Underlying Motivations of Substance Abuse: A Study of Moral Values and Co-Occurring Addictive Behavior in Online Social Networks
Conference Year
January 2019
Abstract
Research Objective
The new opioids crisis reopened a large-scale debate about addiction in the U.S. The Department of Health and Human Services recently declared the opioid epidemic a public health emergency. The crisis may be seen as a syndemic rather than an epidemic because it is an aggregation of multiple and sequential epidemics. Deaths from opioids are interdependent with deaths associated with other epidemics, such as depression, chronic disease and other substance abuse. The distinction made is imperative, because syndemics develop under health disparities caused by poverty, stress, unemployment and (perceived) structural injustice. Indeed, to solve the crisis we must depart from a traditional biomedical approach and focus on the social context that influence decision-making and substance dependence.
Models of (rational) decision-making are routinely based on the notion that agents base their choices on underlying motivations – preferences, desires, needs, and morals. Researchers interested in addictive behaviors have extensively studied the social context but have not looked at morality as a clustering mechanism among types of addiction. The objective of this study is to use cluster analysis to show differences in foundations of morality and types of addiction.
Study Design
We used text mining techniques to investigate the relationship between moral values and moral behavior, we classified individuals by moral values using a data dictionary based on the Moral Foundations Theory developed by scholars in Moral Psychology. Users of Reddit, an online social network, were clustered by addiction tags and morality tags. The posts were aggregated by user and only users who were tagged both with an addiction type and a morality foundation class were kept. We formed a bipartite graph between users and addiction tags and converted this to a network of Reddit users, where edges correspond to a shared addiction tag. We used the Girvan-Newman clustering algorithm which generates disjoint clusters to hierarchically cluster users via edge-betweenness.
Population Studied
We used text data from Reddit and its specific forum or “Subreddit” for Addiction. Through an Application Programming Interface, we gathered the top thousand and most up-voted comments. These comments were collected from 204 unique individuals.
Principal Findings
We found that the clusters did not match. We calculated the Adjusted Rand index and found it to be just slightly above zero (0.01777) meaning that the clusterings of individuals by addiction and morality were approximately as similar as if they had been formed by random chance.
Conclusions
We conclude that more work is needed to empirically test the relationship between moral values and moral behavior and co-occurring addictive behavior in social networks.
Implications for Policy or Practice
This study serves as a catalyst to more research into morality in social networks and substance abuse. If we have more information about what drives individuals in their addictive behaviors, the results can be used to design more effective prevention programs and to monitor and evaluate these programs more effectively. It could help the National Academy of Medicine’s coalition of government and private sector parties, including the CDC, CMS, FDA and addiction groups, to better understand the complex drivers of the opioid syndemic and addiction patterns.
Primary Faculty Mentor Name
Eline van den Broek-Altenburg
Faculty/Staff Collaborators
Elizabeth Tripp (Collaborator), Niokhor Dione (Collaborator)
Status
Undergraduate
Student College
College of Engineering and Mathematical Sciences
Program/Major
Data Science
Primary Research Category
Health Sciences
Secondary Research Category
Social Sciences
Underlying Motivations of Substance Abuse: A Study of Moral Values and Co-Occurring Addictive Behavior in Online Social Networks
Research Objective
The new opioids crisis reopened a large-scale debate about addiction in the U.S. The Department of Health and Human Services recently declared the opioid epidemic a public health emergency. The crisis may be seen as a syndemic rather than an epidemic because it is an aggregation of multiple and sequential epidemics. Deaths from opioids are interdependent with deaths associated with other epidemics, such as depression, chronic disease and other substance abuse. The distinction made is imperative, because syndemics develop under health disparities caused by poverty, stress, unemployment and (perceived) structural injustice. Indeed, to solve the crisis we must depart from a traditional biomedical approach and focus on the social context that influence decision-making and substance dependence.
Models of (rational) decision-making are routinely based on the notion that agents base their choices on underlying motivations – preferences, desires, needs, and morals. Researchers interested in addictive behaviors have extensively studied the social context but have not looked at morality as a clustering mechanism among types of addiction. The objective of this study is to use cluster analysis to show differences in foundations of morality and types of addiction.
Study Design
We used text mining techniques to investigate the relationship between moral values and moral behavior, we classified individuals by moral values using a data dictionary based on the Moral Foundations Theory developed by scholars in Moral Psychology. Users of Reddit, an online social network, were clustered by addiction tags and morality tags. The posts were aggregated by user and only users who were tagged both with an addiction type and a morality foundation class were kept. We formed a bipartite graph between users and addiction tags and converted this to a network of Reddit users, where edges correspond to a shared addiction tag. We used the Girvan-Newman clustering algorithm which generates disjoint clusters to hierarchically cluster users via edge-betweenness.
Population Studied
We used text data from Reddit and its specific forum or “Subreddit” for Addiction. Through an Application Programming Interface, we gathered the top thousand and most up-voted comments. These comments were collected from 204 unique individuals.
Principal Findings
We found that the clusters did not match. We calculated the Adjusted Rand index and found it to be just slightly above zero (0.01777) meaning that the clusterings of individuals by addiction and morality were approximately as similar as if they had been formed by random chance.
Conclusions
We conclude that more work is needed to empirically test the relationship between moral values and moral behavior and co-occurring addictive behavior in social networks.
Implications for Policy or Practice
This study serves as a catalyst to more research into morality in social networks and substance abuse. If we have more information about what drives individuals in their addictive behaviors, the results can be used to design more effective prevention programs and to monitor and evaluate these programs more effectively. It could help the National Academy of Medicine’s coalition of government and private sector parties, including the CDC, CMS, FDA and addiction groups, to better understand the complex drivers of the opioid syndemic and addiction patterns.