Dissertation - 2023

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.

ProQuest -

UPenn Repository -

Partisans are divided in their non-political preferences, leading to prejudiced interpersonal interactions. In particular, significant attention has been given to how partisans are divided by their entertainment-media preferences, which may indicate deep divisions in taste associated with partisanship. However, this divide may only be minimal and associated with the abandonment of preferences in the face of the group-identity misrecognition and competing influences. To test both possibilities, I asked partisans to recommend films every American should see in their lifetimes and to justify their recommendations. These recommendations were then the basis for constructing an audience network, for which I assessed fragmentation. Similarly, I coded respondents’ justifications and then modeled the occurrence of themes on respondent features. Results provide little support for the existence of deep cultural divides. Democrats and Republicans recommended many of the same films and used the same justifications for doing so, suggesting that arguments over cultural divergence may be overstated.

Co-Authored with Do Eon Lee

There is little doubt over the existence of affective polarization, but findings on the related causal effects of party cues on non-political behavior may be affected by design decisions related to the volume and types of information used in experiments, as well as the trust framework employed. Here, we consider how these effects vary across low and high information environments for a less impactful trust context replicating initial trust conditions. We find that in low trust conditions, the effects of party cues are stable between party groups and are strong relative to the effects of other pieces of information about race, gender, religion, religiosity, policy preferences, and cultural preferences. However, such effects are reduced and become highly moderated by party affiliation within the high information environment. These findings suggest that differences between the implications of earlier research on the behavioral consequences of affective polarization and daily life may be explained by assumptions made inthe design of earlier studies.

MPSA 2019 -

The ANNALS of the American Academy of Political and Social Science - 2024

Co-Authored with Kokil Jaidka, Yphtach Lelkes, and Yifei Wang

News outlets are increasingly nationalizing their presentation of news stories, framing and presenting local news in a broad national context. We investigate how supply-side and curation-side factors of the news cycle contribute to the nationalization of news coverage. Through the computational analysis of 1.05 million Google news results on four days in July–August 2020, that corresponded to 1,581 news stories published on the George Floyd protests in Portland, Oregon, and Kenosha, Wisconsin, we examine the relationship between the nationalization of news coverage, stories’ search rank in Google News, and the geographic distance between the news event and the stories’ reading audience. Further, we explore the role of Google News in curating locally focused news. Our findings help to map the media ecosystem in a digital age, highlighting the influence of algorithmic power in politics and showing that excessive circulation of national news may have a profound negative impact on news diversity and social justice.

ANNALS -

Nature Human Behaviour - 2022

Co-Authored with Kokil Jaidka and Yphtach Lelkes

A submission to Nature Human Behaviour's Matters Arising section responding to Manguson's (2022) critique of Fischer, Jaidka, and Lelkes (2020). This work introduced new analysis of our original local news data to include a more precise assessment of the locality of results in Google News searches.

Nature Human Behaviour -

Nature Human Behaviour - 2020

Co-Authored with Kokil Jaidka and Yphtach Lelkes

Local news outlets have struggled to stay open in the more competitive market of digital media. Some have noted that this decline may be due to the ways in which digital platforms direct attention to some news outlets and not others. To test this theory, we collected 12.29 million responses to Google News searches within all US counties for a set of keywords. We compared the number of local outlets reported in the results against the number of national outlets. We find that, unless consumers are searching specifically for topics of local interest, national outlets dominate search results. Features correlated with local supply and demand, such as the number of local outlets and demographics associated with local news consumption, are not related to the likelihood of finding a local news outlet. Our findings imply that platforms may be diverting web traffic and desperately needed advertising dollars away from local news.

Nature Human Behaviour -

PACSS 2019 -

Working Paper - 2020

The response to the COVID-19 pandemic in the United States has been notably partisan. However, recent evidence suggests that people have also been directing more attention to local newspapers during this period. Given that local newspapers promote pro-social civic behavior, such as turning out to vote, it is possible that this increase in attention is helping communities to adopt necessary social distancing behavior. To test this possibility, I combine data from Google on mobility in thousands of American counties with counts of the number of newspapers available in each county, as well as county-level pandemic and demographic fea- tures, to model changes in staying at home and traveling for retail and recreation purposes. I find that even though behavior change is corre- lated with local newspaper availability, the association disappears when controlling for additional pandemic and demographic features. The lack of an effect persists even when applying covariate balancing propensity score weighting. The lack of a causal effect of local news availability on social distancing uptake suggests that local news is limited in its ability to undo the politicization of national issues.

