Political science is an intensely quantitative discipline, and models are central. Political scientists use models--formal and informal, statistical and qualitative--to investigate and illuminate causal mechanisms, generate comparative data, and understand the conditions under which certain outcomes are expected to occur. But how do we justify and rationalize the method? Why test predictions from a deductive, and thus truth-preserving, system? David Primo and Kevin Clarke tackle these central questions in this novel work of methodology.