The Swedish version of the Bergen Social Media Addiction Scale: A psychometric evaluation among adolescents

The Bergen Social Media Addiction Scale (BSMAS) is one of the most widely used measures to assess problematic social media use (PSMU). However, no validated Swedish version for adolescents is available.

The BSMAS was translated and culturally adapted into Swedish following established guidelines and psychometrically evaluated among a school-based sample of 658 adolescents aged 13-17 years from three schools (332 boys, 313 girls, and 13 with undisclosed gender; 580 Swedish-born and 78 non-Swedish-born). Reliability and construct validity were examined using classical test theory (CTT) and Rasch modeling. Confirmatory factor analysis (CFA) was conducted to assess dimensionality, and measurement invariance was tested across gender and country of birth. Evidence based on relationships to other variables was further explored using structural equation modeling and network analysis, including associations with social media use duration and symptoms of anxiety and depression assessed using the PROMIS Pediatric Depression and Anxiety Scales.

Among the competing CFA models, the best fit was obtained for a two-factor structure in which (i) salience and tolerance reflected peripheral criteria, and (ii) mood modification, relapse, withdrawal, and conflict reflected core criteria. Internal consistency was acceptable for both the peripheral factor (Cronbach's α = 0.793; McDonald's ω = 0.795) and the core factor (Cronbach's α = 0.764; McDonald's ω = 0.786). Rasch analyses indicated acceptable item fit and limited differential item functioning across gender and country of birth. Multigroup CFA supported configural, metric, and scalar invariance across both gender and country of birth. Network analyses showed that social media use duration was related to BSMAS symptoms but was not central in the symptom network. Structural equation modeling further showed that both peripheral and core criteria were positively associated with anxiety and depression, with stronger associations observed for core criteria.

The Swedish version of the BSMAS showed acceptable reliability and validity among adolescents, with findings supporting a two-factor representation that distinguishes peripheral from core criteria. The scale appears suitable for adolescent research, school-based studies, and early screening contexts.