Media Effects on Public Safety following a Natural Disaster: Testing Lagged Dependent Variable Models

Abstract

This study assesses news and media campaign effects during the recovery phase of a catastrophe. With data from a panel telephone survey in New Orleans in 2006, this study tests lagged dependent variable models for safety beliefs and safety behavior in the context of Hurricane Katrina. News attention and media campaign exposure influenced safety behavior. The effects of news attention were synchronous, while those of media campaign exposure were cross-lagged. In contrast, neither news attention nor media campaign exposure influenced safety beliefs, which may be attributable to ceiling effects of the belief measure. Safety beliefs did, however, have a cross-lagged influence on safety behavior.