Most diffusion models have assumed an unstructured pattern to the spread of information, disease, or elements in other substantive contexts. The population dynamics of disease spread only by close personal contact is an important example of a diffusion process for which the assumption that diffusion is unstructured, that the population mixes randomly, is inappropriate. While biological factors regulate the transmission and natural history of the infectious agent, social factors, which have been typically neglected in the field of epidemiological modeling, regulate the pattern of differential association, or social mixing, and thus the structure within which transmission is channelled. This research planning project seeks to develop a plan for producing a general modeling framework that will enable analysis of selective mixing in diffusion processes. Steps leading to this research include the development of a solid collaborative working agenda with potential colleagues overseas, a period of intensive collaboration in order to broaden the sociological focus of the work, and analysis of extant data from several sources on patterns of selective mixing and on patterns of diffusion in disease and other empirical contexts. This project will enhance capability for modeling diffusion processes by developing a plan to examine how the structure of social networks affects the nature and velocity of diffusion. It will also facilitate significant collaboration between the fields of sociology and biology.
