Social Sharing of Negative Emotions in Virtual Travel Communities

Virtual communities offer a social environment in which several emotional processes take place. After a collective emotional event, such as a terrorist attack, these communities can provide individuals a space for social contact and emotional support and an opportunity to reduce the negative emotions arising from these situations (Rimé, 2009). The novelty and importance of this type of event coupled with the broad media coverage elicit strong negative emotions, such as sadness and fear, and stimulate people to engage in social sharing (Luminet et al., 2004). Socializing negative emotions help individuals to obtain cognitive clarity, and to regulate their emotions by venting, obtaining emotional and social support, and being in touch with others (Duprez et al., 2015). Also, people may feel stimulated to express their sympathy, and share feelings of sorrow or concern for those involved in the event (Eisenberg & Miller, 1987). The objective of this article is to analyze the social sharing of negative emotions triggered by a terrorist attack in a virtual travel community. Specifically, we examine the cases of anger, sadness, and fear. We also analyze sympathetic feelings expressed in these contexts. We conducted a multi-method study in two steps. Firstly, we collected all threads and their replies in the TripAdvisor Paris Forum from 01/01/2014 to 31/12/2017 (1,072,499 posts, including 101,393 opening posts and replies), comprising the URL, name of posting person, date and time, index of reply, content, number of previous posts on TripAdvisor, and location. Emotions were extracted with the R package syezhet (Jockers, 2015) using a large dictionary for word-emotions associations developed via crowdsourcing on Amazon’s Mechanical Turk. For the quantitative data analysis, we compared the threads and their replies seven days before and after the Paris terrorist attack on several characteristics by performing t-tests of the mean or the proportion. We also compared sympathy scores of the period after the attack across various groups of threads. Secondly, we conducted a qualitative analysis of the 20 largest threads created within seven days in the Paris Forum after the Paris terrorist attack and the thread that was created after the Nice terrorist attack on July 14, 2015 in the same forum.