Search

This paper examines research on the diffusion of innovation, the final stage in the process of technological change. The focus rests primarily on two traditions in diffusion research: that of economists and that of sociologists. Diffusion researchers in these and related disciplines have made significant contributions to the understanding of the dynamics of processes of change; yet the state of the art in diffusion research is not equal to the sum of its parts. This is due in large measure to disciplinary parochialism: scholars have concentrated on those innovations, diffusion environments, explanatory variables, and analytical methodologies which are most compatible with their particular disciplines, despite the fact that diffusion is not a discipline-specific phenomenon. Deficiencies in current understanding of diffusion are examined in the context of this and other significant problems. The paper concludes by considering the policy relevance of diffusion research and suggesting issues with which future research might productively be concerned. [ABSTRACT FROM AUTHOR]

The rampant acceleration of scientific advance and technological change that seems to be required for national viability unfortunately entails a disconcerting human consequence: an explosive increase in cultural complexity bearing ominous possibilities of massive social disruption. Attainment of an adequate rate of cultural adaptation-as an idealized response to this situation-can be predicated only on the basis of deeper comprehension and creative modification of the social-institutional decisionmaking processes that comprise the cognitive modus operandi of civilized society. Recent methodological developments in management science are presented as promising theoretical resources for extending the conventional paradigm of objective rational analysis to incorporate valuative aspects of policy-level decision problems. Initially the emphasis is placed on an extension of structured rationality which constitutes one of the core-advances contributing toward emergence of the policy sciences as a legitimate supradiscipline. With its secondary emphasis on the concept "institutional self-organization," this paper represents an attempt to (1) capitalize early on a particular feature of optimal organization and (2) bring this feature immediately to bear as a consideration in institutional systems design. The concept of a "national administrative research agency" is put forward as an institutional prototype embodying the innovative organizational format needed in order to connect theoretical and practical aspects of social problem solving. The significance of this prototype lies in its implication for achievement of a deliberately self-transforming society, a purposefully adaptive version of the social order. [ABSTRACT FROM AUTHOR]

The objectives of this paper are to understand what is meant by "better" policymaking and "more efficient" technology transfer, to explore what is needed for their achievement, and to suggest an operational mechanism for improving the two processes. The author introduces a few new terms: (1) "Inter-context information" is defined, and its importance in decisionmaking, policymaking, technology transfer and education is pointed out; (2) a distinction is drawn between "incidental technology transfer"-initiated by the donor-and "organized technology transfer"-initiated by the recipient. The author suggests that National Thinking Laboratories should be established to promote organized technology transfer and to act as catalysts to organized policymaking. Their charter should be to match needs in one context to capabilities in another context. This charter is outlined in operational terms by five general objectives listed by the author. The National Thinking Laboratories are most urgently needed, particularly in the developing countries. [ABSTRACT FROM AUTHOR]

In this article, the authors have considered some simple models for the diffusion of innovations, using as an illustration the adoption of the comprehensive school reform by Norwegian communes. The first aim was to present a kind of catalogue of models showing for which technical reasons some models tend to appear rather often and which could serve to lead, if necessary, to more complex models. The second aim was to define and limit for each model the kind of explanation it suggested. Two possible models, which have been most frequently proposed in the literature, will be emphasized. Descriptive models serve first to summarize data. The shape of the diffusion curve when fitted to a function which has a simple analytical form as well as few parameters is more easily remembered than if it had a more precise empirical description. The exact analytical form is not so important because different functions can serve the purpose of fitting the data reasonably well. Usually, analytical formulas are chosen according to the author's familiarity with them. Probabilistic models allow for more refined hypotheses than deterministic models, and, in principle, they also allow for a better judgement about the quality of fit. This, however, occurs only in the framework of a number of assumptions where control might be difficult. Probabilistic models generate a distribution of diffusion curves: the observed one being one among numerous possibilities.

The article aims to determine how much education--that is, formal schooling--employed workers had in 1950 and 1960, and to relate such data to technological change, as measured by average annual changes in output per worker during the 1950's. Authors emphasize that their concern is with the amount of schooling or training needed for the job rather than with the level of education demanded by employers or personnel men when hiring workers. Information on average annual changes in output per worker during the 1950's, for each of over 60 industries comprising the entire United States labor force, was available from an earlier study of authors. Data on the educational level of the workers were obtained from one-in-a-thousand sample tapes of the 1950 and 1960 United States population censuses. The labor force becomes progressively more educated by the simple fact that youngsters who enter have had more schooling than the older workers, especially those who retire. This can be seen in the accompanying tabulation.