in 1996, roy schmidt, kalle lyytinen, mark keil, and paul chule conducted an international delphi study to identify software project risks. despite the success stories in the literature, many software development projects still end in failure. the study aimed to develop an authoritative list of risks, determine which risks are perceived as more important, and develop risk categories that can support theory development. previous research on risk management relied on descriptions of good practice observed by researchers. recent work attempted to build a theoretical framework for understanding risk management, using this body of research as a starting point. the authors' main concern was to address risk in a way that improves the ability to resolve risks. they aimed to clarify the interplay between risk identification and risk resolution in theory and study it in empirical research. the study used a "ranking-type" delphi survey to elicit opinions from a panel of experts through iterative, controlled feedback, producing a rank-order list of risk items. the risk items were also classified into general categories helpful for developing coping strategies and theoretical linkages. the study also examined how risk perceptions might differ across cultures, organizational settings, and economic conditions by conducting three simultaneous surveys in three different settings: hong kong, finland, and the united states. the panels consisted of very experienced project managers known to the researchers in each country. at the conference, the authors presented a comprehensive list of risk items formed by combining input from all three panels and ranked lists produced independently by each panel. they also discussed their analysis of the lists and contrasted their findings with previous research. they presented a meta-classification of risk items supporting further theoretical development. finally, they engaged the attendees in a discussion of the implications of their findings for further research and practice. an extended working paper with complete data, findings, and analysis was available at the conference in limited quantity. the authors were happy to supply a copy of the working paper by mail to anyone who requested it at or after the conference.in 1996, roy schmidt, kalle lyytinen, mark keil, and paul chule conducted an international delphi study to identify software project risks. despite the success stories in the literature, many software development projects still end in failure. the study aimed to develop an authoritative list of risks, determine which risks are perceived as more important, and develop risk categories that can support theory development. previous research on risk management relied on descriptions of good practice observed by researchers. recent work attempted to build a theoretical framework for understanding risk management, using this body of research as a starting point. the authors' main concern was to address risk in a way that improves the ability to resolve risks. they aimed to clarify the interplay between risk identification and risk resolution in theory and study it in empirical research. the study used a "ranking-type" delphi survey to elicit opinions from a panel of experts through iterative, controlled feedback, producing a rank-order list of risk items. the risk items were also classified into general categories helpful for developing coping strategies and theoretical linkages. the study also examined how risk perceptions might differ across cultures, organizational settings, and economic conditions by conducting three simultaneous surveys in three different settings: hong kong, finland, and the united states. the panels consisted of very experienced project managers known to the researchers in each country. at the conference, the authors presented a comprehensive list of risk items formed by combining input from all three panels and ranked lists produced independently by each panel. they also discussed their analysis of the lists and contrasted their findings with previous research. they presented a meta-classification of risk items supporting further theoretical development. finally, they engaged the attendees in a discussion of the implications of their findings for further research and practice. an extended working paper with complete data, findings, and analysis was available at the conference in limited quantity. the authors were happy to supply a copy of the working paper by mail to anyone who requested it at or after the conference.