A growing strand in the rapidly emerging literature on populism investigates the distribution of so-called populist attitudes among ordinary citizens, the measurement of these attitudes, how they arise, and what their consequences are for different behavioral outcomes, such as party preferences and voting. While the exact definition of populism and its components are still contested, the majority of these recent studies converge on a consensus about the core of populist attitudes: a negative view of established political elites, a Manichean outlook of politics, and favorable view of the people’s will as the foundation for democratic decisions. In this paper, we argue that the adoption of this view can present two main conceptual challenges to so-called populist attitudes that have yet been addressed in the literature.
First, we identify a dissonance between the definition of populist attitudes, which usually follows the Satorian way of taxonomical hierarchies, and how populist attitudes are operationalized, namely in line with the Goertzian family resemblance approach. That is, while the concept of populism is dichotomous and requires the simultaneous presence of all of its facets, the measurement is often continuous and allows substituting the lack of one component by higher values on other aspects. The second challenge is a direct result of the first. Ultimately, populist attitudes borrow extensively from related and long-standing concepts in public opinion, such as political cynicism. These related concepts are bundled under the umbrella of populism. If the concept is one of family resemblance, however, in which, for instance, one dimension such as anti-elitism may be enough to characterize individuals as populist, then we can no longer distinguish populism from its sub-dimensions, further diminishing its theoretical and field utility. Against this backdrop, this paper discusses the conceptual underpinnings of populism as s set of individual-level attitudes. We also demonstrate empirically how different conceptual understandings can lead to diverging empirical findings. Applying structural equation and latent class models, we analyze panel data from the German Longitudinal Election Study (GLES), which contains a multi-item measure of populist attitudes in three survey waves. We investigate the additional value gained from including so-called populist attitudes following the classical Sartorian way by implementing them into basic vote choice models.
