Prevention of violence against children and psychosocial models of intervention: Network analysis of an inter-institutional service

This study describes an Italian service, the Inter-institutional Groups Operating against child Abuse and Maltreatment (GOIAM), which specializes in the treatment of child abuse. The GOIAM service is composed of 53 professionals divided into four different categories: social workers, psychologists, child neuropsychiatrists, and psychopedagogists. With the aim of detecting strengths and weaknesses of the operative model adopted by GOIAM, the whole service was analyzed through the application of social network analysis (SNA). The analysis revealed a low-density network, while the study of centrality yielded different distributions of power indicators (degree, closeness, and betweenness): prestige and pervasiveness were balanced, while influence, reachability, and betweenness were quite unbalanced. The search for subgroups (N-clans, lambda sets, community structure) suggested the presence of many overlapping triads, leading to the formation of large clusters. The inspection of such groups revealed that professionals belonging to the same district tended to cluster. This condition did not promote the diffusion of opinions and expertise, and could be an important factor of weakness in an interdistrict service. The analysis of inter-groups relationships (brokerage) and the study of social roles (REGular Equivalence) showed that most professionals had “diffuse” roles through which they try to perform every task, exchanging assignments, activities, and positions. The flattening of the roles inside the organization might at first appear as facilitation, although this could represent a limit for a complex organization in the long term because it generates a lack of specialization. Finally, the potential evolution of the GOIAM network (P1 model) seems to be characterized by a further decrease in density, more isolation of peripheral nodes, an increase in centralization, and, consequently, a higher level of hierarchy in the network.