Scientific Reasoning and Data Interpretation
Scientific Reasoning and Data Interpretation
This domain tests experimental design, graph reading, uncertainty, and evidence-based conclusions. You must distinguish observation from inference, identify variables, and judge whether a claim follows from the data.
Core rules
- The independent variable is manipulated; the dependent variable is measured; controlled variables are held constant.
- A valid conclusion requires a representative sample, a suitable control, and a repeatable method.
- Correlation does not establish causation; a confounding variable may explain both.
- Report uncertainty with the measurement; use the same precision as the instrument.
- Percentage error=|measured−accepted|/accepted×100%.
- A line of best fit summarizes trend; outliers need a stated reason, not automatic deletion.
Worked example
Four trials of a reaction time give 12.1, 12.3, 11.9, and 12.7 s; the accepted value is 12.0 s. Mean=(12.1+12.3+11.9+12.7)/4=12.25 s. The range is 0.8 s, showing scatter; report the mean to a sensible precision, about 12.3 s, and state the spread. Percentage error=|12.25−12.0|/12.0×100≈2.1%. A claim that the method is “exactly accurate” is unsupported because the trials vary; a claim of small systematic bias would need more evidence.
Traps that cost marks
- Calling a difference “significant” without a statistical test or error estimate.
- Changing two variables in one experiment.
- Reading a graph from a non-zero origin as if it were zero.
- Dropping an outlier without justification.
- Confusing precision (repeatability) with accuracy (closeness to true value).