Teaching Probability, Uncertainty, and Hazard Communication to Undergraduates: A First-Year Seminar Experience

The effective provision of impact-based decision support for high-impact weather events requires effective communication about what is possible, what is probable, and what actions can be taken to prepare for and/or respond to the specific hazards associated with a given weather event. Historically, meteorologists have communicated this information using terms that they understand via products that were developed for specific phenomena. Yet, recent social science research indicates that such communication is most effective when viewed as a process that focuses on end-user needs, interests, and circumstances. This requires that communicators understand how people make decisions, particularly when one or more possible or probable outcomes have a high probability of being negative.

Motivated by the disconnect between how meteorologists and the general public perceive and respond to hazard-related information, in Fall 2016, I taught a First-Year Seminar course titled, “A 30% Chance of What? Communicating the Forecast.” A total of fourteen freshmen, all but one of whom were weather-interested but were not atmospheric science majors, completed the course. Though the cohort had limited prior meteorological knowledge, it was representative of a college-educated general populace to whom meteorologists are often communicating. The course started with a discussion of communication and decision-making for a specific high-impact weather event, Hurricane Sandy in 2012, before broadening to consider weather-related risk communication and behavior best practices and, finally, the psychology of decision-making independent of weather-related context.

In this presentation, I will discuss the hands-on methods used to help students understand and give them experience with putting into practice key course concepts. Student and instructor perspectives on the course’s content and structure will be discussed, including a documentation of best practices for similar classes and broader implications for impact-based decision support.