Identifying and Assessing the Effects of Forecast and Warning Communication (In)Consistency on Different Recipients

There are many places where people can find weather forecast and warning information. In addition to the NWS and local and cable television, there are private sector companies that provide information, much of which can be obtained on mobile phones and the Internet. Anyone seeking such information knows that there can be inconsistencies in the messages provided, and consistency can depend on audience, scale, and message purpose. If the goal of the weather enterprise is to provide information that leads to protection of life and property, inconsistent or confusing messages can be a hindrance. This is complicated by plans to provide more forecast uncertainty and probabilistic information. Our panel will discuss scholarship and paths forward not only from meteorology but also from the fields of medicine, health communication, and psychology. Prepared remarks will be given by four panelists, followed by open discussion. Panelists include: Castle Williams is a Ph.D. student in the Department of Geography at the University of Georgia. His research interests include examining how we communicate weather terminology and hazards to the public, as well as increasing the amount of interdisciplinary projects within the atmospheric sciences. His dissertation work is focused on the role of message consistency in the weather enterprise, specifically assessing the importance of having a “consistent” visual design when communicating weather-related risk, uncertainty, and probabilistic information. Castle’s presentation is titled “Inconsistent ‘Consistency’ Conversations: The Past, Present, and Future of Message Consistency in the Weather Enterprise.” According to Mileti and Sorensen (1990), a warning’s ability to encourage an individual to perform a given protective action is best evaluated among the following dimensions: warning source; warning channel; the consistency, credibility, accuracy, and understandability of the message; and the warning frequency. While most of those factors have been thoroughly investigated in the context of the weather enterprise, the consistency of weather messages has become a growing concern due to the mass availability of weather information via Internet and mobile-based devices. Though several professional panels of meteorologists have attempted to tackle this intricate concept, the idea of consistency within the weather community remains a formidable hurdle and each discussion only adds more complexity to this issue. His presentation will provide an overview of previous “consistency” conversations held in the weather enterprise, explore the concept of “consistency” through several different disciplines, and look ahead to identify research gaps and operational challenges associated with having a “consistent” message. Dr. Delesha Carpenter is an Assistant Professor in the Division of Pharmaceutical Outcomes and Policy at the Eshelman School of Pharmacy at the University of North Carolina-Chapel Hill. Her educational background includes clinical training in respiratory therapy and graduate training in health behavior and health education. She is particularly interested in understanding how individuals process and react to inconsistent or conflicting health information. She has authored more than 80 peer-reviewed publications and has led multiple research studies as a principal investigator. Dr. Carpenter’s presentation, “Understanding how people react to conflicting information: Insights from the field of pharmacy,” will explore data from two studies. The first examined how patients living with arthritis reacted to hypothetical scenarios in which conflicting medication information was received. Specifically, she will discuss four common reactions to conflicting information and how these reactions may also apply to inconsistent weather information. Second, using data from a longitudinal study of arthritis patients, she will discuss how patients’ ability to accurately recall medication information is related to their perceptions of conflicting information and information-seeking behaviors. She will conclude by presenting a definition and conceptual typology of conflicting information that is based on results from these and other studies. Dr. Paul K. J. Han, MD, MA, MPH, is the Director of the Center for Outcomes Research at Maine Medical Center, Associate Professor of Medicine at Tufts University School of Medicine, and a behavioral and health services researcher and board-certified general internist and palliative care physician. He has conducted several conceptual and empirical research studies examining the nature, communication, and management of uncertainty in healthcare. Dr. Han’s presentation, “Conceptualizing and communicating inconsistency and other uncertainties in prognostic information,” will explore the nature and effects of inconsistency and other forms of uncertainty in prognostic information, drawing on conceptual and empirical work from health care and other fields. He will describe how inconsistent prognostic information about the weather and other important phenomena relates to other types of uncertainty. He will then discuss potential psychological effects of uncertainty arising from inconsistent information and other sources. Finally, he will outline potential strategies for communicating and managing uncertainty arising from inconsistent prognostic information. Dr. Susan Joslyn is an Associate Professor of Psychology at University of Washington. She is a cognitive psychologist whose work addresses real-world decisions made under uncertainty, such as those involving severe weather, climate change, and medical treatment. Her research is largely experimental, bringing the principles of information processing and decision theory to bear on a series of applied questions. Her interdisciplinary research makes important theoretical contributions to the psychology of decision making as well as having implications for best practices in uncertainty communication techniques in several domains. She serves on the Pacific Northwest Integrated Weather Team and the Advisory Board for the Alliance for Integrative Approaches to Extreme Weather Events. Dr. Joslyn’s presentation is titled “What is Forecast Consistency?” Consistency between forecasts for the same target date is thought to increase user trust, although the empirical evidence for this effect is sparse. Moreover, in the case of sequential forecasts, a difference in subsequent forecasts might be due to updated information that is more accurate than the information upon which the previous forecast was based. In this case maintaining consistency (by making the subsequent forecast align with the earlier one) may be at the sacrifice of accuracy. A series of five experimental studies was conducted to systematically test the impact on user trust of forecast consistency and forecast accuracy. The implications of this research will be discussed in the context of forecast policy and consistency in general.