This project is a study of how organizations collect information about and learn from near accidents, events which under slightly different circumstances could have resulted in an accident. Organizations monitor these potential dangers to learn how to cope with dangers, i.e., low probability, high negative consequence events. In this research, a comparison is made to how near accident monitoring systems operate in the medical device, nuclear power, space shuttle, and air transportation industries. Specifically, the research is based on an analysis of interviews with key participants in a variety of safety information systems, and reports generated within previous studies in the air transportation industry. The proposed research draws on and seeks to contribute to the study of risk management, information processing, and organizational learning. Additionally, it has broader implications for organizational management of other areas of risky events, such as the extreme weather events that may become more frequent or unpredictable under global environmental change. The purpose of the projects is to identify factors which constrain (or enhance) the organizational capacity to collect information about and learn from near accidents (or near disasters). The focus is on three main research questions: 1) How does the organization of safety information systems affect the volume (e.g., reliability) and contents (e.g., validity) of information collected and the organizational capacity to learn from near accident information? 2) To what extent do attention-related factors affect the volume of near accident reporting over time? and 3) How do organizations learn from near accidents? Specific research questions are developed for each of these main research questions. To illustrate, the first research question includes a discussion of the definition and interpretation of near accidents in which the dimensions of vulnerability vs. robustness, learning vs. controlling behavior, and ambiguous vs. clearly -defined definitions are considered. The research will help address such policy questions as: If an organization collects information for purposes of controlling behavior, does it lessen the amount of near accident reporting and increase the self-interested bias that distorts accounts of potential dangers?
