Ideology Among the Mass Public

• In Event: Reconsidering Ideology in Mass Publics

Abstract

What role do preferences of the mass public play in the American political system? The predominant view among scholars of public opinion has long been that mass-level preferences cannot play much of a role because, for the most part, such preferences do not exist. In this view, ideologies are creatures of political elites — understood poorly if at all by most members of the public (Converse 1964, Zaller 1992, Noel 2013, Achen and Bartels 2016, and Kinder and Kalmoe 2017) Since so few members of the public have coherent political views, they cannot be the source of major American political ideologies.

We argue that the conclusions of prior research arise from a challenge of measurement. Converse and subsequent researchers have defined ideology as a coherent, stable set of political beliefs, and they have sought to measure such a belief-set through policy questions. We argue that doing so collapses information about elite-level politics with political preferences. Indeed, much evidence shows that members of the American public have low levels of information about policy. Hence, to measure ideology through policy questions inevitably indicates an apparent absence of ideology. But this approach reveals that many survey respondents lack stable positions on policy issues, not that respondents lack stable political beliefs in general. If respondents were to exhibit stability in (non-policy) political beliefs, this opens the door to future research on the content and development of mass-level ideology, which previous research would appear to foreclose.

We measure ideology through beliefs about the power of different social groups. Specifically, we provide a list of groups and ask survey respondents about the level of power in American politics held by each group (descriptively) and how much power each group ought to have (normatively). Analogous to feeling-thermometers, we label these “power thermometers”. In a two-wave panel survey, we evaluate stability in respondents' answers to these questions. We show response stability over multiple survey waves to be high (in the range of response stability on group feeling thermometers). In this way, we replicate Converse's approach to observing ideology among members of the mass public, but rather than seeking to measure ideology through policy questions, we do so through questions about group power.

Authors

• Joseph Warren, University of California, Berkeley

• Alan Nigel Yan, University of California, Berkeley