Inserção e interdependência multinível entre redes sociais interpessoais e interorganizacionais: capital social e desempenho no caso do cluster industrial Parque Tecnológico BH-TEC

• Ano de defesa: 2018
• Autor(a) principal: Luciano Mattar
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; FAFICH - FACULDADE DE FILOSOFIA E CIENCIAS HUMANAS; Programa de Pós-Graduação em Sociologia; UFMG
• Palavras-chave em Português: redes sociais; análise de redes sociais; embeddedness; capital social; cooperação; relações estado-sociedade; clusteres produtivos; modelos de grafos aleatórios exponenciais; Sociologia; Capital social; Redes de relações sociais
• Link de acesso: http://hdl.handle.net/1843/31685

Studies on the insertion of economic relations in social relations in terms of social networks, guided by the concept of embeddedness, cover researches on the relationship between structured patterns of interpersonal and interorganizational behavior and the performance of industrial environments. These patterns, according to the concept of social capital, allow us to understand sociability as a support to the ability to seek benefits through association in networks. The purpose of this research is to examine, in a case study, how local interaction processes define network patterns at the levels of individuals and companies, both within and between levels, and consequently how these processes affect the creation of social capital via the emergence of network structures. The main research hypothesis is that the interdependence between social networks at two levels, which results from their embeddedness, impacts the structuration of similar patterns of behavior at each of the two levels. In this case study of the BH-TEC Technological Park, interpersonal relations of advice among directors and interorganizational relations of cooperation between firms were investigated. Probabilistic multilevel exponential random graphs models are employed to estimate the effects of local processes of network formation. Such probabilistic models for social network analysis are appropriate for the test of hypotheses about the structuration of networks that result from nonrandom social process. The main finding is the identification of significant co-occurrence of tie creation processes at the individual and the corporate levels; it is nevertheless worthy to notice that specific processes take place at each level. Specifically, dense cyclical relationships tend to guide the formation of networks among directors, while dense hierarchical relationships tend to guide the formation of networks among firms. In terms of social capital, dense networks among firms reduce uncertainty and foster interorganizational cooperation. For the directors, according to the analysis, network density is potentially more beneficial for it tends to be associated with structural autonomy, that is, the agents create and maintain ties with nodes external to their groups, which is crucial to obtain new information and opportunities for themselves and their groups, and thus increase the performance of the group. The overall conclusion is the presence of distinct interaction patterns for each of the two levels. At the interorganizational, the interactions between the firms are defined by the decisions of their directors, however, the relationships are not identical to their relationships.