EPP Research: Risk Analysis and Risk Communication
Combining expertise in science, engineering, economics, decision analysis, psychology, and behavioral decision theory, EPP is home to one of the world's leading research programs in risk analysis and risk communication. Mike DeKay, Scott Farrow, Baruch Fischhoff, Paul Fischbeck, Keith Florig, Carol Goldburg, Lester Lave, Indira Nair, Granger Morgan, Henry Piehler, Mitchell Small, and Elizabeth Casman all do research in this area.
Risk-related work in the department ranges from the analysis of new or under appreciated risks (home sprinkler fire suppression systems, alternative motor vehicle fuels, off-shore platforms for oil and gas) to studies of public perceptions and valuation, how best to improve public understanding of risks, how to facilitate public participation in risk-related decision making, and how to improve the management and regulation of risks. Much of this research is done collaboratively in interdisciplinary teams.
The field of risk analysis has undergone major changes over the past 25 years. Much of the early work involved straight technical assessment by people in fields such as engineering and public health. This began with fairly simple deterministic methods and, with time, moved to increasingly sophisticated probabilistic methods. However, it soon became apparent that issues of risk involve far more than just technical assessment. Issues of public communication, public understanding and perception, valuation, and citizen participation, as well as policy questions related to the choice and implementation of risk management strategies, can all be as or more important than technical assessment. As a result, economists, psychologists, and other social scientists began to work on risk. Unfortunately, even today, the perspectives of these various disciplines are often not combined in an integrated approach to risk analysis. Fortunately, at Carnegie Mellon, EPP has long enjoyed an institutional environment that allows a close integration of these perspectives both in our research and in the education of our students.
One recent example of this integrated approach has been work in the Department on risk ranking. In the Fall of 1993, the White House Office of Science and Technology Policy asked faculty in EPP to propose a detailed method by which federal agencies might rank the risks they manage. A group of four faculty and seven graduate students in the department published a detailed description of how this might be done. Since then, with support from NSF and EPA, a group of researchers in the department, including Granger Morgan, Paul Fischbeck, Mike DeKay, Keith Florig, Baruch Fischhoff and others, have worked to refine, test, and demonstrate these ideas using citizen groups and a realistic test bed that involved health and safety risks to students in a hypothetical middle school. Now the method is being extended to include ecological risks as well. Developing these methods has spun off basic research into the nature of risk and the logic of prioritization.
Baruch Fischhoff was among the pioneers of behavioral decision making and the psychology of risk perception. In addition to his work as part of the effort in risk ranking he is part of the Center on the Human Dimensions of Global Change, studying how best to support people in valuing outcomes and situations with which they have had little or no prior experience. He is also conducting work on topics that range from adolescent risk taking to people's willingness to pay for environmental protection, involving citizens in risk decision making, getting relevant science used in risk decision making, and communicating with laypeople about risks to public health (AIDS, cryptosporidiosis, radon, breast cancer, sexual assault, etc.). He is also doing work on research priority setting and evaluating the usefulness of basic research.
Much of the risk-related work in EPP overlaps with work in the area of energy and environmental systems. For example, Lester Lave has done extensive work on the health risks of air pollution, recently extending the focus to include temperature effects and climate change. As part of the Center of the Study and Improvement of Regulation, Lave is studying the development of improved techniques for screening for chemical carcinogens and the evolution of air pollution regulations for motor vehicles and their fuels. As noted above, he leads the university-wide Green Design Initiative which is pioneering methods in life-cycle analysis. Other recent risk studies have ranged from an examination of the risks from truck drivers who have diabetes to work on the risk from large dams.
Paul Fischbeck also conducts research on a wide variety of topics in risk assessment. Examples of recent projects include risks posed by air pollution from ships, development of performance-based fire codes, design of improved inspection methods for the protective heat shield tiles on the space shuttle, studies of the risks of off-shore oil platforms, and the clean-up and reuse of old contaminated industrial sites.
Like Fischhoff, Lave and Fischbeck, Granger Morgan has addressed a wide variety of issues in risk assessment, from power-frequency electric and magnetic fields to motor vehicle crashes. As part of the activities of the Center for the Study and Improvement of Regulation he is exploring issues such as how to measure the performance of current regulatory systems, how to do a better job of balancing equity and efficiency in the design of regulation, and whether and how to undertake fundamental redesign of environmental enabling legislation. On the analytical front, much of his work has involved the development and demonstration of methods for characterizing and treating uncertainty in quantitative policy analysis.
Mitchell Small's research on environmental fate and transport, exposure, and risk considers the role of risk perception and risk communication on individual behaviors which influence exposure. Recently he has collaborated with Baruch Fischhoff to develop integrated methods for household chemical exposures and is currently involved in a project to develop protocols for risk communication for microbial pathogens in drinking water. He collaborates extensively with colleagues in the Department of Statistics on the development and application of Bayesian methods for environmental modeling and risk assessment.
In the area of risk communication, Baruch Fischhoff and Granger Morgan have co-directed a series of studies of the problems of communicating to semi-technical and non-technical people about technological risks. These studies have shown that people do not process and interpret such information in isolation. Rather, they filter and interpret it in the context of their existing knowledge structures. Thus, risk communications that ignore what people already believe can both confuse and fail to inform them. As a result, one focus of this efforts has been on the development of a set of techniques for eliciting the "mental models" that people use when thinking about risk issues. This research has provided a disciplined basis for developing and empirically testing risk communications. For more information on activities in EPP and elsewhere at Carnegie Mellon on Risk Perception and Communication see http://www.hss.cmu.edu/departments/sds/risk/.
For example, mental model interview methods have recently been used to conduct studies of what laypeople know about climate change. These studies found that most people do not understand the key role played by carbon dioxide produced when fossil fuel is burned. Many people appear to have a mental model that general pollution causes climate change and good "green" practice will somehow solve the problem. In order to help people develop a more informed basis for participating in public debate on this issue, a hierarchically organized set of brochures on climate change has been developed for the general public.
Mike DeKay is working on developing strategies to support laypeople in evaluating ecological impacts in risk ranking and other contexts. His other environmental interests include the prioritization of endangered species and their habitats for preservation. He is also interested in health-related risks and decision making. Projects in this area have involved decisions about risky treatments for individual patients versus groups of similar patients; the effects of malpractice liability on physicians’ decisions about diagnostic testing; cost-effectiveness analysis of population-based genetic screening programs; and prioritization of candidates for organ transplantation.
Hadi Dowlatabadi and Elizabeth Casman are conducting a series of risk analytic and econometric projects exploring the connections between climate change and vector-borne disease. These projects are examining the socio-economic context of disease transmission in an effort to quantify various contributions to overall risk.