New America - 2021

Co-Authored with Amber Hye-Yon Lee and Yphtach Lelkes

The quality of a democracy is, in part, determined by citizen attitudes. In particular, electoral winners and losers should believe that elections are fair, and interparty animosity should be minimal. While scholars have argued that disproportional electoral institutions increase the perceived system legitimacy gap between electoral winners and losers and increase affective polarization, they have relied on cross-sectional observational data. As correlates of electoral systems are also correlated with these attitudes, causal statements linking systems to attitudes are problematic. We also do not know whether people react to unfairness endemic to plurality systems or the downstream effects of these institutions, such as more vitriolic campaigns or elite polarization. Using a novel large-scale behavioral game that randomized participants to different electoral systems with other participants and that varied both in rules and number of parties to choose from, we examine whether electoral institutions directly affect subsequent attitudes. We find that non-plurality systems with many parties have the smallest winner-loser gap. While increasing the number of parties decrease interparty animosity, we found, surprisingly, that plurality systems on their own had the lowest levels of interparty animosty. However, within proportional systems, increasing the number of parties decreases interparty animosity.

New America -

New America (Brief) -

SSRN -

Book Chapter - 2023

In Emotions in the Digital World: Exploring Affective Experience and Expression in Online Interactions, edited by Robin L. Nabi and Jessica Gall Myrick

Co-Authored with Yphtach Lelkes

This book chapter reviews the links between emotion, affective polarization, and online communication already observerd in the literature. We highlight how certain negative emotions have been shown to promote the types of engagement with digital media that are associated with affective polarization. We also highlight opportunities for future research, in part by working through a case study showing that discussions in political forums on Reddit are more emotionally charged when discussing members of the outparty.

Journal of Medical Internet Research - 2023

Co-Authored with Chris Williams, Elissa Kranzler, Joe Luchman, Ben Denison, Tom Wonder, Ronne Otsby, Monica Vines, Jessica Weinberg, Elizabeth Petrun Sayers, Allison Kurti, Sarah Trigger, Leah Hoffman, and Josh Peck

Results from this study provide initial evidence of the We Can Do This campaign’s digital impact on vaccine uptake. The size and length of the Department of Health and Human Services We Can Do This public education campaign make it uniquely situated to examine the impact of a digital campaign on COVID-19 vaccination, which may help inform future vaccine communication efforts and broader public education efforts. These findings suggest that campaign digital dose is positively associated with COVID-19 vaccination uptake among US adults; future research assessing campaign impact on reduced COVID-19–attributed morbidity and mortality and other benefits is recommended. This study indicates that digital channels have played an important role in the COVID-19 pandemic response. Digital outreach may be integral in addressing future pandemics and could even play a role in addressing nonpandemic public health crises.

JMIR -

Joint Statistical Meetings - 2023

Much of the published work (peer-reviewed or otherwise) in sports analytics is focused on measurement and prediction tasks, with little spent focusing on full causal evaluations of decisions or policies. This presentation details why a focus on causal inference, particularly design-based causal inference developed in econometrics, can be useful in sports domains with examples from cross country and track and field.

JSM -

SportRxiv Preprint - 2021

Recent shifts in professional basketball have led teams to place more urgency in drafting as well as possible. Draft picks must play out their initial years under team-friendly contracts that provide teams with increased salary cap flexibility. Yet, while this urgency has led to widespread discussion and research of how to improve teams' draft decisions, little attention has been given to identifying what teams can do to maximize the performance and potential of their draft picks once they are added to their roster. However, theories of learning and ecological psychology suggest that giving young players as much playing time as possible should lead to concrete improvements in their development and future performance. In this study, I test this causal theory by evaluating the relationship between the minutes a player receives in their first two seasons in the NBA and their fourth-year performance using a novel method of propensity score weighting that enables weighting for continuous treatment variables. I find that players who receive more minutes in their first two seasons have better fourth seasons and make larger jumps from their first two seasons to their fourth season, controlling for a broad set of potential confounders. These results have important implications for teams as they develop organizational strategies for the short-and medium-term.

Preprint -

NESSIS 2021 -