How Traceability Information and Live-Stream Credibility Shape Agricultural Product Purchase Intention: An Online Experiment and SEM Analysis

Live-streaming commerce has emerged as an important channel for agricultural products, yet the unverifiable nature of quality attributes viewed only through a screen leaves consumers suspended between trust and doubt. Grounded in the Stimulus-Organism-Response framework and signal theory, this study examines how traceability information (a product signal) and streamer credibility (a source signal) jointly shape online purchase intention through the parallel mediators of product trust and perceived risk. A two-study design was adopted. Study 1, a 2×2 between-subjects online experiment (N = 163), established causal evidence for the main and interaction effects. Study 2, a field survey of experienced agricultural live-streaming consumers (N = 363), was analyzed via PLS-SEM to assess the mediation structure, explanatory power, and predictive relevance. Both signals exerted significant positive effects on purchase intention, and their interaction was nonadditive and reinforcing rather than compensatory. Product trust and perceived risk each mediated the two signal-to-intention paths, with product trust consistently carrying the larger indirect effect across both studies. The model explained 48.7 percent of the variance in purchase intention, with positive out-of-sample predictive relevance. Traceability information and streamer credibility function as complementary rather than substitutable signals, operating primarily through product trust. These findings offer practical guidance for platform design, streamer certification, and traceability-system development in agricultural live commerce.