THE TIES THAT BIND: A SOCIAL NETWORK ANALYSIS OF A LARGE GANG SEX TRAFFICKING NETWORK IN THE UNITED STATES

This dissertation examines the effect of network position on the probability that a sex trafficking network member avoided indictment in this gang RICO prosecution, and what that says about the functioning of trafficking networks and their resistance to fragmentation strategies by law enforcement. Ties between members, particularly involving members that broker relationships between otherwise disconnected individuals or groups, are hypothesized to be conduits through which benefits flow, such as power, information, or influence. Others hypothesize that brokerage positions can hold disadvantages, so that an advantageous position in one network may be disadvantageous in another. An individual in a brokerage position might also use his/her power strategically to block the flow of benefits between others.This study used police and court data to examine how brokerage, measured in different ways, influenced the probability of avoiding indictment. This is the first social network analysis of an individual human trafficking network in the United States, and it is the largest such study of a human trafficking network in terms of network size examined. It is distinctive in that it includes extensive data on both indicted and unindicted individuals in sufficient numbers to use broad boundaries and include the impact of porous network borders. Avoiding indictment is the dependent variable and is used as a proxy for survival, which is an important utility to be maximized when examining criminal trafficking networks as business enterprises. Results support the notion that relying on betweenness centrality alone to measure brokerage does not work as predictably well in larger networks; its effect was miniscule and statistically insignificant in every model where degree centrality, or the sheer number of people one is connected to, was included. This is contrary to the results of several network analyses involving smaller networks for a few important reasons, including the presence of multiple redundant ties and the lengths of paths between individuals in larger, more mature networks. Clustering coefficient emerged as a measure of brokerage worthy of more study.Important insights are also offered regarding the symbiotic relationships between perpetrators and victims. On a practical level, the goal of this study was to explore the extent to which identifying brokers in a sex trafficking network could help police target network members that would best fragment a network, thus reducing its ability to operate and exploit victims.