Analyzing risk response dynamics on the web: the case of Hurricane Katrina

Critical Time, Space, and Decision-Making Agent Considerations in Human-Centered Interdisciplinary Hurricane-Related Research

In hazard and disaster contexts, human-centered approaches are promising for interdisciplinary research since humans and communities feature prominently in many definitions of disaster and the built environment is designed and constructed by humans to serve their needs. With a human-centered approach, the decision-making agent becomes a critical consideration. This article discusses and illustrates the need for alignment of decision-making agents, time, and space for interdisciplinary research on hurricanes, particularly evacuation and the immediate aftermath. We specifically consider the fields of sociobehavioral science, transportation engineering, power systems engineering, and decision support systems in this context. These disciplines have historically adopted different decision-making agents, ranging from individuals to households to utilities and government agencies. The fields largely converged to the local level for studies' spatial scales, with some extensions based on the physical construction and operation of some systems. Greater discrepancy across the fields is found in the frequency of data collection, which ranges from one time (e.g., surveys) to continuous monitoring systems (e.g., sensors). Resolving these differences is important for the success of interdisciplinary teams in protective-action-related disaster research.

Factors Affecting Hurricane Evacuation Intentions

Protective actions for hurricane threats are a function of the environmental and information context; individual and household characteristics, including cultural worldviews, past hurricane experiences, and risk perceptions; and motivations and barriers to actions. Using survey data from the Miami-Dade and Houston-Galveston areas, we regress individuals' stated evacuation intentions on these factors in two information conditions: (1) seeing a forecast that a hurricane will hit one's area, and (2) receiving an evacuation order. In both information conditions having an evacuation plan, wanting to keep one's family safe, and viewing one's home as vulnerable to wind damage predict increased evacuation intentions. Some predictors of evacuation intentions differ between locations; for example, Florida respondents with more egalitarian worldviews are more likely to evacuate under both information conditions, and Florida respondents with more individualist worldviews are less likely to evacuate under an evacuation order, but worldview was not significantly associated with evacuation intention for Texas respondents. Differences by information condition also emerge, including: (1) evacuation intentions decrease with age in the evacuation order condition but increase with age in the saw forecast condition, and (2) evacuation intention in the evacuation order condition increases among those who rely on public sources of information on hurricane threats, whereas in the saw forecast condition evacuation intention increases among those who rely on personal sources. Results reinforce the value of focusing hurricane information efforts on evacuation plans and residential vulnerability and suggest avenues for future research on how hurricane contexts shape decision making.

Dynamic simulation as an approach to understanding hurricane risk response: insights from the Stormview lab

This article investigates the use of dynamic laboratory simulations as a tool for studying decisions to prepare for hurricane threats. A prototype web-based simulation named Stormview is described that allows individuals to experience the approach of a hurricane in a computer-based environment. In Stormview participants can gather storm information through various media, hear the opinions of neighbors, and indicate intentions to take protective action. We illustrate how the ability to exert experimental control over the information viewed by participants can be used to provide insights into decision making that would be difficult to gain from field studies, such as how preparedness decisions are affected by the nature of news coverage of prior storms, how a storm's movement is depicted in graphics, and the content of word-of-mouth communications. Data from an initial application involving a sample of Florida residents reveal a number of unexpected findings about hurricane risk response. Participants who viewed forecast graphics, which contained track lines depicting the most likely path of the storm, for example, had higher levels of preparation than those who saw graphics that showed only uncertainty cones-even among those living far from the predicted center path. Similarly, the participants who were most likely to express worry about an approaching storm and fastest to undertake preparatory action were those who, ironically, had never experienced one. Finally, external validity is evidenced by a close rank-order correspondence between patterns of information use revealed in the lab and that found in previous cross-sectional field studies.

Applications of functional data analysis: A systematic review

BACKGROUND

Functional data analysis (FDA) is increasingly being used to better analyze, model and predict time series data. Key aspects of FDA include the choice of smoothing technique, data reduction, adjustment for clustering, functional linear modeling and forecasting methods.

METHODS

A systematic review using 11 electronic databases was conducted to identify FDA application studies published in the peer-review literature during 1995-2010. Papers reporting methodological considerations only were excluded, as were non-English articles.

RESULTS

In total, 84 FDA application articles were identified; 75.0% of the reviewed articles have been published since 2005. Application of FDA has appeared in a large number of publications across various fields of sciences; the majority is related to biomedicine applications (21.4%). Overall, 72 studies (85.7%) provided information about the type of smoothing techniques used, with B-spline smoothing (29.8%) being the most popular. Functional principal component analysis (FPCA) for extracting information from functional data was reported in 51 (60.7%) studies. One-quarter (25.0%) of the published studies used functional linear models to describe relationships between explanatory and outcome variables and only 8.3% used FDA for forecasting time series data.

CONCLUSIONS

Despite its clear benefits for analyzing time series data, full appreciation of the key features and value of FDA have been limited to date, though the applications show its relevance to many public health and biomedical problems. Wider application of FDA to all studies involving correlated measurements should allow better modeling of, and predictions from, such data in the future especially as FDA makes no a priori age and time effects assumptions.