Response to a WEA Tornado Warning Text Message

You are outside of your house in the spring when your phone receives a Wireless Emergency Alert (WEA). The alert reads: “Tornado Warning in this area till 3:49 pm. Take shelter now. Check local media. –NWS”. This study focuses on individuals’ perceptions of, and reactions to, such a message. Tornadoes are a dangerous experience that can lead to loss of life and property. Although those who live in Tornado Alley, which includes Midwestern states, are at a higher risk of experiencing a tornado, not everyone follows the recommendation to seek shelter when they receive a tornado warning message. In order to create more effective weather messages, it is imperative to understand how people perceive weather alerts, and what they do in response to those perceptions. The purpose of this study was to examine how beliefs about a tornado alert message, and communicative concerns, shape shelter-seeking intentions. The health belief model concerns how responses to health threats are a function of the perceived severity of, and personal susceptibility to, a health risk, as well as the perceived benefits of, and barriers to, taking action or following prescribed recommendations. Though much utilized in the health field, we believe the variables it highlights have high applicability to weather-related threats and risks. Additionally, given that weather-related threats often place relational partners at risk, we are concerned with the tendency to seek out others in order to verify, as well as provide information, about a weather-related threat (a construct we refer to as “secondary communication”). An online survey was conducted comprising 189 participants. The sample included 125 (66.1%) females and 64 (33.9%) males, with the average age being 37.4 years old. The survey included items concerning secondary communication (motivation to share information with others), health beliefs (perceived vulnerability, barriers and benefits to action), perceived threat and past experience as they relate to intention to seek shelter. Analyzing the data, we found that 108 (57.1%) participants said “yes” to having past experience with a tornado. Correlation analyses revealed that intention to take shelter was related to severity, susceptibility, benefits, barriers, past experience, and secondary communication. In regression analyses, the health belief model variables, along with past experience, accounted for 36.5% of the variance in shelter-seeking intentions, with past experience, severity and susceptibility being significant predictors. Additionally, severity, susceptibility, and benefits predicted messages cues and message directness. In mediation analyses, message cues mediated the relationships of severity, susceptibility, benefits, and barriers shelter-seeking intentions. Message directness mediated susceptibility and benefits on intentions. Regarding secondary communication, messages cues mediated the relationship between severity, susceptibility, benefits, and barriers on seeking out others following receiving the alert. Based on these results, three conclusions are offered. First, the health belief model variables are relevant to weather-related risks and could be targeted for educational purposes. Second, text alert messages do play a role in increasing intentions to seek shelter, especially when pre-existing beliefs about severity, susceptibility, benefits, and barriers are strong. Third, the tendency to seek out others when an alert has been received is a relevant factor; safety is not just an issue of personal concern. Further research on this tendency offers opportunities to create more effective messages by educating the public about how and when to share information in their social networks.