THE FLOW OF INFORMATION THROUGH SOCIAL NETWORKS - DIAGONAL-FREE MEASURES OF INEFFICIENCY AND THE STRUCTURAL DETERMINANTS OF INEFFICIENCY

The aim of this paper is to (1) propose specific diagonal-free measure s of the inefficiency of information flow through social networks and (2) present an analysis of the structural determinants of inefficiency . We employ a formal model of collective action developed by Coleman ( 1973, 1990); link it to the Markov model of interpersonal influence de veloped by among others French (1956), Harary (1959), and Friedkin (19 91; Friedkin and Johnsen 1990); and extend it to measure inefficiency in information flow based on the concept of mean first passage times. It is shown that structural characteristics of social networks - such as global density, local density, diameter, number of bridges, and ine quality in centrality, and certain interactions among these variables - strongly affect inefficiency. Theoretical implications of the result s are discussed.