Scenario-Based Evaluation of Farmers’ Acceptance of Crop-Switching Policy in Kurdistan Province: Integrating Structural Equation Modeling, Bayesian Belief Networks, and Agent-Based Modeling

Abstract

This study aimed to develop an integrated framework for evaluating the acceptance of crop-switching policy under conditions of uncertainty. The study employed an exploratory sequential mixed-methods design. In the qualitative phase, data obtained from semi-structured interviews with 20 farmers and agricultural experts were analyzed using thematic analysis. In the quantitative phase, data collected from 300 farmers across four counties in Kurdistan Province were analyzed using partial least squares structural equation modeling (PLS-SEM). Subsequently, the validated constructs and relationships were modeled within a Bayesian belief network, and six policy scenarios were evaluated alongside the baseline scenario. The network’s probabilistic outputs were then transferred to an agent-based model to simulate the crop-switching adoption rate at the level of individual farmer agents. The qualitative findings identified seven major themes: economics and markets, risk and decision-making, social influence and trust, policy and institutional support, technical and extension capacity, environment and health, and livelihoods and crop-switching behavior. The structural model results indicated that economic advantage, market security, institutional trust, social influence, and technical capacity positively affected attitudes, intentions, or crop-switching behavior, depending on the specified model pathways. In contrast, perceived costs and risks and environmental and health consequences exerted inhibitory effects. The coefficient of determination was 0.644 for attitudes and 0.716 for intentions and behavior, while the standardized root mean square residual was 0.049. The Bayesian belief network results showed that the probability of a high intention to switch crops increased from 25.87% under the baseline scenario to 54.23% under the favorable scenario, whereas it decreased to 11.01% under the unfavorable scenario. Furthermore, the agent-based model showed that the crop-switching adoption rate increased from 53% under the baseline scenario to 74% under the favorable scenario and declined to 31.67% under the unfavorable scenario. Overall, the findings indicate that the success of crop-switching policy depends on designing and implementing a coordinated policy package encompassing economic profitability, market security, risk reduction, institutional support, practical training, and trust-building.