Citation

Krassa MA. Soc. Netw. 1988; 10(2): 109-136.

Copyright

(Copyright © 1988, Elsevier Publishing)

DOI

10.1016/0378-8733(88)90018-4

PMID

unavailable

Abstract

The paper describes the results of computer simulations of behavioral contagion in the realm of public opinion. Using threshold models to describe how and why some people assert opinions and others do not, the paper argues that to better describe the dynamics of opinion assertion and quiescence the model must be able to account for the fact that different individuals differ in the importance they place on the actions of others, and in the networks in which they are embedded. The paper reports on simulations incorporating these aspects into the model, yielding an improved representation of public opinion dynamics. Further, the incorporation of networks and weighting systems in the model alleviates one of the problematic features of the simple threshold models described by Granovetter, namely that certain individuals play unrealistically key roles in the contagion process. When networks are accounted for, the threshold formulation is less likely to make the group dynamic so dependent upon the behaviors of specific individuals.