Abstract: “Where do I go from here?” Support Seeking and Message Characteristics
Social support plays a significant role in human life, and receiving social support during times of crisis may be vital to human well-being. Social support is no longer bound to friends and family members, nor to any proximity boundaries. Online environments offer unique benefits to people in need of social support, including flexibility, convenience, and a sense of anonymity (Wright, 2016). Social support forums are an important resource for people experiencing an illness who lack others to communicate with (Wicks et al., 2010). Social support scholars have recently turned to the implications of computer-mediated-communication (CMC) and research shows that social support via CMC has the potential to have a larger impact than face-to-face interaction. Rains et al. (2016) found that compared to the face-to-face interaction, participants provided with social support via CMC experienced the most benefits, including the strongest motivation to receive support and the greatest changes in worry discrepancy. Due to the positive findings surrounding mediated support interactions, it is clear that an extension to this knowledge, examining how people seek support online and the nature of these support-seeking messages, is warranted. This study incorporates computational social science methods, including data scraping, data mining, and association rules, to obtain and analyze data from the prominent social support website dailystrength.org. The data set includes approximately 1500 original posts, and the first associated responses, from 15 forum topics (e.g., breast cancer, smoking addiction, depression). The posts were run through the Linguistic Inquiry and Word Count (LIWC) program, which is currently perceived as the gold standard in computerized text analysis. LIWC provides a broad range of social and psychological insights and, in our study of support forums, allows us to understand the underlying rules and associations regarding people’s written interaction. This analysis included 32 additional variables such as title word count, mood level, number of comments, reply delay, and word count of the first reply. This study sought to understand: (a) what aspects of an initial post cause more people to provide support, (b) what types of support are offered by the first person to respond, and (c) if there are differences between forums. Results indicate that 92% of posts (approximatelly 22,000) received at least one response, responses that included emotion words were more likely to focus on positive emotions than negative, and support providers were significantly more likely to help the support seeker cognitively process their current situation. A key finding of this study indicates that people who are seeking support online are likely to receive it. This is especially important for those with a stigmatized illness, who are unable to obtain support from their offline network, and who may not be in close proximity to physical support groups. In addition, multiple variables indicated differences between forums. This finding suggests that research in support seeking and support behaviors needs to move beyond illnesses, in a general sense, into context-specific support behaviors for different types of illnesses. Other implications of this study, practical applicability, and future directions are discussed.