Social network models to understand the relationships between protective and risk factors in problematic gambling

To understand addictive behavior and develop preventive interventions, it is important to know the characteristics of people who exhibit such conduct. In recent years, a progressive growth trend of gambling behavior and problematic gambling has been observed and regarding problematic gambling behavior the prevalence rates range from 0.1% to 6% among the adult population (Abbott, 2020). This study focuses on gambling and studying the characteristics of gamblers who show different levels of gambling severity by the utilization of a new methodology of analysis that makes it possible to observe how different psychological characteristics such as: attribution of responsibility toward gambling (personal or casino responsibility), beliefs related to gambling, and knowledge about gambling are shaped in profiles of gamblers showing different levels of gambling severity.

We used data from The Transparency Project website (Gray, LaPlante, Abarbanel & Bernhard, 2021)  that included a sample of 3748 American adults who attended a Casino and responded to an online questionnaire.

To analyze data we employ a branch of social network analysis called “community analysis” and to design the network, each person and each answer is a node, giving a bipartite network. Two nodes that belong to the set of the answers to the questions are linked if there exist at least one respondent that has selected both answers. Two respondents are linked if they have at least one answer in common. The links are weighted: the more answers have in common the stronger the link between two nodes. The “Louvain method” (Blondel, Guillaume, Lambiotte & Lefebvre, 2008) allows to identify 3 communities of person nodes. An ANOVA test is conducted to check for significant differences in the average values of the variables that characterize the nodes in each community.

Community 1 represents almost 60% of the sample, has 97% of individuals with no problematic gambling behavior with high gambling literacy, high personal responsibility, but low casino responsibility.

Community 2 represents almost 30% of the sample, has 56% of individuals with no problematic gambling behavior and 44% with problematic gambling behavior. The main characteristics of this community are: low personal responsibility, a factor that is shared by both gamblers at risk and problematic gamblers.

Community 3 represents roughly 15% of the total, has 91% of individuals with no problematic gambling behavior, low gambling literacy, this community shows both high casino responsibility and personal responsibility and it is the less stable community of the three. This study found that factors such as attribution of personal responsibility toward gambling behavior could be considered as risks toward engaging in such behavior. Future prevention programs should consider elements such as perceived responsibility and beliefs toward gambling.