Tweeting negative emotion: An investigation of Twitter data in the aftermath of violence on college campuses.

Abstract

Studying communities impacted by traumatic events is often costly, requires swift action to enter the field when disaster strikes, and may be invasive for some traumatized respondents. Typically, individuals are studied after the traumatic event with no baseline data against which to compare their postdisaster responses. Given these challenges, we used longitudinal Twitter data across 3 case studies to examine the impact of violence near or on college campuses in the communities of Isla Vista, CA, Flagstaff, AZ, and Roseburg, OR, compared with control communities, between 2014 and 2015. To identify users likely to live in each community, we sought Twitter accounts local to those communities and downloaded tweets of their respective followers. Tweets were then coded for the presence of event-related negative emotion words using a computerized text analysis method (Linguistic Inquiry and Word Count, LIWC). In Case Study 1, we observed an increase in postevent negative emotion expression among sampled followers after mass violence, and show how patterns of response appear differently based on the timeframe under scrutiny. In Case Study 2, we replicate the pattern of results among users in the control group from Case Study 1 after a campus shooting in that community killed 1 student. In Case Study 3, we replicate this pattern in another group of Twitter users likely to live in a community affected by a mass shooting. We discuss conducting trauma-related research using Twitter data and provide guidance to researchers interested in using Twitter to answer their own research questions in this domain. (PsycINFO Database Record

Insights into Adolescent Online Conflict through Qualitative Analysis of Online Messages

“…For example, preliminary work has described hand-coded patterns of online conflict among gang-involved adults, suggesting that it is possible to identify high-level patterns of conflict using social media [10]. Others have measured the sentiment of communities, as expressed through social media, after exposure to violence [11,12]. The LIWC (Linguistic Inquiry Word Count) method measures occurrences of words defined in the dictionary to comprise a finite set of categories, such as negative emotion, positive emotion, anxiety, anger, conflict, or sadness, and may be able to be applied to identify online conflict [13].…”

Using Big Data to Study the Impact of Mass Violence: Opportunities for the Traumatic Stress Field

“…This method offers total certainty about where a tweet was generated and, in using this approach, a researcher can capture tweets from residents and visitors in a target location affected by a collective trauma. However, as some Twitter researchers have noted (e.g., Jones, Wojcik, Sweeting, & Silver, ), very few users “opt in” to having their tweets geocoded. In fact, estimates of the percentage of Twitter users who opt in to having their tweet location made public range between ∼6% and ∼8% (Jones et al., ; Lin et al., ), although the true number is only known by Twitter.…”

“…To target likely residents in a community that has been impacted by mass violence, researchers can first identify Twitter accounts of local organizations or agencies in that community and then download the list of users who follow those accounts (cf. Jones et al., ). The rationale is that users who follow the Twitter account of the Riverside, CA, City Hall, for example, are likely to be residents of Riverside, CA, and it is less likely that residents of New York City would follow this Twitter account.…”

“…After compiling lists of users who follow local accounts, researchers can download the most recent 3,200 tweets from each user who possesses a public account (9%–13% of users have private accounts; see Jones et al., ). An additional advantage of this method is that it lends itself to the inclusion of data from a closely matched control community, an important feature of any observational study attempting to make a strong inferential claim about the impact of a mass violence event.…”

“…The utility of Twitter data has also been demonstrated when examining community responses to college campus shootings. For example, Jones and colleagues () sourced Twitter data by identifying Twitter accounts local to communities that experienced incidents of college campus violence (e.g., the Isla Vista killings near the University of California, Santa Barbara, campus in 2014) and downloading tweets from users who followed those accounts. The authors also sourced data from control communities to allow for a comparison of patterns between the target and control communities.…”