Aragao Pereira, Mariana de2011-09-212011https://hdl.handle.net/10182/3866This study draws on social-psychology in an attempt to identify the various motivations for technology adoption (TA), including both economic and non-economic, and to gain insights into how and why Brazilian innovative beef farmers make decisions about whether or not to adopt particular technologies. Three major research questions are addressed: (1) is there diversity of major goals and values amongst Brazilian innovative beef farmers, and if so, how can this diversity be characterised?; (2) how does diversity within innovative beef farmers’ goals and values affect adoption and non-adoption of technologies?; and (3) do innovative beef farmers use a different set of constructs when assessing different types of technologies? If so, why? Innovative farmers were targeted given their openness to new ideas, including innovations, and their social role in importing innovations from institutions onto farms. Innovative farmers from Mato Grosso do Sul State (MS), Brazil, were purposively selected based on their self-enrolment in the Good Agricultural Practices Programme (BrazilianGAP) or in the Association of Producers of Young Steers (APYS), which are initiatives that promote good farming practices among beef farmers. The 15 farmers enrolled in BrazilianGAP who ran commercial family farms in MS were selected and six who agreed were interviewed. From APYS’s 120 members, 30 cases were selected through a stratified random sampling based on herd size (small, medium and large). Some 21APYS’ members were interviewed, resulting in 20 valid interviews. Using a constructivist-interpretivist philosophy and a case study strategy, investigations of 26 farmers about their goals, farming systems, and the rationale as to why specific technologies were or were not applied, were undertaken through semi-structured in-depth interviews, which were conducted on their farms. This study employed Ethnographic Decision Tree Modelling, Q-Methodology and Personal Construct Theory, and elements of Soft Systems Thinking and Grounded Theory. Four main sets of goals and values were identified amongst the farmers through the sorting of 49 statements (Q-methodology), and were labelled the Professional Farmer (PF), the Committed Environmentalist (CE), the Profit Maximiser (PM) and the Aspirant Top Farmer (ATF). The PF aimed at running the farm in a professional way, based on sound technical and managerial practices. The CE put emphasis on the long-term sustainability of his farming system. The PM focused on technical issues to pursue his economic and lifestyle objectives. The ATF was seeking excellence and sought recognition for this. Analyses of the aggregate adoption rates of these farmer types showed that, on average, they adopted 27 (60%) of the 45 technologies analysed, with production and managerial technologies having higher levels of adoption relative to environmental technologies. The levels of technology adoption found in this study were considerably higher than those of average Brazilian farmers, as the Brazilian Agricultural Census show. This confirmed the innovative character of the interviewed farmers and validated the sampling strategy to identify such farmers. Although no relationship between the farmer types and the use of individual technologies can be claimed, results suggested that the farmers’ goals (i.e., represented by the farmer types) tended to generally orientate technology adoption. Farmer types who were production-oriented (PF, PM and ATF) adopted more production technologies than the environmentally-driven type (CE). This CE type, in turn, had the highest adoption rates of environmental technologies of all farmer types. Although important for adoption behaviour, the farmers’ goals were insufficient by themselves to determine their technology adoption behaviours, with multiple influencing additional factors identified. Among these factors were the five technology attributes proposed by Rogers’s (2003) adoption of innovations theory: compatibility, complexity, relative advantages, observability and trialability. Compatibility and the relative advantages of technologies were the most important attributes while observability and trialability were relevant, but of secondary importance. Complexity seemed to be considered alongside other aspects of technologies (e.g., cash returns) that define their relative advantages, rather than an attribute in itself. This study, therefore, expands Rogers’ (ibid) propositions by identifying a hierarchy among the technology attributes. Ethnographic decision tree models on a dry season supplementation for rearing cattle (‘hard’ production technology) and on beef cost analysis (‘soft’ managerial technology) showed that farmers construed these technologies differently, using multiple criteria both economic and non-economic. They also demonstrated that both adoption and non-adoption resulted from elaborate decision processes and were rational given the farmers’ understanding of these technologies and their current resource set. Both adoption and non-adoption occurred for diverse reasons. Reasons for non-adoption included the technology incompatibility with the farmers’ goals and values or with their farming systems, constraints to adoption or because the technology was perceived as less advantageous than other alternatives. These findings contribute to decision making and technology adoption theories, drawing attention to the need of a ‘farmer-centric’ approach in the development and diffusion of technologies. Under a ‘farmer-centric’ approach, it is acknowledged that farmers are unique, have diverse goals and farming systems and these impact on how they perceive technologies. It is argued, therefore, that by better understanding the decision frameworks of these innovative farmers, research institutions can design more effective research and extension strategies.ensocial psychologyQ methodologyfarmer's goalsadoption of innovationagricultural researchbeef cattledecision makinggrounded theorysoft systems thinkingtechnology adoptiontechnology transferethnographic decision tree modelingThesisQ112885557