Abstract
Using Structural Equation Modelling (SEM), this study investigated peer acceptance, peer pressure,
childhood neglect, cyber-stalking, and depression as predictors of social media dependence among in-school
adolescents in Ibadan, Nigeria. A multi-stage sampling approach was employed to sample 508 adolescents
from six randomly selected local government areas. Standardised instruments, including the Social Media
Disorder Scale, Childhood Trauma Questionnaire, and PHQ-9, were administered. The results revealed that
peer pressure (β = .23, p < .001), childhood neglect (β = .17, p < .001), cyber-stalking (β = .25, p < .001), and
depression (β = .26, p < .001) significantly and positively predicted social media dependence, while peer
acceptance (β = –.16, p = .001) had a significant negative effect. Depression significantly mediated the
relationships between the predictor variables and social media dependence, with all indirect paths
statistically significant (e.g., peer pressure: β = .09, childhood neglect: β = .08, cyber-stalking: β = .06, peer
acceptance: β = –.07; all p < .01). Moreover, peer acceptance moderated the relationship between peer
pressure (β = -0.12; p = 0.001), childhood neglect (β = -0.09; p = 0.011), and cyber-stalking (β = -0.10; p =
0.006), indicating a buffering relationship. The structural model goodness-of-fit indices indicated good fit -
χ²/df = 2.31, CFI = 0.957, TLI = 0.944, RMSEA = 0.051, SRMR = 0.042 - and accounted for 49% variance in
social media dependence (R² = 0.491). This research demonstrates the intersection of social variables,
emotion, and digital behaviours that affect adolescent behaviours, and underscores the need for targeted
interventions promoting peer support and emotional regulation.