DUCES, BRIGITTE; PHD

                         UNIVERSITY OF MARYLAND COLLEGE PARK, 1985

                         INFORMATION SCIENCE (0723)
 

                         The purpose of this dissertation is to study the role of information in decisionmaking as exemplified in the
                         adoption of agricultural innovations. The specific objectives are (1) To test and refine a descriptive model
                         of the decision process in the adoption of innovations based on Rogers and Shoemaker (1971). (2) To
                         test and refine a typology of agricultural innovations. Data were collected through 24 case studies in
                         different counties across the U.S. A case study consisted of on-site visits of one week with interviews of
                         6-7 innovators, 2-3 non-innovators, and 2-10 change agents. Data were analyzed through two
                         complementary methods: statistical tests of hypothesis and content analysis of the interview
                         descriptions. The statistical analysis saw no conclusive trends emerging; the innovations studied are too
                         dissimilar in nature and the phenomena behave differently depending on the cases. The content
                         analysis identified relevant variables and lead to a typology of innovations based on their difficulty, as
                         determined by the degree of consonance between the innovator's values and those of the innovation,
                         together with the degree of continuity between the innovator's previous practices and those required by
                         the innovation. A statistical analysis that controls for type produces better results: Innovations with a high
                         degree of consonance of value are adopted without extensive amounts of information. Innovations with
                         both high dissonance and discontinuity are adopted only when forced (as by law) or after processing
                         extensive amounts of information, particularly from trusted sources. These results suggest that the
                         nature of the decision problem is of paramount importance in analyzing the decision process and that the
                         search for all-encompassing general patterns may not lead very far. The practical implications of these
                         results are that dissonant innovations require a well-orchestrated information effort on the part of the
                         change agents. On the other hand, highly consonant innovations require minimal effort in information
                         provision.

Social Systems Simulation Group P.O. Box 6904 San Diego, CA  92166-0904 Roland Werner, Principal Phone/FAX  (619) 660-1603