Tell Me How You Really Feel about Drug Users: An Examination of the Social Context of Stigma Toward those with a Substance Use Disorder using a Targeted Sample of An Online Social Network
Glenn Sterner, Sean Parsons
Stigmatizing language and negative sentiment towards those with a substance use disorder and substance use more broadly can negatively impact outcomes associated with treatment and recovery for affected populations. Negative social sentiment toward substance use disorder can create a context in which those with substance use issues may delay treatment seeking behaviors, including reaching out to friends and family as well as pursuing professional services. In addition, this social context may influence their own self image and internalized stigm that can reduce their prospects of recovery. There is evidence that online “anonymous” or identity obscured activity may reflect personal sentiment in a more raw and accurate nature than those opinions presented in unobscured interpersonal communications, indicating the sentiment of these communications may be more accurately capturing contextual perspectives.
This presentation examines the social context toward substance use in the Philadelphia metropolitan area of the United States, a particularly negatively affected area of the opioid crisis in the country with persistently high rates of overdose deaths. To examine sentiment toward substance use and those with a substance use disorder or addiction, we utilized postings from an online social network called Reddit. This online social network allows individuals to post on message boards, allowing comments and subcomments to build from these postings. Within this social network, one can create communities of interest called subreddits, where the topic is focused on areas relevant to those creating the postings ranging from topical issues to geographic localities. We specifically collected postings from subreddits that are based around topical and geographic interests associated with the Philadelphia metropolitan region. To identify postings, comments, and subcomments relevant to our topic, we utilized a Python program that searched the Reddit API utilizing key words associated with substance use. To ensure validity, we screened these data to ensure relevance to our topic matter. We then utilized the Stanford language processing library to determine the sentiment (negative or positive) of these data utilizing a Python script.
In our analysis, we examine three networks. First, we explore the one-mode networks of postings to explore sentiment spread toward substance use issues within online communities. Second, we examine the one-mode network of users posting on substance use issues across online communities. Finally, we examine the two-mode network of users and their postings across online communities to examine influencers of online sentiment toward substance use issues. We also include descriptive data to explore the context of sentiment toward substance use and substance use disorder in the Philadelphia metropolitan region.← Schedule