Systematic review and meta-analysis of adoption, acceptance, and perceptions of precision livestock farming technology

Abstract

Digital technologies are increasingly reshaping livestock production systems, with precision livestock farming (PLF) emerging as a key innovation that uses sensor-based monitoring and data analytics to support animal management, welfare oversight, and decision-making on farms. Despite PLF’s technical potential, adoption across the livestock sector remains uneven, and stakeholder perceptions of these technologies vary widely. This study provides a meta-analysis of empirical research examining adoption, acceptance, and stakeholder perceptions of PLF technologies, aiming to identify key social determinants shaping the diffusion of digital livestock technologies. Following PRISMA 2020 guidelines and a registered protocol (PROSPERO: CRD420261282807), systematic literature searches were conducted in Scopus, Web of Science, and PubMed/MEDLINE. Twenty-nine peer-reviewed studies met the inclusion criteria and were included in the quantitative synthesis, contributing 421 study-level effect-size estimates that constituted the unit of analysis. Random-effects meta-analysis (restricted maximum likelihood, REML) was used to estimate pooled adoption rates, acceptance scores, and perception outcomes across stakeholder groups, with heterogeneity assessed using I2 statistics, and within-study dependence addressed through cluster-robust variance estimation (RVE, CR2). As approximate summary patterns rather than firm benchmarks given the still-limited evidence base — the pooled estimate indicates that adoption of any PLF technology pooled to approximately 53% (95% CI: 38–68%; five studies), whereas adoption of individual technologies pooled to only ∼17% (95% CI: 12–22%; 146 estimates from 16 studies) — most stakeholders use at least one PLF tool, but uptake of specific technologies remains low. In this study, acceptance refers to stakeholder attitudes or evaluations of PLF technologies measured on Likert-type scales (additionally rescaled to a percent-of-maximum-possible (POMP, 0–100) metric to enable cross-scale comparability), while perception refers to the direction and valence of stakeholder views, including the proportion expressing positive assessments. Perception analyses suggest that approximately 69.0% (95% CI: 63.5–74.0%; k = 69 effect sizes from 7 studies) of respondents expressed positive views of PLF technologies, and pooled acceptance and perception scores on the POMP metric were approximately 71.5% (95% CI: 68.2–74.8%) and 63.2% (95% CI: 59.1–67.3%), respectively. However, substantial heterogeneity was observed across all outcome categories (I 2 > 90%), and all pooled estimates should be interpreted with appropriate caution as indicative patterns rather than precise benchmarks. A principal methodological limitation of this synthesis is the very high between-study heterogeneity and the small number of eligible studies (k = 29; 421 effect-size estimates clustered within studies). The results suggest a gap between positive stakeholder attitudes and actual technology uptake. These findings highlight the importance of addressing economic, technical, and institutional barriers to facilitate broader adoption of PLF technologies and support socially responsive digital transformation in livestock systems. To the best of our knowledge, this synthesis provides the first quantitative estimates of PLF adoption and perception rates across stakeholder groups and livestock sectors, though the small evidence base warrants cautious interpretation. The findings offer actionable insights for policymakers, technology developers, and agricultural advisors seeking to align digital livestock innovations with stakeholder needs, institutional capacities, and societal expectations for responsible agricultural transformation.