Answers are grounded in this paper and cite their sources.
Climate change and population growth pose significant challenges to agricultural production. Although Climate-Smart Agriculture (CSA) offers great potential to enhance productivity and resilience, its adoption remains low in Henan Province, China. This study develops a ratio-based CSA adoption index that captures adoption depth and breadth, and applies a combined econometric-machine learning framework to identify heterogeneous factors across CSA adoption levels. By integrating generalized ordered logit regression (GOLR) with XGBoost, the research reveals both persistent and heterogeneous factors influencing adoption. Persistent factors, including weather information accessibility, flood control infrastructure, and agricultural extension services, consistently support CSA adoption across all levels. However, heterogeneous factors play crucial roles at different adoption levels. Technical support and maize selling channels are pivotal during the transition to medium-high adoption level, while profitability perception, age, and household size become more influential in the shift to high adoption level. The results demonstrate that farmers face evolving constraints across adoption stages, highlighting the need for stage-specific rather than uniform CSA interventions. Taken together, the results provide actionable insights for developing stage-sensitive and resource-efficient CSA strategies that enhance policy effectiveness and support sustainable and widespread adoption.
No takes yet. Share an insight, caveat, or question.
Zhao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: