Research Interests

Automation, Artificial Intelligence, and Politics

My postdoctoral research can be organized into two overarching themes: the first examines people’s political and policy preferences in response to labor dislocation as a result of automation and artificial intelligence (AAI) and globalization; the second analyzes people’s attitudes towards artificial intelligence.

  • I lead a project that investigates whether providing more encompassing information about the trade-offs associated with AAI (highlighting both its costs and benefits) affects not only its support but also the policy preferences in response to it.
  • In other projects with my co-authors we investigate whether citizens are similarly concerned about economic displacement caused by automation as compared to other labor market changes, such as those arising from trade and demand changes, whether these effects vary by gender, and finally whether these fears might generate support for populist and radical right parties.
  • Another project addresses three main questions: 1) when should governments use AI according to citizens, 2) what do citizens know about AI and algorithms, and 3) what are their biggest concerns around government use of AI?

Explaining the causes and consequences of populism

  • Motivated by the global rise of populism and the increasing backlash against globalization, my dissertation explores the role that financial and economic literacy plays in shaping individual economic policy preferences in Italy and the UK, and teases apart the mechanisms behind this relationship.
  • I address the question of how different socio-economic crises affect support for democratic institutions and voting behavior. I studied the interaction between the Eurozone crisis and external interventions such as IMF and EU conditionality on the decline in citizens’ support for democratic institutions, and further expanded the analysis along the gender dimension. Similarly, I analyzed the political success of anti-immigration parties in Italy, focusing on the relative importance of immigration preferences vs issue salience.

Sports and Politics

  • In a paper recently accepted at Sociology, with my co-author Morgan Wack we developed and deployed a unique dataset of over 6,500 player-year observations from the Italian Serie A created using an extensive set of skin tone data developed to improve the verisimilitude of an online interactive game. This allowed us to uncover substantial evidence of racial biases from both fans and officials in the Italian football league.
  • Two other projects with Kevin Aslett and Morgan Wack focus on the process of “sports-washing”, in the context of two recent examples: the role of Russian oligarch Roman Abramovich and his ownership of Chelsea Football Club during Russia’s invasion of Ukraine, and the 2022 FIFA Men World Cup in Qatar. To answer these questions we analyze tweets employing a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) machine learning model.

COVID-19

  • From April 2020 to June 2021 I worked as a research assistant on the UW COVID-19 State Policy Project, the nation-leading effort in collecting daily data on social distancing policies in response to COVID-19 in the US. Our data has been used, among others, by Tableau, Facebook, The COVID Mobility Network, IHME, the Imperial College of London, and COVID Exit Strategy. Our collaborative research effort has resulted in a number of publications in The Lancet, Nature Medicine, Perspectives on Politics and State Politics & Policy Quarterly.
  • In the latter two manuscripts, we analyze what factors determine different social distancing policies at the state level. We find that the most important predictor of introducing and easing social distancing policies is the governor’s party affiliation, with Republican governors introducing measures later and easing them quicker. More strikingly, governors of states with larger Black populations were quicker to ease, and this effect is especially strong in states with Republican governors, suggesting neglect of the disproportionate impact of the epidemic on this population.
%d bloggers like this: