A Social Network Analysis of Alcohol-Impaired Drivers in Maryland---An Egocentric Approach: Applications of Logistic Regression and Poisson Regression

• Activity Number: 696
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Ashraf Ahmed*+ and Andrew Farkas and Kenneth Beck
• Companies: Morgan State University and MORGAN STATE UNIVERSITY and University of Maryland
• Keywords: Social network ; alcohol-impaired drivers ; drinking behavior ; ego and alter

Abstract

Social networks influence human behaviors, including alcohol consumption and travel. This study examined the personal, household, and social structural attributes of alcohol impaired drivers in Maryland, in particular, whether drinking behavior is influenced by social network. It used an egocentric approach of social network analysis, focusing on alcohol-impaired drivers (egos), and those with whom the egos interact and participate in social drinking (alters). A telephone survey was used to collect data from egos, and alters. The data was analyzed in both bivariate and multivariate settings using logistic and Poisson regressions. The results indicate that egos with larger social networks are more likely to drink in a social context, and those with smaller networks are more likely to drink alone because of depression. Egos with many people in their social network are less likely to drink in a context of emotional pain and more likely to drink in context of social facilitation. Egos and alters both exhibit risky driving behaviors, but alters somewhat less with moving violations. Both egos and alters influence each other to drink less, but alters have shown greater influence on egos.