Smart Farm Extension Platform for Smallholder Potato Farmers via Mobile App

Thesis Abstract

The rapid digitization of agricultural advisory services in smallholder contexts has created a gap between information needs and farmer access, particularly for potato producers who confront variable weather, disease pressures, and market volatility. This study addresses the persistent information asymmetry faced by smallholder potato farmers by developing and evaluating a Smart Farm Extension Platform delivered through a mobile application designed to integrate real-time agronomic guidance, pest and disease alerts, input recommendations, yield forecasting, and market linkages. The aim is to enhance advisory effectiveness, farmer decision-making, and productivity through an ICT-driven extension solution. Specific objectives are (i) to design and pilot-test a mobile app-based extension platform tailored to smallholder potato farming systems; (ii) to assess adoption, usage patterns, and user satisfaction among potato farmers and extension agents; (iii) to evaluate the platform’s impact on agronomic practices, input efficiency, and yield outcomes; (iv) to identify socio-economic and contextual determinants of platform use; and (v) to generate policy and practice recommendations for scalable ICT-enabled extension. The study adopts a mixed-methods research design, combining a quasi-experimental component with qualitative explorations. The population comprises 1,200 smallholder potato farmers across three districts in a temperate-agroecological zone with established mobile connectivity. A multistage sampling approach selects 600 farmers for the quantitative arm, evenly split into intervention and control groups, and 24 extension agents for qualitative interviews. Data collection instruments include a structured survey assessing technology readiness, access to inputs, agronomic practices, yield history, and perceived usefulness; platform usage logs and agronomic records; semi-structured interviews with farmers and extension workers; and focus group discussions to capture contextual factors. Instrument validity and reliability are established through content validation by agronomy and ICT-for-agriculture experts, pilot testing with 60 farmers, and a Cronbach’s alpha threshold of 0.70 for multi-item scales. Data analysis employs (i) difference-in-differences regression to estimate the causal impact of the platform on yield, input efficiency, and adoption of recommended practices, controlling for covariates; (ii) multilevel modeling to account for clustering at village and district levels; (iii) propensity score matching to enhance comparability between intervention and control groups; (iv) thematic analysis of interview and focus group data to elucidate barriers, enablers, and user experiences; and (v) a cost-benefit analysis to assess economic viability. The study also tests theoretical propositions derived from Diffusion of Innovations (Rogers) and Technology Acceptance Model constructs (perceived usefulness, perceived ease of use) to explain adoption patterns. Expected findings indicate that the Smart Farm Extension Platform significantly improves the uptake of best management practices, timelier pest and disease interventions, precise input use (reduced fertilizer and pesticide overuse), and a measurable increase in potato yield and net income in the intervention group relative to controls over a 12-month cycle. Usage analyses are anticipated to reveal that frequent engagement with crop health alerts and decision-support recommendations correlates with higher practice adoption, moderated by factors such as household education, land tenure, and network reliability. Qualitative insights are expected to reveal critical success factors, including trusted local content, offline functionality with periodic synchronization, and user-centered interface design, as well as barriers such as digital literacy gaps and data costs. The study contributes to knowledge by empirically validating an ICT-enabled extension model that links real-time agronomic guidance with farmer decision-making, bridging the gap between research recommendations and on-farm practices. It extends theory by testing Diffusion of Innovations and TAM in a smallholder potato context and by integrating a cost-benefit perspective into ICT-enabled extension evaluation. The findings will inform policymakers, development agencies, and agricultural service providers on scalable, cost-effective, and context-sensitive approaches to ICT-driven extension. Recommendations emphasize strengthening digital literacy training, ensuring offline-capable modules, establishing public-private partnerships for sustainable data costs, and integrating the platform with existing extension systems to maximize reach and impact. The study concludes that a well-designed smart extension platform can enhance resilience and productivity of smallholder potato farming while contributing to broader food-security and rural-income objectives.