The skein of political life in democracies is complex and tightly woven. In this chapter we attempt to unravel several of its most important strands. We begin by developing and testing a multivariate model of the support process in which the dependent variables are the authorities, regime and community support factors identified in Chapter 4. The significance of economic and more general government effectiveness evaluations in the support model prompts us to assess their generality by investigating the impact of economic and other variables on support for democratic regimes in a broader comparative context. One of the problems that has bedeviled previous studies of political support is a failure to demonstrate that it has consequences for important political attitudes and behavior in a democracy. Because one of the proudest claims of democracies is that they are governments of “laws not of men,” we next consider how people's willingness to voluntarily comply with authoritative edicts of government are affected by differences in support for the political authorities, regime, and community, as well as by other factors relevant to the support process. Finally, because citizen involvement is a sine qua non of democratic politics, we examine participation in conventional electorally oriented activities as well as more unconventional ones such as sit-ins and potentially violent protests, and determine the extent to which these are affected by variations in political support and related factors. These analyses demonstrate that support does indeed have significant consequences for two of the most important aspects of political life of a democracy.

THE FOUNDATIONS OF POLITICAL SUPPORT! A MULTIVARIATE ANALYSIS

Analyses in presented Chapter 4 reveal that Canadians distinguish between support for the national political community and regime and support for incumbent authorities. Moreover, at the regime level, they distinguish support for parliament and the civil service, on the one hand, and support for the judiciary, on the other. We employed several predictor variables in our multivariate model of these four support factors.