Dan Myers: Current Research Projects


My principle research interests lie in the area of collective behavior and social movements. More specifically, my current work focuses on rioting, diffusion models, and formal/mathematical models of collective action.

Details on Current Projects:

Diffusion Related
Riot Related
Media Bias in the Coverage of Protests, Demonstrations, and Collective Violence

Measuring Individual Activist Orientation

Game Theory

Methodological Issues


Diffusion Related

My work on riot diffusion examines the wave of racial rioting in the U.S. that began in the 1960s and continued thorugh the early 1970s. Work done to this point documents that diffusion/contagion was a powerful force during the wave and that there is great variability in contagion between riots and cities depending on a number of factors. Specifically, the contagious influence from a riot decays over distance and over time. Furthermore, the infectiousness of a riot is keyed strongly to its severity. These findings are dcumented in a working paper, "Collective Violence and Heterogeneous Diffusion Processes: U.S. Racial Rioting from 1964 to 1971." A second component of the project further examines what makes some cities more susceptible to contagious influence and makes their riots more contagious. It also examines mass media distribution as a mechanism for the transmission of the information that makes a riot contagious. A second paper, "The Diffusion of Collective Violence: Infectiousness, Susceptibility, and Mass Media Networks" reports these results.

In a separate line of research, I have developed a new deterministic diffusion model for collective violence in the tradition of classical diffusion analyses. This model is called the "Opposing Forces Diffusion Model," and provides an advances over previous model because it treats provocation and repression as two seperately diffusion phenomena. The model is introduced in "The Opposing Forces Diffusion Model: A Theory of Collective Violence Diffusion," and its empirical advantages over previous models are documented more fully using empirical and simulated data in "A Comparative Test of Three Diffusion Models for Collective Violence" and "Systematic Testing of Deterministic Diffusion Models."

Riot Related

Beyond diffusion analysis, my riot research is also concerned with structural and economic conditions that contribute to riot propensity. One paper recently published in the American Sociological Review, ("Racial Rioting in the 1960s: An Event History Analysis of Local Conditions") finds support for economic competition models. My most recent riot research (conducted with advance graduate student Mike Davern) examines the relevance of local economic conditions for campus race riots (as opposed to the urban "ghetto" riots). This research is also examining diffusion influences between campuses and urban areas. Another graduate student, Mike Gibbons, is working with me on a project examining weather patterns and rioting. This project is producing interesting new results regarding the effects of temperature, humidity, and region on riot rates.

Information on the Ideological Foundations project will be posted soon.

Media Bias in the Coverage of Protests, Demonstrations, and Collective Violence

Along with Pam Oliver and other collaborators at the University of Wisconsin, I continue to work on a large project tracking the media coverage of protests and demonstrations. This is a long-term project that has involved collecting traces of protest through multiple sources including mulitple police agencies, other governmental bodies, newspapers, and television news. The data gathering process was a massive undertaking and the results of this study are just now beginning to be compiled. A paper examining all events and sources for Madison, WI for 1994 reports the first set of results from this project ("Media Coverage of Political and Nonpolitical Public Events").

Measuring Individual Activist Orientation

This project is concerned with identifying the sources and correlates of "activist orientation." In other words, who thinks of themself as an activist, why, and what impact does this have on their behavior. The first step in the process was to produce a valid and reliable measure of activist orientation that we call the Activist Orientation Scale. The development of this scale, undertaken with psychologist Alexandra Corning, is documented in "Activist Orientation and its Relationship to Relative Deprivation, Locus of Control, and Activist Behavior." Studies of the predictors of activist orientation using this scale are currently under way.

Game Theory

N-person game theory is a mathematical method of examining competition and cooperation between multiple actors. Using these games, we can examine underlying dimensions of important social processes such as conflict resolution, the provision of public goods, decision-making in risky situations, and the formation of coalitions of actors in competitive situations. My research in this area is currently focused on the testing and refinement of a new theory or solution concept called the Central Union Theory. This solution concept represents an advance over others in it class for a number of reasons but most importantly because it make specific probabilistic predictions about which coaltions will form. Two papers written with H. Andrew Michener are in press and provide initial competitive tests demonstrating the effectiveness of the theory ("A Test of Probabilistic Coalition Structure Theories in 3-Person Sidepayment Games" in Theory and Decision and "Probabilistic Coalition Structure Theories: A Test in 4-Person Superadditive Sidepayment Games" in the Journal of Conflict Resolution. This fall semester, graduate student Mike Gibbons and I will be collecting a new data set for a study of the CU Theory focusing specifically on the coalition formation issue.

Methodological Issues

As research progresses, new methodological issues inevitably arise. Three papers are currently under way. The first demonstrates a series of techniques to thoroughly examine and test deterministic diffusion models. The second examines number of goodness-of-fit indexes used to evaluate the predictions of game theory solution concepts. These indexes incorporate both the accuracy of predictions about payoffs to players and predictions about coalition structure formation. The third discusses the use of log-linear event history models with historical data pointing out the special considerations which must be invoked with the sort of survival data historians are most likely to encounter.