Published online by Cambridge University Press: 25 January 2021
The American states provide, arguably, the world's most advantageous venue in which to test general propositions about political behavior and policymaking (Jewell 1982; Brace and Jewett 1995). One would be hardpressed to design a better system of polities for conducting political research. The states represent fifty units of analysis with broadly similar political structures, cultures, and populations, but with significant variation across a range of social, policy, and institutional characteristics important to many political theories. Unlike the examination of single units of government, such as the U.S. federal government, the variation in these characteristics allows scholars to test hypotheses of cause-and-effect relationships of political behavior and policymaking in a valid way. But unlike the comparison of different countries, the comparison of states does not produce so much variation that it overwhelms scholars' ability to identify relationships that might exist. Just as the states are a natural laboratory for public policy (New State Ice Co. v. Liebmann 1932), they are also a natural laboratory for testing hypotheses derived from political theory. Furthermore, the accessibility of state government officials makes quite feasible many research strategies that would be unthinkable in the study of federal officials and institutions.