Understanding farmers’ participation in agricultural marketing networks through an exploratory sequential mixed-methods study integrating qualitative barrier identification and structural equation modeling

Agricultural marketing networks such as cooperatives are widely promoted as mechanisms to improve farmers’ market access and income; however, adoption rates remain low in many developing contexts, including Iran. Despite extensive literature based on the Theory of Planned Behavior (TPB), existing studies often fail to fully capture context-specific institutional and social barriers, and rarely integrate qualitative insights into quantitative model development. This study addresses these gaps by employing an exploratory sequential mixed-methods design to develop and test an integrated behavioral model of farmers’ participation in agricultural marketing networks in Fars Province, Iran. In the qualitative phase, semi-structured interviews with 32 farmers were conducted to identify context-specific barriers and underlying mechanisms shaping participation decisions. Thematic analysis revealed four main categories of barriers: economic, institutional, social-informational, and psychological. These findings were systematically transformed into measurement constructs and used to develop a structured questionnaire for the quantitative phase. In the second phase, data from 332 randomly selected farmers were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results show that attitude (β = 0.30, p 0.05) and knowledge (β = 0.01, p > 0.05) do not. Trust significantly shapes farmers’ attitudes (β = 0.60, p < 0.001) and also has a weaker direct effect on intention (β = 0.09, p < 0.05). Importantly, the composite perceived barriers construct exerts a significant negative effect on intention (β = − 0.06, p < 0.05), reflecting the structural constraints identified in the qualitative phase. The study contributes to the literature by demonstrating how qualitative insights can be systematically translated into measurable constructs within a TPB-based framework, thereby improving contextual validity and explanatory power. It further distinguishes between different forms of social influence, showing that behavioral effects are driven more by observed peer participation through social networks rather than perceived normative pressure. These findings provide important implications for designing trust-based, socially embedded agricultural policies that extend beyond information-driven interventions.