Resumen

BENKRIZI, N. et al. Uncovering motivations for honey consumption in northwestern Algeria: A multivariate and CART-Based Modeling Approach. Agrocienc. Urug. [online]. 2025, vol.29, e1820.  Epub 01-Dic-2025. ISSN 2730-5066.  https://doi.org/10.31285/agro.29.1820.

Honey, as a culturally rooted and multifunctional food, provides a compelling case to explore how motivations, purchasing habits, and sensory preferences interact across diverse consumer groups. A questionnaire with thirteen categorical variables was administered to a sample of 303 individuals and then analyzed to examine honey consumption behaviors in relation to socio-demographic factors, purchasing patterns, sensory attributes, and underlying motivations. The sampling targeted the northwestern Algerian region. Rare categories were merged based on frequency, and bivariate associations were tested using Chi-square and Cramér’s V. Multiple Correspondence Analysis (MCA) extracted latent behavioral dimensions, which were then used in a Hierarchical Clustering on Principal Components (HCPC) to identify consumer profiles. Additionally, a Classification and Regression Tree (CART) model was applied to predict the primary motivation for honey consumption (nutritional vs. therapeutic). The MCA revealed two orthogonal axes, while HCPC identified three distinct clusters: (1) a health-conscious, tradition-oriented group, (2) a pragmatic segment, and (3) a price- and brand-sensitive group. CART achieved 67.3% overall accuracy, with the therapeutic segment yielding the best performance (F1-score = 0.697). These findings highlight a multidimensional structure of honey consumption, offering actionable insights for targeted marketing and nutrition-oriented interventions.

Palabras clave : CART; consumer driver prediction; hierarchical clustering; honey demand; MCA.

        · resumen en Español | Portugués     · texto en Inglés     · Inglés ( pdf )