Risk Analysis, Vol. 00, No. 00, 2017 DOI: 10.1111/risa.12845 Risk Communication Emergency Response Preparedness: Contextual Assessment of the Protective Action Decision Model Robert L. Heath, 1,* Jaesub Lee, 1 Michael J. Palenchar, 2 and Laura L. Lemon 3 Studies are continuously performed to improve risk communication campaign designs to bet- ter prepare residents to act in the safest manner during an emergency. To that end, this arti- cle investigates the predictive ability of the protective action decision model (PADM), which links environmental and social cues, predecision processes (attention, exposure, and com- prehension), and risk decision perceptions (threat, alternative protective actions, and stake- holder norms) with protective action decision making. This current quasi-longitudinal study of residents (N = 400 for each year) in a high-risk (chemical release) petrochemical man- ufacturing community investigated whether PADM core risk perceptions predict protective action decision making. Telephone survey data collected at four intervals (1995, 1998, 2002, 2012) reveal that perceptions of protective actions and stakeholder norms, but not of threat, currently predict protective action decision making (intention to shelter in place). Of signif- icance, rather than threat perceptions, perception of Wally Wise Guy (a spokes-character who advocates shelter in place) correlates with perceptions of protective action, stakeholder norms, and protective action decision making. Wally’s response-efficacy advice predicts res- idents’ behavioral intentions to shelter in place, thereby offering contextually sensitive sup- port and refinement for PADM. KEY WORDS: Emergency response; protective action decision model; risk spokes-characters: risk com- munication; theory of reasoned action 1. INTRODUCTION “A risk communication effort will succeed only by making information personally relevant and ap- propriate and by conveying methods of control- ling the risks (personal efficacy),” so concluded ter Huurne and Gutteling (p. 820). (1) In light of that advice, it is intriguing to investigate how well the 1 Valenti School of Communication, University of Houston, Hous- ton, TX, USA. 2 School of Advertising & Public Relations, University of Ten- nessee, Knoxville, TN, USA. 3 Department of Advertising and Public Relations, University of Alabama, Tuscaloosa, AL, USA. ∗ Address correspondence to Robert L. Heath, Valenti School of Communication, University of Houston, Houston, TX 77204, USA; rheath1941@me.com. protective action decision model (PADM) (2) justi- fies the use of emergency response messages that are personally relevant, appropriate, and focused on personal efficacy. PADM should shed light on (1) how residents learn environmental safety cues and (2) use core risk perceptions of threat, protec- tive action, and stakeholder norms to (3) motivate their behavioral intentions (protective action deci- sion making) and thereby (4) predict whether, when, and how citizens will respond (shelter in place, SIP) during a risk event of the type (chemical release) featured in this study. Also, this study answers a call by Lindell and Whitney (3) to “use longitudi- nal experimental designs involving credible sources delivering persuasive messages directed toward re- ceivers’ beliefs about those adjustment attributes to 1 0272-4332/17/0100-0001$22.00/1 C  2017 Society for Risk Analysis