Problematic social media use and associated factors among undergraduate students at a Nigerian university: a cross-sectional study

Abstract

Problematic social media use (PSMU) involves dysregulated engagement characterised by impaired control and adverse consequences. Estimates among university students vary by instrument and threshold, while evidence from Nigerian universities remains limited. We estimated PSMU symptoms and associated factors among undergraduate students at the University of Benin, Nigeria. We conducted an analytical cross-sectional study among 540 full-time undergraduates at Ugbowo Campus, University of Benin, from January to March 2024. Multistage stratified random sampling was used across seven faculty-level academic units and 14 departments. PSMU was assessed using the six-item Bergen Social Media Addiction Scale (BSMAS; range 6–30). A score ≥ 18 was the prespecified screening threshold; ≥24 and continuous BSMAS scores were examined in sensitivity analyses. Crude and adjusted prevalence ratios (aPRs) were estimated using Poisson regression with a log link and robust variance within generalised estimating equations, accounting for department-level clustering. Statistical significance was set at two-sided p < 0.05. The mean BSMAS score was 18.78 (SD 6.29; Cronbach’s α = 0.889). Overall, 295/540 students (54.6%; 95% CI 50.4–58.8) scored ≥ 18 and 148/540 (27.4%; 95% CI 23.8–31.3) scored ≥ 24. The proportion scoring ≥ 18 increased from 15.0% among students reporting 0–1 h/day to 86.4% among those reporting > 10 h/day. Male sex was associated with higher prevalence of elevated symptoms (aPR 1.62; 95% CI 1.12–2.33; p = 0.010), as was each one-category increase in daily social-media-use duration (aPR 1.38; 95% CI 1.22–1.55; p < 0.001). Age, family structure and residence were not independently associated. Findings were directionally consistent using the ≥ 24 threshold and continuous BSMAS score. Elevated PSMU symptoms were common in this single-university sample, but estimates were threshold-dependent and should not be interpreted as clinical prevalence. Longer daily social-media use and male sex were independently associated with elevated symptoms. Multicentre longitudinal studies using locally validated measures are needed in Nigeria.