Many important topics in social science involve interactions between people. These topics include peer influence during childhood and adolescence, cooperation in the allocation of resources, formation of social movements and group and residential treatment of psychological and substance abuse disorders. The theory and analysis of interactions between individuals poses unique challenges both because the individuals are not independent of each other and because the interactions are dynamic and often nonlinear in nature. The Social Interactions Working Group is interested in developing and applying methodologies to address these challenges.
"Mitigating Unintended Consequences of Mandatory Domestic Violence Policies: Using Computer Simulations for Coordinating Interventions"
Dr. Peter Hovmand, George Warren Brown School of Social Work, Washington University in St. Louis
April 30, 2008
In the United States, the failure of communities and police departments to intervene resulted in a push to adopt and implement pro and mandatory arrest policies for domestic violence. These policies have led to an unexpected increase in the number of arrests of women, with women now representing approximately 20% of the intimate partner arrests. Various competing explanations have been offered in the literature for this unexpected increase in women arrests with conflicting implications for community responses to domestic violence. This talk first presents a system dynamics computer simulation model of the increase in women arrested for domestic violence. The model is based on data from a single case study of a coordinated community response to domestic violence. Data sources include multiple time series, arrest records, key informant interviews, and public documents. After replicating the observed system behavior, behavioral analysis approaches are applied to identify the mechanisms driving the increase in women arrests. Results show how mandatory arrest policies may have created or strengthened a “crossover” mechanism shifting the risk of arrest from men to women and the changing role of cooperation between victim advocates and police as central to explaining the increase in women arrests. The model is then used to explore how the sequence and timing of three interventions—mandatory arrest, victim advocate in the prosecutor’s office, and collaboration—affect increases in women arrests. Implications for domestic violence policy and research on nonlinear dynamic systems are discussed.
"Murder by Structure: Dominance Relations and the Social Order of Gang Homicide in Chicago"
Dr. Andrew Papachristos
February 1, 2008
Most sociological theories consider murder an outcome of the differential distribution of individual, neighborhood, or social characteristics. While such studies explain variation in aggregate homicide rates, they do not explain the social order of murder i.e., who kills whom, when, where, and for what reason. This paper argues that gang murder is best understood not by searching for its individual determinants but by examining the social networks of action and reaction that create it. Gang murder occurs through an epidemic-like process of social contagion as competing groups jockey for positions of dominance. In short, the social structure of gang murder is defined by the manner in which social networks are constructed and by people’s placement in them. I use a network approach and incident level homicide records to recreate and analyze the structure of gang murders in Chicago. Descriptive, non-parametric, and multivariate analyses demonstrate that individual murders between gangs create an institutionalized network of group conflict, net of any individual’s participation or motive. Within this network, murders spread as gangs respond to threats by evaluating the highly visible actions of others in their local networks. Gangs must constantly (re)establish the social order through highly visible displays of solidarity which, in turn, merely strengthen these murder networks.
"Social Interactions and Endogenous Association"
Dr. Bruce Weinberg
February 1, 2007
Stillman Hall 115
This paper develops a model of social interactions with endogenous association. People are assumed to invest in relationships to maximize their utility. Even in a linear-in-means model, when associations are endogenous, the effect of macro-group composition on behavior is non-linear and varies across individuals. We also show that larger groups facilitate sorting. Using data on associations among high school students, we provide a range of evidence consistent with our model. Individuals associate with people whose behaviors and characteristics are similar to their own. This tendency is stronger in large groups. We also show that behaviors vary within and between macro-groups in the way predicted by endogenous association.