Content Mapping: An Experimental Analysis of Missing Message Content
Jeannette Sutton, Hugh Walpole, Micki Olson, Lauren B. Cain, Michele M. WoodABSTRACT
The Warning‐Response Model recommends the inclusion of five key content areas in warning messages to reduce information seeking that can lead to warning response delay: the message source, hazard (including its impact), affected location, recommended guidance, and time. Research has found that complete messages, that is, those that include all five recommended content areas, help people make decisions about the protective actions they may take. Unfortunately, a large majority of Wireless Emergency Alert (WEA) messages omit information from one or more of the recommended content areas. This study examined the impact of omitting each of the five recommended content areas on message outcomes from the Warning‐Response Model (understanding, believing, personalising, and deciding), self‐efficacy, response efficacy, information seeking, and organisational trust. To this end, we conducted a 2 × 7 experiment with two hazard types (wildfire and chemical release), and we compared a complete message containing all five message elements (control) to messages that were missing either the message source, hazard name (plus hazard impact statement), location, time, or guidance to determine the effect of the missing content on Warning‐Response Model and other message outcomes. Findings indicate that the omission of different types of content affects message outcomes in different ways for different hazards, shedding light on a hierarchy of content functions within theories of warning communication. This is the first study that has focused entirely on examining which content areas are the most critical drivers of Warning‐Response Model outcomes across multiple hazard types.