Nature of Science
Domain I examines how Earth and space scientists build and defend knowledge: the logic of hypothesis testing, the design of controlled investigations, measurement quality, and the interpretive skills used to read data, maps, and models. Expect items that ask you to identify variables, judge whether a conclusion is supported, and separate observation from inference.
Hypothesis, theory, and law
- A hypothesis is a testable, falsifiable prediction about a specific situation; a theory is a broad, repeatedly tested explanatory framework (plate tectonics, biological evolution); a law is a concise statement of a regularity (Kepler's laws, superposition). They are different kinds of statements, not rungs on a ladder toward proof.
- Science supports or refutes; it does not prove. A single contradictory result is weighed against the total evidence, and replication plus peer review carry the load.
Experimental design and measurement
- The independent variable is manipulated; the dependent variable is measured; controlled variables are held constant; a control group isolates the treatment.
- Precision is reproducibility (tight clustering); accuracy is closeness to the accepted value. Random error scatters data; systematic error shifts it in one direction, and averaging more trials will not remove it.
- Correlation alone cannot establish causation - look for confounding variables.
Historical science and deep time
- Uniformitarianism/actualism: processes observed today (erosion, deposition, volcanism) are used to interpret ancient rocks.
- Relative dating uses superposition, original horizontality, cross-cutting relationships, and inclusions; absolute dating uses radiometric decay and half-lives.
Worked example
Five density trials of a mineral give 2.6, 2.7, 2.6, 2.7, and 2.6 g/cm³; the accepted value is 3.0 g/cm³. The spread is only 0.1 g/cm³, so precision is high. The mean, 2.64, misses the accepted value by 0.36, so accuracy is poor. Because every trial is low, the cause is likely systematic - a miscalibrated balance or a consistently misread displacement volume - not random scatter. More trials would tighten the cluster around the wrong value, so the instrument or procedure must be fixed first.
Traps that cost marks
Calling a well-supported theory just a theory or claiming a hypothesis is proven; swapping accuracy and precision; ignoring a confounding variable; describing an inference (this rock was deposited in water) as if it were direct observation; and treating a graph's axis scale as neutral when it exaggerates or hides a trend.