AI Perception and Nanostore Performance: Moderating Effects of Institutional Heterogeneity in Honduras and Colombia (#2522)

Authors

Ortega-Jimenez, Cesar H

Paque Salazar, Ana Lucia

Melgar-Martínez, Narciso A

Ramirez Zarta, Jorge Bernardo

Abstract

This study investigates how relational capability, perceived quality, and perceived AI benefits influence operational performance in nanostores across Honduras and Colombia, addressing AI perception dynamics in resource-constrained Latin American retail contexts. The research is framed by an integrated theoretical model addressing limitations of traditional technology acceptance theories by incorporating Resource-Based View and Institutional Theory lenses, positioning relational capability as a networked resource contingent on institutional context. A cross-sectional dataset of 608 nanostores (204 Honduras, 404 Colombia) was analyzed using Multi-Group Structural Equation Modeling with Maximum Likelihood Estimation. Measurement validity was assessed through confirmatory factor analysis and invariance testing. Relational capability (β = 0.34, p < 0.001) and perceived AI benefits (β = 0.41, p < 0.001) significantly predict operational performance, while perceived quality exerts a moderate effect (β = 0.23, p < 0.01). Multi-group analysis reveals relational capability is more influential in Colombia, whereas perceived quality is more salient in Honduras. Contextual factors explain additional variance (R² = 0.42 Colombia, 0.36 Honduras). The cross-sectional design limits causal inference; common method variance was non-significant. Future research should incorporate longitudinal designs and objective metrics. Results suggest context-specific AI adoption strategies: Colombian nanostores benefit from ecosystem-driven approaches emphasizing supplier networks, while Honduran ones require quality-centered strategies focusing on service reliability. Policymakers should tailor digital transformation initiatives to local institutional realities and develop data privacy frameworks for low-resource settings. This study advances technology adoption theory by introducing relational capability as a critical networked resource in informal contexts and provides empirical evidence of cross-national heterogeneity in AI perception–performance relationships in informal retail.