Psychometric properties of the Bergen Social Media Addiction Scale (BSMAS): An item response theory and classical test theory analysis

Over the past two decades, social networking services have become a pervasive and integral part of young adults’ lives. At the same time, a growing body of evidence points to their negative effects on mental health and their potentially addictive nature. In this context, the Bergen Social Media Addiction Scale (BSMAS) was developed to assess problematic social media use (PSMU). Although the BSMAS has been translated and validated in Spanish, its validity and reliability have yet to be examined in an Ecuadorian population. Using a cross-sectional design, the present study examined the psychometric properties of the BSMAS in Ecuadorian university students using both Classical Test Theory (CTT) and Item Response Theory (IRT). The sample consisted of 580 university students aged 18 to 54 years (M = 20.09, SD = 3.27), 58.4% of whom were female. CTT supported an adequate fit for a unidimensional model (CFI = .992, TLI = .987, SRMR = .046), with high internal consistency (α = .88; ω = .85), moderate associations with anxiety (r = .41), depression (r = .43), and hours of SNS use (r = .40), and a weak negative association with well-being (r = -.16), assessed with the Patient Health Questionnaire (PHQ-4) and World Health Organization Well-being Index (WHO-5), and strict measurement invariance across sex. IRT showed adequate discrimination and difficulty parameters for the items. The instrument also reached its peak information in the range from -1 to +2. Overall, the findings suggest that the BSMAS is a suitable instrument for measuring PSMU among Ecuadorian university students.