Do Democrats and Republicans listen to the same music? Watch the same movies? Read the same books? It would be reasonable to conclude, based on popular press reporting, that members from the two parties effectively in their own cultural bubbles with few cultural preferences crossing over between them. If true, this condition threatens to erode the social fabric that we rely upon to smooth over partisan conflict and disagreement, as apolitical lifestyle and cultural preferences act as a sort of social glue. However, the academic research behind these claims is mixed, with credible evidence on both sides of the debate. Jumping into this research gap, this dissertation attempts to improve upon the methods used in prior studies to clarify the extent to which partisans share cross-party cultural preferences. The main methodological improvement comes from a reworking of how audience networks are constructed from a combination of survey and digital trace data. Across three pilot studies, I demonstrate how this method works and how it suggests that partisans' cultural preferences are not substantively polarized. But, all of the datasets analyzed in these pilot studies are flawed in crucial ways, making it hard to draw generalizable conclusions from them. As such, I follow up with this dissertation's main study: a network analysis of an original dataset pairing behavioral data from Spotify with survey responses. Applied to this more robust dataset, my network-based approach finds that shared cultural preferences span across party lines, especially when people only share a few preferences. Larger sets of shared preferences are more likely to run between those with shared partisan identities, but these types of strong ties are very rare, even among co-partisans. However, these observational network-based results leave open an important question: what might be causing notable acute cases of partisan cultural polarization? I test one possible answer: elite political cues. I test this possibility via a simple survey experiment placing the musician Jon Bon Jovi in different contexts and asking subjects to evaluate him and his music. In doing so, I find that even an extreme political cue given through the survey is unable to generate strong reactions among subjects. Ultimately, this dissertation makes two important points. First, that partisans actually appear to share a substantial number of cultural preferences, undermining claims of partisan cultural bubbles. Second, that for partisanship to influence cultural preferences, people likely need to be exposed to persistent, strong, and diverse cues from both elites and peers. Together, these conclusions point to an important limit in the spillover of partisanship and its influences on the nonpolitical parts of our lives.