with precision.
- Drawing a cause-and-effect conclusion from correlation.
- Omitting units or labelling axes inconsistently.
- Treating a model as the real system rather than a limited representation.