AS91187 Earth and Space Science 2.1 - Practical Earth and Space Science Investigation
What this standard tests
AS91187 assesses a practical investigation. You plan a method, collect primary data, process it, interpret the pattern and evaluate the investigation. The context may be geological, atmospheric, astronomical or ecological, but most marks sit in variable control, evidence handling and uncertainty - not in choosing an impressive topic.
Core rules and concepts
- A testable aim links the independent variable (IV) to the dependent variable (DV). A hypothesis predicts a direction or relationship, not just “it will change”.
- Control variables must be held constant because they could otherwise explain the DV. A control group is only needed when comparing against a no-treatment or baseline condition.
- Reliability improves with repeats and larger samples. Validity means the method measures what it claims. Accuracy is closeness to a true value; precision is repeatability or instrument resolution.
- Record raw data in a table with units and consistent significant figures. Calculate means only from repeats, keep the range, and graph IV on the x-axis with labelled units.
- A conclusion states the pattern, direction, magnitude and uncertainty. An evaluation identifies one real limitation and a specific improvement.
Worked example
A student tests how slope angle affects infiltration time. Three trays use the same dry sand, compaction, water volume (200 mL) and pour height; slope angles are 5°, 15° and 25°. The DV is time until ponding. Repeats give mean times of 42 s, 29 s and 18 s, with ±2 s timing uncertainty. The conclusion is that ponding occurs sooner as slope increases, consistent with faster downslope flow leaving less time for infiltration. The reasoning is sound because sand, compaction and water volume were controlled; the main limitation is the small tray scale, so the result should not be extrapolated to a hillslope without further testing.
Traps that cost marks
- Changing two variables and then attributing the result to one.
- Calling repeats a validity improvement; repeats mainly improve reliability.
- Deleting an anomaly without justification or ignoring uncertainty.
- Drawing a conclusion outside the range tested.
- Giving a limitation without saying how it affects the result.