Influence at work: A 25-year program of research

Organization of any kind, from prehistoric hunting societies to companies w orking through the worldwide web, operate with a distribution of influence and power among their members. This distribution of influence has consequen ces at three levels: for the people working in the organization, for the or ganization itself, and, from time to time, for members of society outside t he organization. A series of action- and policy-oriented projects on the di stribution of influence were developed by or in collaboration with the Cent re for Decision Making Studies of The Tavistock Institute over a quarter of a century. They started with a seven-country comparative research on top m anagement decision making, followed by two 12-country studies on Industrial Democracy and a 5-year longitudinal program in seven companies in three co untries. These and two longitudinal projects in Britian, one on a motor car manufacturer and the other on an airport, used a similar conceptual framew ork. The article draws on the evidence from this program of work, describes the evolving theoretical model and concludes that organizational influence sharing appears to have made only limited progress during the last 50 year s. Four explanations are put forward: overidealistic expectations; a tenden cy to ignore the need for certain necessary antecedents, like competence; a tendency to act as if influence sharing is not subject to contingencies li ke the nature of tasks; and probably most importantly, the almost universal tendency to design influence sharing measures through uncoordinated mechan istic social engineering.