Toward an Empirical Framework for Studying Gangs as Groups: A Case Study of Gangs in the Texas Prison System

• In Event: Pulling Back the Curtain on Gangs in American and Canadian Prisons

Abstract

The most recent estimates from law enforcement indicate that there are roughly 850,000 gang members and 30,000 gangs in the United States. Social scientists concerned with the etiology of, and responses to, gangs generally study the former and ignore the latter. That is, even though inferences typically operate at the group-level, the gangs themselves are usually nowhere to be found in gang research, owing to trends in the quantitative social sciences that disaggregated down to study individuals or aggregated up to study neighborhoods. Our goal is to put “gangs” back into gang research. Based on data from the LoneStar Project, a study of 802 prisoners interviewed in Texas, half of whom were officially designated as gang members, we offer an new empirical framework for the study of gangs as groups. We address three questions: (1) Is it possible to research the group-level characteristics of gangs by relying on the self-reports of gang members? (2) To what extent is there heterogeneity across gangs in their instrumental, expressive, profit-generation, and communicative features? (3) What explains differences in the group-level characteristics of gangs? We discuss the implications of this research for theory, methods, and responses to gangs.

Authors

• David Pyrooz, University of Colorado Boulder

• Scott Decker, Arizona State University