A Farmer-Cocalibrated Extension Interaction Model for Adoption

Thesis Abstract

This study investigates how farmer-cocalibrated extension interactions influence adoption of agricultural innovations, addressing persistent gaps between extension messaging and farmer realities that constrain uptake in smallholder systems. The problem stems from misalignment between top-down extension advisories and farmers’ experiential knowledge, resource constraints, and local risk perceptions, which undermines timely and sustained adoption of improved practices. The aim is to develop and validate a Farmer-Cocalibrated Extension Interaction Model (FCEIM) that integrates farmer agency, co-design of extension content, and iterative feedback into the adoption process. Specific objectives include (1) identifying determinants of farmer-initiated co-calibration behaviors in extension encounters, (2) assessing the relative influence of co-calibrated interactions on adoption intention and actual uptake of improved agronomic practices, (3) examining how contextual factors (land tenure, access to credit, market signals) modulate participation in co-calibration processes, and (4) articulating a parsimonious theoretical model linking interaction quality, farmer empowerment, and adoption outcomes. Methodologically, the study adopts a sequential explanatory mixed-methods design conducted in a purposively selected agro-ecological zone with smallholder farmers (n=420) and extension personnel (n=28) across 14 communities. The quantitative phase employs a structured survey to measure constructs such as interaction quality, perceived co-calibration, adoption intention, and actual adoption rates of two benchmark innovations (improved seed varieties and precision fertilizer application) over two cropping seasons. Data will be analyzed using structural equation modeling (SEM) to test the hypothesized paths among interaction quality, farmer empowerment, and adoption outcomes, complemented by multilevel logistic regression to account for clustering at village level. The qualitative phase uses semi-structured interviews and focus group discussions with farmers and extension agents (n=60 interviews, 6 focus groups) to elucidate mechanisms of co-calibration, perceived benefits and constraints, and context-specific nuances. Thematic analysis will be conducted with a framework approach, enabling triangulation with quantitative findings. Instrument validity will be established through content validity indices and pilot testing, with reliability assessed via Cronbach’s alpha and composite reliability. The expected findings indicate that higher-quality, farmer-influenced extension interactions—characterized by collaborative content development, iterative feedback, joint problem-solving, and culturally congruent communication—significantly enhance adoption intention and actual uptake, mediated by farmer empowerment and perceived relevance. Variations are anticipated across gender, literacy, and asset endowment, with more pronounced effects in communities exhibiting stronger social networks and higher trust in extension personnel. The model is expected to reveal that co-calibrated interactions moderate the relationship between perceived risk and adoption decisions, reducing information asymmetry and improving resource mobilization for practice change. Contributions to knowledge include (i) the formalization of a Farmer-Cocalibrated Extension Interaction Model (FCEIM) that integrates farmer agency into extension theory and practice, (ii) empirical evidence on the mechanisms by which co-calibration processes influence adoption dynamics, (iii) methodological advancement through a robust mixed-methods protocol suitable for testing interaction-focused theories in extension contexts, and (iv) actionable guidance for policy and program design to institutionalize farmer-driven co-design and feedback loops within extension systems. The study also expands theoretical understanding by integrating social capital and empowerment theories with diffusion of innovations and interaction frameworks, providing a coherent explanation for variances in adoption across heterogeneous farming communities. The main conclusion anticipated is that embedding farmer-led co-calibration within extension interactions markedly improves the relevance and timeliness of advisory services and accelerates the adoption of improved agricultural practices. Recommendations include establishing formal co-design platforms for farmer–extension dialogues, capacity-building initiatives to strengthen farmers’ problem-definition and feedback skills, and policy measures that incentivize adaptive extension models, continuous monitoring of co-calibration processes, and scalable digital tools to document and analyze interaction outcomes. Further research directions include comparative cross-regional validation of FCEIM, exploration of long-term sustainability of adopted practices, and integration of cost–benefit analyses to quantify economic impacts of co-calibrated extension interventions.