Drivers of farmers' adoption of organic coconut farming in the Mekong Delta of Vietnam: PLS-SEM with an ANN-based sensitivity analysis

This study investigates the psychological, social, policy-related, and farm-level drivers of farmers’ adoption of organic coconut farming in Vinh Long province, a major coconut-growing area in Vietnam’s Mekong Delta. Grounded in the extended Theory of Planned Behavior (TPB), the study incorporates descriptive norms, policy satisfaction, and farm size into the original TPB framework to explain both adoption intention and current adoption practices. Data were collected from 205 coconut-farming households across 18 communes and analyzed primarily using partial least squares structural equation modeling (PLS-SEM), with an artificial neural network (ANN) used as an exploratory sensitivity analysis to assess the relative importance of variables.. The PLS-SEM results show that attitude, subjective norms, perceived behavioral control, descriptive norms, and policy satisfaction are positively and significantly associated with farmers’ intention to adopt organic coconut farming. Intention shows the strongest contemporaneous association with current organic coconut adoption practices, while farm size and perceived behavioral control are also positively associated with current adoption practices. As an exploratory sensitivity check, an ANN-based importance analysis was also conducted; it showed partial agreement with the PLS-SEM ranking for adoption intention and complete agreement for the ranking of factors associated with current adoption practices. The multigroup analysis indicates broad stability of the structural relationships across conventional and transitioning farmers, with no statistically significant between-group differences in the examined paths.