Stephen J. Andriole

Toward the Development of an Integrated Crisis Warning System

Research directed toward the development of user-oriented integrated crisis warning systems should incorporate quantitative military, political, and economic indicators for crisis warning; indicators of U.S. military, political and economic interests abroad; and a unified multi-method forecasting capability into an interactive computer base. Initial movement toward this goal has occurred in the development and testing of a first-stage prototype crisis warning system. Tests indicate that the prototype would have contributed to predictions of the Soviet invasion of Czechoslovakia and the Indo-Pakistani war, and would enable analysts to track crisis and noncrisis events and conditions and to interact productively with a computer-based system. The second-stage prototype will continue to integrate, test, modify, and apply existing and ongoing research to the overall warning and management problem.

The use of events/interaction research by the intelligence community

This review essay applies academic and operational research community criteria of evaluation to a project founded on events/interaction data. The project, the Early Warning and Monitoring System (EWAMS), is built on solid academic research but also takes into account requirements of the operational community. Differences in form and content from typical academic events/interaction research have been instrumental in putting the EWAMS into operational use.

The levels of analysis problems and the study of foreign international, and global affairs: A review critique, and another final solution

Discussions about the “levels of analysis problem” have proliferated the field for almost two decades. In 1961, J. David Singer began the discussion in a now classic article. In response to Singer, analysts like Robert Isaak have attacked the Singer interpretation while many others have carelessly or obliviously skirted the issues surrounding the levels of analysis problem. Much of the confusion can be traced to a failure to view the importance of analytical levels in the aggregate analytical process and a failure to differentiate between causal and effectual levels of analysis. Causal levels of analysis are those whose variables or factors are antecedent to a particular behavior; effectual levels are those on which the behavior or phenomena in question actually occur, that is, on which the effect or result of the causal variables manifests itself. With reference to the aggregate analytical process, whenever scholars engage in inquiry they necessarily employ at least one causal and one effectual level of ...

Graphic Equivalence, Graphic Explanations, and Embedded Process Modeling for Enhanced User-System Interaction

Usersystem interaction research has traveled many paths over the past two decades. Human factors research now stands along side of cognitive modeling and artificial intelligence as promising new ventures into enhanced user-computer performance. Some notions developed in the cognitive modeling and artificial intelligence communities are presented, along with three techniques that we believe hold promise for enhanced user-system interaction. The techniques include 1) graphic equivalence, 2) graphic explanations, and 3) embedded process modeling. Graphic equivalence refers to the use of split or dual screen hardware configurations that permit users to interact with graphic and alphanumeric screens and create equivalence on the other. Graphic explanations are those that permit users to see graphically the analytical components of system-and user-generated problem solutions, while embedded process modeling refers to a technique designed to help users know where they have been in the problem-solving process, where they are going in the process, and what the whole process looks like. The domain for testing the techniques is Army tactical planning at the Corps level.

The process, outcomes, & impact of regime change in the third world, 1959–1981∗

(1986). The process, outcomes, & impact of regime change in the third world, 1959–1981. International Interactions: Vol. 12, No. 4, pp. 363-392.

The quiet revolution in the scientific study of international politics and foreign policy

The application of quantitative international political and foreign policy research to real decision-making and forecasting problems has been quietly progressing over the past ten years. However, this “quest for relevance” has not been targeted at the highest policy-making levels; rather the applications have occurred at the middle- and low-level policy support and analysis levels. Specifically, applied analysts have profiled crisis management problem areas, developed a computer-based early warning and monitoring system, developed and applied several Bayesian forecasting techniques, and developed a computer-based crisis management decision aid which have all been used to solve specific analytical problems within the U.S. Department of Defense. Nearly all of the applied research is interdisciplinary. All is problem-oriented, and a great deal is computer-based. In all likelihood the number of applications will increase as the complexity of international and foreign affairs increases. The national security decision- and policy-making analytical process will thus become more pluralized and, because of the computer-based nature of much of the research, more accelerated and concentrated. Thusfar the impact can be said to be positive; the future will enlarge or shrink the role of applied quantitative analysis in response to how beneficial such analyses actually are to real national security problems.

Decision process models and the needs of policy-makers: Thoughts on the foreign policy interface

Many political scientists are not at all convinced that empirically verifiable generalizations about political behavior can ever be produced. There are thus many who believe that political scientists can never exert a significant impact upon the public policy-making process. Perhaps one way of narrowing the gap between the researcher and the policy-maker is to examine the public (foreign) policy decision-making process, pinpoint the leverage points within the process where social and political scientists (who engage in foreign policy analysis) might help, and then list the kinds and methods of analysis that would contribute the most to the process. The work of Harold D. Lasswell, James E. Anderson, E. S. Quade, William D. Coplin, and Charles F. Hermann can help us define the (foreign) policy-making process and locate the points at which scientific analysis can help the (foreign) policy-maker. The matching of these leverage points with types and instances of “relevant” analyses, as well as the methodologies necessary to carry out the analyses results in a matrix of decision tasks and research tasks which may be viewed as an agenda for the conduct of policy relevant research.

High order corporate crisis management

Abstract Much research has shown corporation managers generally to be unprepared for such crisis as political upheavals, kidnapping, dramatic changes in foreign exchange transactions, etc. Three related approaches to collecting and structuring information which warns of possible impending crises, or helps in their management once erupted are described: INFER, OPINT and EVAL. They draw on the experience of the US Department of Defence and intelligence agencies.

Comparative forecasting: Hindsighting the Czechoslovakian invasion

Abstract Two dissimilar information-processing techniques may be used to forecast the same phenomenon. Event-pattern recognition, objective in nature, uses predetermined indicators and is dependent upon selected data flows; the Bayesian method rests more heavily upon subjective interpretation. These two methods may be used to forecast a variety of phenomena—here they are retrospectively tested for their ability to forecast the August 1968 Soviet invasion of Czechoslovakia. Objective and subjective methods may be used to generate combined, or at least cross-validated, forecasts; and the author suggests how—and at what costs—comparative forecasting may be used productively for international political analysis.

Knowledge, Action, and the Pursuit of Global Happiness

The recent proposal by Richard C. Snyder, Charles F. Hermann, and Harold D. Lasswell (1976) for the development and application of a global monitoring system (GMS) for the realization of human dignity is unique in several very important respects. In the following few pages I will try to confine my comments to this uniqueness and suggest why I believe that the Snyder, Hermann, and Lasswell proposal is long overdue, although very much in the formative stage.