Technological priorities in rice production among smallholder farmers in Ghana. (1st December 2017)

• Record Type: Journal Article
• Authors: Tsinigo, Edward; Behrman, Jere R.
• Is Part Of: Netherlands journal of agricultural science. Volume 83(2017)
• Page Start: 47
• Page End: 56
• Publication Date: 2017-12-01
• Subjects: Multinomial probit model -- Technology adoption -- Productivity-enhancing technologies -- Simultaneity -- Interdependencies; Agriculture -- Periodicals; Agriculture -- Research -- Periodicals; 630
• DOI: 10.1016/j.njas.2017.07.004
• Languages: English
• ISSNs: 1573-5214

Abstract

Highlights: Rice farmers were more likely to adopt fertilizer only after adopting the improved rice variety but not vice versa. Adopters of the improved rice varieties were more likely to use inputs separately or simultaneously adopt all technologies. Education and complementary input availability consistently predicted the adoption of five technology bundles. Extension contacts and farm size consistently and significantly predicted the adoption of all technology bundles. Complementary input availability and participation in field demonstrations were the strongest predictors of adoption. Abstract: Technological innovations in agriculture have transformed the farming systems of smallholder farmers, leading to the realization of economic incentives of higher outputs, profits, and sustainability. However, attempts to model their adoption behaviors have failed to consider the possible interdependencies and simultaneities in the adoption decision process. This may cause invalid inferences and incorrect conclusions to be made regarding smallholder farmers' adoption decisions. This study, therefore, models smallholder farmers' adoption decisions while taking into account the possibility of interdependencies and simultaneities of adopting rice productivity-enhancing agricultural technologies in Ghana. Cross-sectional survey data on 1016 smallholder rice farmers sampled using a three-stage sampling method from four municipalities in two regions in Ghana were used for the study. We fit a multinomial probit model to analyze the decisions to adopt or not to adopt varied combinations of improved rice variety, fertilizer, and herbicide. Our study finds that when faced with improved rice variety-fertilizer-herbicide technological options, smallholder rice farmers' adoption decisions fell into any one of four possibilities: (a) reject all technological options, (b) adopt a particular technology only, (c) adopt two of the technologies, and (d) adopt all technological choices. We show that such adoption decisions were consistently and most significantly predicted by farm size, extension contacts, education, participation in field demonstrations, complementary input availability, and extent of commercialization. Modeling smallholder farmers' adoption decisions to capture interdependencies and simultaneities provides useful information for integrating diverse perspectives into the design and promotion of technological innovations based on their nature of adoption, the specific technological innovations, and influencing factors. We recommend that to achieve the maximum benefits, future technology promotion activities should concurrently reduce farmers' subjective uncertainties and risks, promote credit access and subsidies, as well as ensure effective input delivery systems while targeting specific technological innovations.