This study assesses improved maize variety (IMV) adoption, as well as yield effects among smallholder farmers in Benue State, Nigeria, having implications for sustainable crop management and resource-use efficiency. Benue state is commonly known as the “Food Basket of the Nation,” but the average maize yield remains less than 2 t/ha, compared to 7–10 t/ha when achieved under improved technologies, and it shows a key sustainability challenge for food security and land-use efficiency. With primary cross-sectional survey data from 205 smallholder farmers with 107 adopters and 98 non-adopters, selected across Local Government Area (LGAs) in Benue State, this study adopts Propensity Score Matching (PSM) for controlling selection bias and estimating the Average Treatment Effect on the Treated (ATET). Nearest Neighbour Matching acts as a primary estimator through robustness checks while using Radius and Kernel Matching. However, the logit model shows that IMV is greatly determined by gender, use of fertilizer, formal education, cooperative membership, access to irrigation, and extension contact, highlighting the crucial parts of human capital, complementary inputs, and institutional support in promoting sustainable adoption of technology. Following the control for observable differences across matching, a 0.399 log-unit yield gain was achieved by adopters, which is equivalent to approximately 49% higher output per hectare compared to non-adopters, an effect that is robust throughout alternative matching algorithms, and it surpasses the 38.7% national-level yield increase, indicating a regional sustainability premium in Benue State. The gains in productivity can promote land-use efficiency, decrease pressure for agricultural intensification on vulnerable lands, and enhance the case for integrated crop management. But adoption remains limited by access to quality seeds, complementary inputs, credit, and sustained gender barriers. Improving input supply chains, extension services, and institutional support is therefore crucial for developing productivity, resource-use efficiency, and food security across smallholder farming systems.

This study examines the impact of improved maize seed varieties (IMVs) on farm yield among smallholder Benue state, Nigeria and identifies key determinants of adoption. Benue State is often referred to as “Food Basket”, but has an average yield of less than 2 tons per hectare, compared to 8-10 tons per hectare that can be achieved under improved technologies. While previous nationally representative studies disguise local heterogeneity, this study focuses specifically on Benue State using primary cross-sectional data from 205 maize farmers. However, minimizing selection bias was carried out by matching adopters and non-adopters with similar observable characteristics and this method was introduced by using Propensity Score Matching (PSM) to estimate the causal impact of improved maize seed varieties (IMVs) adoption on maize yield. Nearest Neighbour Matching is used to compute the Average Treatment Effect on the Treated (ATET), with robustness checks using Radius and Kernel Matching. The results indicated that IMV adoption is significantly determined by gender (heads of male household), formal education, use of fertilizer, irrigation access, members of cooperative, and extension contact, emphasizing the significant roles of human capital, complementary inputs, as well as institutional support. Afterwards, the control of observable differences through matching led adopters to achieving a yield gain of 0.399 log-units which is relative to non-adopters that were not matched, and this is equivalent to 49% increase in output per hectare. The robustness across alternative matching algorithms is effective, compared with national-level evidence reporting a 38.7% yield increase [11]. Our finding suggests that the productivity of premium for IMVs may be greater in regions like Benue. The reliability of this treatment effect is confirmed using alternative matching algorithms in Robustness checks. Conclusively, the study of IMVs full potential is limited by inadequate access to quality seeds, complimentary inputs, funds, and gender-specific interventions.