Households disaster memory recollection after the 2013 Colorado flood

Abstract

There is some evidence supporting that surveys conducted 12–18 months after a disaster can provide accurate assessments of people’s disaster responses during disasters. Studies suggest that people have good memories of events that are personally relevant to them and that there appear to be reasonable justifications for taking post-event survey data at face value. Nonetheless, according to the Protective Action Decision Model, people’s disaster response activities include behavioral and emotional responses. Since these two types of responses are different in nature, it is unclear whether people have good memory recollection of both types of responses. Thus, it is important to obtain additional evidence to test survey respondents’ memory recollection over time. To do this, this study collected the 2013 Colorado flood household response data 7 months and 14 months after the event. Box’s homogeneity test is used to test the equivalence of covariance matrices. The results indicate that survey respondents’ behavioral responses follow similar patterns between two survey samples, but the emotional responses do not. This finding suggests that survey studies are able to acquire accurate disaster behavioral response data 14 months after a disaster; however, emotional response is considered ephemeral data.