Societal threat, authoritarianism, conservatism, and U.S. state death penalty sentencing (1977-2004).

On the basis of K. Stenner's (2005) authoritarian dynamic theory, it was hypothesized that the number of death sentences and executions would be higher in more threatened conservative states than in less threatened conservative states, and would be lower in more threatened liberal states than in less threatened liberal states. Threat was based on state homicide rate, violent crime rate, and non-White percentage of population. Conservatism was based on state voter ideological identification, Democratic and Republican Party elite liberalism-conservatism, policy liberalism-conservatism, religious fundamentalism, degree of economic freedom, and 2004 presidential election results. For 1977-2004, with controls for state population and years with a death penalty provision, the interactive hypothesis received consistent support using the state conservatism composite and voter ideological identification alone. As well, state conservatism was related to death penalties and executions, but state threat was not. The temporal stability of the findings was demonstrated with a split-half internal replication using the periods 1977-1990 and 1991-2004. The interactive hypothesis and the results also are discussed in the context of other threat-authoritarianism theories and terror management theory.