Structural Selectivity in Digital Dairy Farming: Evidence from Sensor-Collar Adoption Among Family Farms in Southern Brazil

Digital technologies are increasingly used in dairy farming to improve efficiency, sustainability, and animal management, yet adoption may vary across heterogeneous farms. This study examined sensor-collar adoption among family dairy farms in Rio Grande do Sul, southern Brazil, focusing on farm-structural, operator-level, and infrastructural characteristics. Primary data were obtained from 142 questionnaires; 140 complete cases were analyzed, including 39 adopters and 101 non-adopters. Adopters and non-adopters were compared using Mann–Whitney U and chi-square tests, followed by multivariable logistic regression. Sensitivity analyses included backward elimination and log-transformed herd size. Adopters differed significantly across all six observed characteristics. In the full model, herd size was the strongest and most robust correlate of adoption. In the parsimonious model, herd size remained strongly associated with adoption (OR = 1.013 per animal; p < 0.001), intensified feeding systems showed approximately threefold higher odds than pasture-based systems (OR = 3.20; p = 0.053), and better road conditions were positively associated with adoption (OR = 1.46 per level; p = 0.082). Overall, adoption showed a structurally differentiated profile, dominated by productive scale but complemented by production intensity and physical accessibility, consistent with structurally selective digital technology diffusion.