This project studies the role of appearance in politics as a politically contested and strategically adjustable attribute. The project asserts that politicians' self-expression via stylistic choices shapes voters' perception of their social, cultural, and political identity. The project argues that politicians act as rational actors when making appearance choices, trying to maximize positive and minimize negative reactions. Their decisions are guided by appearance norms, which are defined as group-level expectations about the approval and disapproval of appearance choices, and allow them to anticipate audience responses. However, politicians value positive reactions from varying groups, all of which can have their own, potentially diverging appearance norms. This creates tensions that can be difficult to navigate. Since appearance norms in politics are biased towards those of overrepresented groups, these tensions are particularly pronounced for members of historically underrepresented groups.
The project sets out to study these dynamics through a systematic account of politicians' appearances and the politics behind their stylistic decisions. To do so, I compile novel and comprehensive cross-national image data capturing the appearances of politicians over time. Next, I leverage advancements in machine learning to measure appearance norms and norm deviation based on these images. I use the resulting data to (1) study the evolution of appearance norms, (2) understand how politicians of over- and underrepresented groups navigate the tensions between the appearance norms of varying groups, and (3) investigate political correlates of varying appearance strategies.
