Paul Vicinanza
Paul Vicinanza
Research Statement
My research studies how social structures shape the emergence, spread, and evolution of knowledge and culture with novel methodologies rooted in natural language processing and deep learning. One paper (forthcoming, American Sociological Review) traces the origins of conspiratorial beliefs on twitter to mortality threat and their spread to the further social reinforcement of extreme belief. Another research project demonstrates prescient ideas—those which are incommensurate with existing logics but later reach widespread acceptance—emerge from peripheral actors in fields as diverse as politics, law, and business. In my job market paper, I analyze over 50,000 quarterly earnings calls between sell-side analysts and the executives of publicly traded firms. I theoretically frame the interaction as a type of "interrogation game" and demonstrate how asymmetric acts of conversational deference and alignment by executives predict future analyst stock upgrades and stock returns. A separate series of papers examines the structural origins of financial misconduct among U.S. financial advisors and entrepreneurial ventures. You can find me at divey music joints, hiking through redwoods, or on the 2022-2023 academic job market. For a more detailed summary please refer the to research page.
Research Interests
- computational social science
- culture
- misconduct
- natural language processing
Job Market Paper
Many forms of social interaction in market economies, from job interviews to press conferences, are a type of interrogation. In these “interrogation games,” the interrogator seeks accurate and replete information from a subject who aims to both appear forthcoming while also concealing sensitive material. In this manuscript, I define the interrogation game and develop a structural account of how the game should unfold, rooted in Goffman’s theory of the interaction order, to produce a desired outcome for the subject. I argue two constructs deemed universally essential to successful conversation–interpersonal alignment and conversational flow–instead produce dissonant interaction when undertaken by the wrong party. I apply my theory to study the conversations of sell-side analysts and executives in over 50,000 quarterly earnings calls (QECs) and predict analysts’ future stock recommendations for the firm. I measure alignment and flow by projecting each utterance in the conversation to a high dimensional space using word embeddings and validate the measures against hand-annotated data. Consistent with my theory, I find analysts upgrade the stock when executives exhibit deference by aligning with analyst questions and letting analysts control the flow of conversation. In contrast to theories of topical control, I find that analysts are more likely to downgrade the firm when analysts align with executives and when executives drive the conversation. I discuss the implications of these results for the literatures on QECs and firm valuation, computational conversation analysis, and the micro-sociology of strategic interaction.
Publications
Henrich Greve, Hayagreeva Rao, Paul Vicinanza, and Echo Zhou
Working Papers
Paul Vicinanza, Amir Goldberg, and Sameer Srivastava
Paul Vicinanza, Hayagreeva Rao, Mark Egan, Gregor Matvos, and Amit Seru
Paul Vicinanza, Hayagreeva Rao, Mark Egan, Gregor Matvos, and Amit Seru
Work in Progress
with Henrich Greve, Hayagreeva Rao, and Echo Zhou