Fighting Together: Discovering the Antecedents of Social Support and Helpful Discussion Threads in Online Support Forums for Cannabis Quitters

Abstract

An increasing number of people are using online cannabis support forums as a source of help for their cannabis quit attempts. In order to assist support seekers dealing with emotional and physical-behavioral difficulties associated with their cannabis abstinence, it is important to identify the factors that facilitate social support provisions by forum members, as well as the overall helpfulness of discussion threads. In this combined qualitative and quantitative study, we propose a model hypothesizing and testing these factors, based on variables generated using Natural Language Processing and Machine Learning techniques. The result shows that linguistic and content characteristics of thread-initiating messages are important predictors of the receptions of informational and emotional support from other forum members, and of the overall helpfulness of discussion threads.