The title of this blog post is KOD, which is the most recent album released by J. Cole. His album covers the dangerous reality of addictive behaviors as seen among his friends and family. I attached a 20 second snippet of his song FRIENDS in the album picture below. So what does this have to do with SNA? As you read through this post, you will understand how addiction can tie with family and friendship SNA.
The aim of this study was to apply social network analysis to pathological gambling in a comparison to recreational gamblers. For clarity, pathological gambling is an impulse control disorder defined by symptoms such as loss of control of gambling, preoccupation with gambling, and persistence despite negative consequences (1994). At the time of this study, a social network analysis was never done towards the subject matter of pathological gambling. Social network analysis methodology can provide insight as to whether individuals with pathological gambling seek out those with similar complications or whether social groups can directly deteriorate the risk for developing pathological gambling
Does an individual’s social network play a role in their pathological gambling behavior?
Is there a difference in network composition between pathological gamblers and non-pathological gamblers?
Pathological gamblers would have social networks that were denser with gamblers and be structurally different than non-pathological gamblers.
The study collected data using an egocentric network analysis approach from forty frequent-gambling adults recruited from the Athens, GA community. They were instructed to list their 30 closest social associates (ex. Friends, family members, romantic partners, co-workers) and demographics about each one of them (ex. Sex, race, years known, how close they were, etc.). What was truly measured was how frequently they gambled with or without the study participants. The assessment was done via EgoNet, a program used for collecting egocentric social network data.
Each research participant had their own egocentric network, with nodes representing their social associates and edges being their connections (ex. Who knows who). Figure 1 shows example social networks of gambling in two participants of the forty studied. (a) Non-pathological gamblers network’s gambling and (b) Pathological gamblers network’s gambling. The darker colors and larger nodes reflect more frequent gambling behavior. They found that there was a significantly greater occurrence of gamblers in the networks of pathological gambler participants compared to those of a non-pathological gambler’s network.
Diagnostic and statistical manual of mental disorders. (1994). Washington, DC: American Psychiatric Association.
Meisel, M. K., Clifton, A. D., MacKillop, J., Miller, J. D., Campbell, W. K., & Goodie, A. S. (2012). Egocentric social network analysis of pathological gambling. Addiction, 108(3), 584-591. doi:10.1111/add.12014
I didn’t realize this until nearing completion of the blog, but this study is by the same authors of the Gambling study but done at a later year. The aim of this study was to apply social network analysis to addictive behaviors through the various levels of alcohol usage. In other words, authors wanted to see if at-risk drinkers had networks of peers with addictive behaviors. Social network analysis methodology can help identify clusters in the network so that interventions can take place to address their behaviors.
Is addictive behavior spread evenly throughout networks or more localized in clusters?
Is there a difference in network composition between individuals with various alcohol usage levels?
At-risk drinkers would report greater frequencies of addictive behaviors among their peers and be structurally different between alcohol usage levels.
The study collected data from 281 undergraduate students enrolled in a psychology course at a large university in the Southeastern United States. The students were given an ego questionnaire with the Alcohol Use Disorders Identification Test (to assess their alcohol use and related problems) and a social network questionnaire (like the one given in the Gambling study above). The addictive behaviors included in this study were alcohol use, marijuana use, tobacco use, and gambling.
Pearson correlations were conducted to test whether each participant’s listed associates engage in addictive behavior. The study also utilized the Markov Clustering algorithm to maximize ties between participants and their associates within clusters while minimizing ties between clusters. Figure 2 displays a graphic representation of the different drinker types and their networks. Darker colors reflect higher prevalence of alcohol use by associates. And because the participant had ties between their listed associates, ties were only shown among associates who knew each other. The study found significant differences in drinking frequencies between the overall alcohol use networks among the participants. I would personally like to express my appreciation of this study visually categorizing each participant’s associates (as seen in the Key of Figure 2).
Meisel, M. K., Clifton, A. D., MacKillop, J., & Goodie, A. S. (2015). A social network analysis approach to alcohol use and co-occurring addictive behavior in young adults. Addictive Behaviors, 51, 72-79. doi:10.1016/j.addbeh.2015.07.009