Complex agent network model of youth violence in Singapore

• Main Author: Sun, Kaixuan
• Other Authors: Cheong Siew Ann
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Science::Physics
• Online Access: http://hdl.handle.net/10356/63141
• Institution: Nanyang Technological University

Nowadays, youth violence problem is becoming more serious. Although Singapore is a relatively safe place, we are nonetheless concerned that the youth violence problem might undermine the future of our nation. To understand how familial, social, and psychological factors play out in this problem, and to test the effectiveness of intervention measures, we built in this project a data-driven complex agent network model to simulate how the different factors affect each other and contribute ultimately to violence tendencies in the agents. Because of the large number of variables and model parameters, we start by doing a sensitivity analysis to examine to what extent each parameter will affect the rate of violent incidents. We found the model outcomes to be most sensitive to five parameters: (1) non-intact family to gang involvement, (2) school engagement to gang involvement, (3) general delinquency to gang involvement, (4) peer delinquency to gang involvement, and (5) friends in gang to gang involvement. We then choose to intervene by changing the school engagement level, varying the timing, intensity, and number of target agents of the intervention. Our results show that it is critical to intervene early, and successful interventions tip the scale between the competing psychological and social factors. This unique insight from the model can guide the development of more effective interventions by school and social counsellors.