On the salience of political issues: The problem with 'most important problem'

Salience is an important concept throughout political science. Traditionally, the word has been used to designate the importance of issues, particularly for voters. To measure salience in political behavior research, scholars often rely on people's responses to the survey question that asks about the “most important problem” (MIP) facing the nation. In this manuscript, I argue that the measure confuses at least two different characteristics of salience: The importance of issues and the degree to which issues are a problem. It even may be that issues and problems are themselves fundamentally different things, one relating to public policy and the other to conditions. Herein, I conceptualize the different characteristics of MIP. I then undertake an empirical analysis of MIP responses over time in the US. It shows that most of the variation in MIP responses reflects variation in problem status. To assess whether these responses still capture meaningful information about the variation in importance over time, I turn to an analysis of opinion-policy dynamics, specifically on the defense spending domain. It shows that changes in MIP mentions do not structure public responsiveness to policy or policymaker responsiveness to public preferences themselves. Based on the results, it appears that the political importance of defense has remained largely unchanged over time. The same may be true in other domains. Regardless, relying on MIP responses as indicators of issue importance of any domain is fundamentally flawed. The responses may tell us something about the “prominence” of issues, but we simply do not know. The implication is clear: We first need to decide what we want to measure and then design the instruments to measure it.