Democratic and aggressive political participation: Estimation of a nonrecursive model

Previous empirical research concerning political participation has made either one of two errors. In the first case, the research has omitted aggressive or extralegal forms of political participation. In the second case, the research included aggressive forms of participation but estimated models of aggressive and democratic participation separately despite the strong positive correlation between them. Each error leads to an incorrectly specified model of political participation. To correct this situation, we specify a nonrecursive model and test it using data from two disparate sample surveys. In each test the results indicate that earlier participation models were incorrectly specified in that aggressive behavior is a significant cause of democratic participation while the reverse is not true.