Persistent Multidimensional Mobile Broadband Disadvantage and Population Exposure in Nigeria

Mobile broadband inequality can persist even when national averages appear stable, and the population significance of recurring disadvantage is not captured by state rankings alone. This study examines the magnitude, temporal persistence, and population exposure of state-level mobile broadband disadvantage across Nigeria's 36 states and the Federal Capital Territory using four consecutive quarters of Ookla Open Data from Q2 2025 to Q1 2026. Device-weighted state estimates were calculated for download throughput, upload throughput, and latency, with test weighting retained for robustness. Inequality was summarized with dispersion measures and cross-metric rank correlations. Temporal persistence was assessed with quarter-to-quarter Spearman correlations, Kendall's coefficient of concordance, Friedman tests, and random-intercept mixed-effects models. A state was classified as persistently disadvantaged when it occupied the adverse quartile in at least three of four quarters, and WorldPop 2025 estimates were used to quantify population exposure. Download and latency were more unequal than upload. Kendall's W was 0.867 for download, 0.750 for upload, and 0.775 for latency, while mixed-model intraclass correlations ranged from 0.610 to 0.836. Fifteen states were persistently disadvantaged in at least one dimension, representing 38.3% of the modeled population; six states were disadvantaged in at least two dimensions, representing 13.4%. The findings show that mobile broadband disadvantage in Nigeria is multidimensional, persistent, and socially consequential at population scale.