Understanding farmers’ acceptance of digital technologies: a meta-analytic structural equation modeling approach based on the technology acceptance model

Abstract

Understanding how farmers accept and interact with digital technologies is crucial for transforming agriculture and improving smallholder livelihoods. Despite widespread use of the technology acceptance model (TAM) to understand farmers’ adoption of digital technologies, findings remain inconsistent. We applied a two-stage meta-analytic structural equation modeling approach to synthesize 53 correlation matrices from 33 TAM studies involving 13,660 farmers across 17 countries. While the analysis confirms TAM’s overall robustness in digital agricultural contexts, it also reveals unexpected findings. First, despite TAM’s assumption that perceived ease of use (PEU) strongly predicts behavioral intention (BI), the meta-analytic results show that PEU has a weak and, in some models, non-significant effect on BI, challenging a core TAM hypothesis. In contrast, perceived usefulness (PU) consistently emerges as a strong determinant of attitude (AT) and remains central to farmers’ acceptance of digital tools. The integration of external variables, particularly facilitating conditions (FC) and social influence (SI), meaningfully strengthens key TAM pathways, highlighting the importance of supportive environments and social dynamics. Unexpectedly, the type of digital technology does not moderate TAM relationships, indicating that farmers evaluate digital technologies more on perceived value than on technological complexity. Finally, despite generally positive behavioral intentions reported in primary studies, the meta-analysis identifies a persistent intention–use gap, with actual digital technology use remaining low. These findings underscore the need to tailor digital agriculture policies to farmers’ perceptions, emphasizing practical benefits and supportive conditions to enhance adoption and contribute to the transformation of sustainable food systems.