From 1997 to 2006 an average of 649 people per year died from weather-related events in the United States. This included deaths due to lightning, hurricanes, floods, tornadoes, winter storms, and extremes of heat and cold (National Weather Service, NWS, 2007). Severe or extreme weather events resulted in annual average damages and losses of over 230 million dollars for the previous decade. In the most recent year of 2006, 566 people died from weather-related events and another 3,489 people were injured; 26% of weather-related deaths involved children aged nine years or younger (NWS, 2007). These statistics do not include the 4,362 people who died in motor vehicle crashes in the United States in 2005 during inclement weather conditions (e. g., rain, sleet); an additional 216,000 people were injured in motor vehicle crashes that occurred under these conditions (National Highway Transportation Safety Administration, 2006).

Some of these deaths and damages were unavoidable because of the rapid onset of the severe weather events (e. g., tornadoes occurring late at night). In other cases such as floods or hurricanes where the longer forecast lead times allow for securing property, making necessary preparations, and evacuating a threatened area, it is possible to reduce the rates of injury, death and destruction. Yet, despite receiving severe weather warnings and having time to prepare, not everyone in an affected area will take steps to assure their safety.

What are some of the possible reasons for this? As was evident when hurricane Katrina made landfall in New Orleans, some people who received the warnings to evacuate did not have resources (vehicles, money for travel, acquaintances that could assist them) to evacuate. Other people had elderly or infirm family members which led to their choosing to stay in harms way when others evacuated. Still other people with prior experience in weathering past Gulf Coast hurricanes felt seasoned by their experiences and thus decided to ride the storm out, protect their property, and avoid the difficulties associated with trying to return following an evacuation (Blendon, 2006; Dow & Cutter, 2000). Additional reasons for peoples' weather-related risk taking obviously could be found. Ultimately, peoples' decisions about whether to heed weather warnings or to disregard or downplay them stems from their weighing the risks posed by severe/extreme events against the perceived benefits from taking some actions other than those safety measures prescribed by meteorologists, law enforcement, and emergency managers, among others.

One construct that has not been examined previously in the psychology literature that bears upon the safety decisions people make involves weather risk-taking. Prior research suggests that risk-taking is highly domain specific (i. e., someone who takes risks recreationally may be highly averse to taking socially-based risks, Weber, Blais, Betz, 2002). By extension, then, it is likely that people differ with respect to their evaluation of various weather-related risks, their likelihood of taking such risks, and the benefits/outcomes they expect by taking such risks.

The present project describes the construction of the Weather Risk-Taking Scale (WRTS). This measure consists of 32 items whose content embodies risks or lack of known safe behaviors that are associated with injuries of deaths from thunderstorms, lightning, floods, tornadoes, hurricanes, winter weather, heat, and cold. In this regard sources such as the National Lightning Safety Institute, the National Weather Service, and the Tropical Prediction Center were consulted to embody the relevant content in the items. The 32 items are evaluated using a seven-point rating scale with respect to: 1. the respondent's likelihood of engaging in the behavior, 2. their evaluation of the risks associated with the behaviors and 3. the expected benefits they might experience from engaging in the behavior.

The research presentation will discuss the psychometric properties of the WRTS (including Cronbach's α , item factor structure, and the WRTS correlations with selected scales of the Domain Specific Risk-Taking Scale (Blais & Weber, 2006). Using Protection Motivation Theory the presentation will show how the likelihood of performign risky weather behaviors is related to perceptions of weather risk, perceived benefits, fear of weather, and weather salience.