This study presents a multidimensional analysis of poverty in Nigeria using quantile regression, based on data from the National Social Registry (NSR) of Poor and Vulnerable Households (PVHHs). Focusing on six high-poverty states which are Ebonyi, Cross River, Ekiti, Sokoto, Taraba, and Niger. It investigates the influence of education, employment, location, geographic zone, and household characteristics on poverty, measured through Proxy Means Test (PMT) scores at the 25th, 50th, and 75th quantiles. Unlike Ordinary Least Squares (OLS), which may misestimate effects across poverty levels, quantile regression captures heterogeneous impacts across the distribution.Findings reveal that higher education, urban residence, and waged employment significantly reduce poverty, with stronger effects observed at higher quantiles. In contrast, larger household size, female-headed households, and residence in the North-East and North-Central zones are associated with greater poverty. Regional disparities are pronounced, with southern zones exhibiting greater poverty resilience. These results highlight the need for targeted interventions, including vocational training, urban infrastructure development, agricultural support in northern zones, and social protection for female-headed households. This study adds depth to the poverty literature by analyzing NSR data through a distributional lens, yielding rich insights for policy interventions tailored to specific needs. In conclusion, we recommend that dynamic analyses using panel quantile regression and explore vulnerabilities among specific subgroups, such as rural female-headed households in high-poverty areas.
N et al. (Mon,) studied this question.