Comparative Analysis of Agricultural Extension Methods on Smallholder Adoption Rates

Thesis Abstract

The adoption of modern agricultural practices among smallholder farmers remains uneven in the face of diverse extension delivery mechanisms, with significant implications for productivity, resilience, and rural livelihoods. This study investigates how contrasting agricultural extension methods influence smallholder adoption rates of sustainable practices, with a view to identifying effective mechanisms that accelerate uptake while remaining contextually appropriate. The aim is to compare traditional, participatory, digital, and integrated extension approaches to determine their relative impact on adoption velocity, breadth, and sustainability, and to unpack the mediating roles of farmer literacy, trust in extension agents, and perceived usefulness. Specific objectives are (i) to quantify adoption rates of selected sustainable practices under each extension method; (ii) to identify factors predicting adoption intensity among smallholders; (iii) to examine the mechanisms through which extension methods influence decision-making and learning; (iv) to compare the cost-effectiveness of the extension approaches; and (v) to develop a framework for tailoring extension services to farmer characteristics and local agro-ecologies. A mixed-methods design is employed, combining quantitative and qualitative strands to enable robust inference and contextual interpretation. The study is conducted in three agro-ecological zones with distinct farming systems, sampling 1,200 smallholder farmers stratified by gender, farm size, and access to information channels. A multi-stage sampling procedure ensures representation across districts implementing different extension modalities traditional top-down extension, participatory farmer field schools, digital advisory platforms, and an integrated approach that blends in-person and digital elements. Quantitative data are collected via structured interviews using a validated instrument capturing adoption status of target practices, exposure to extension, perceived benefits and barriers, socioeconomic variables, and farm performance indicators. Reliability of instruments is assessed with Cronbach’s alpha and test-retest methods. Qualitative data are gathered through 40 in-depth interviews and 12 focus group discussions to explore experiential factors, trust dynamics, and learning processes. Data analysis proceeds in two parallel streams. Quantitatively, hierarchical logistic regression models estimate the likelihood of adopting each practice as a function of extension method, controlling for covariates, with propensity score matching used to address potential selection bias. Structural equation modeling (SEM) tests the hypothesized pathways linking extension modality to adoption through mediators such as perceived usefulness, knowledge gain, peer influence, and extension agent credibility. Cost-effectiveness analysis compares per-adoption and per-harvest impacts across modalities. Qualitatively, thematic analysis identifies recurrent themes regarding learning experiences, engagement quality, and constraints, using a framework grounded in the Theory of Planned Behavior and Social Cognitive Theory, and triangulates findings with quantitative results to illuminate causal mechanisms. Theories employed include the Diffusion of Innovations, Technology Acceptance Model, and Social Learning Theory to interpret adoption dynamics and channel-specific effects. Expected findings indicate that participatory and integrated extension approaches yield higher adoption rates and more rapid diffusion of practices than traditional methods, with digital platforms augmenting reach and efficiency but requiring supportive in-person components for trust-building and hands-on demonstration. The integrated approach is anticipated to be the most cost-effective over a five-year horizon, balancing scalability with experiential learning. Key determinants of adoption are expected to include perceived usefulness, compatibility with farm routines, extension agent credibility, and access to timely information, with gender and farm size moderating effects. The study would contribute to knowledge by delineating how extension design shapes adoption trajectories, offering a comparative cost-benefit perspective, and proposing a framework for optimizing extension portfolios in diverse settings. The conclusion emphasizes that no single extension method suffices; rather, a hybrid model tailored to local contexts and farmer profiles optimally enhances adoption. Recommendations include prioritizing participatory co-design of extension content, strengthening trust-building and demonstration components within digital platforms, investing in agent training and gender-responsive outreach, and implementing monitoring systems that link extension activities to measurable adoption and productivity outcomes.