Data Science/Computational Social Science Seminar: Amir Goldberg

A computational linguistic model of organizational identification

Abstract:
Identification, the quality of associating the self with a group, is a fundamental human tendency. In contemporary market societies, organizations often serve as important sources of identification. Why is it that some people identify with the organization that employs them more strongly than others? Whereas existing work often attributes this variation to differences between organizational attributes or individual personalities, we trace it to differences in the network positions people occupy. To do so, we use person-specific word embeddings to develop a method for detecting organizational identification in interpersonal email communication. We validate our measure using established survey-based methods and apply it to members of three different organizations. Drawing on structuralist sociological theories, we demonstrate that individuals embedded in dense networks tend to exhibit higher organizational identification. Our findings and methods apply to other settings where the strength of group identification strongly predicts individual behavior. 

Speaker bio:
 

Amir is an associate professor of organizational behavior and (by courtesy) sociology at the Stanford Graduate School of Business, where he is the founder and co-director of the Computational Culture Lab. His work and interests lie at the intersection of computation and social science and are not easily described using standard labels. He uses computationally intensive language- and network-based methods applied to big data to model cultural dynamics in organizations, markets and informal social groups. His research projects all share an overarching theme: the desire to understand the social mechanisms that underlie how people construct meaning and consequently pursue action. His work has been published in The American Journal of Sociology, The American Sociological Review, Administrative Science Quarterly, Management Science and The Review of Financial Studies. He holds a PhD in sociology from Princeton University, an MA in sociology from Goldsmiths College and a BA in computer science and film from Tel Aviv University. 

Register to attend DS/CSS events at umsi.info/DSCSS