Information and the speed of innovation adoption

Within a Bayesian framework, a random-effects model is developed and applied to adoption of new wheat varieties in South Australia. In this model, not all pieces of information add equally to knowledge about the- innovation. The model shows the acquisition of information to be much slower than has been suggested by previous Bayesian models and can also explain laggards and partial adoption. The results have important practical implications for farmers and support agencies. The paper's theoretical contributions are to highlight the structure of information, and to demonstrate how qualitative results can be obtained where the posterior Bayesian distribution is intractable. Key words: adoption, Bayesian, effective information, innovation, random-effects model. One of the recently developed models of the adoption process for agricultural and other in-novations is the Bayesian learning model. An attractive feature of this type of model is that it can account for the empirically observed lags between the time when farmers first hear about an innovation and the time when they adopt it. These lags occur because farmers take time to accumulate the information they need to decide whether to adopt an innovation. Such models feature a mathematical analogue of the learning process which assumes that farmers adjust ini-tial beliefs in the light of new information in accordance with Bayes's theorem. Models of this kind have been used by O'Mara (1971