Characterization of Goat Production Systems in the Northern Dry Forest of Peru Using a Multivariate Analysis

Goat production in the dry forest of northern Peru is essential for rural livelihoods but remains poorly characterized regarding its productivity and sustainability. This study used multivariate techniques-a multiple correspondence analysis (MCA), principal component analysis (PCA), factor analysis of mixed data (FAMD), and hierarchical cluster analysis (HCA)-to analyze data from 284 producers in Tumbes, Piura, and Lambayeque. Surveys captured 48 variables (41 qualitative, seven quantitative) on productivity, socioeconomics, and management. The MCA explained 22.07% of the variability in two dimensions, while the PCA accounted for 63.9%, focusing on productivity and diversification. The FAMD integrated these variables, explaining 51.12% of variability across five dimensions, emphasizing socioeconomic and management differences. The HCA identified three clusters: cluster 1 featured intensive systems with advanced management and commercial focus, cluster 2 included extensive systems limited by water scarcity, and cluster 3 reflected semi-intensive systems with irrigation and diversified production. These findings provide a detailed understanding of goat systems in northern Peru, identifying opportunities to improve resource use and tailor strategies to enhance sustainability. The multivariate analysis proved effective in capturing the complexity of these systems, supporting productivity and improving livelihoods in rural areas.