Maartje Boer

CHAPTER 3 74 222,253 adolescents from 44 countries showed that the SMD-scale has good psychometric properties within a broad international context and demonstrates its suitability for cross-national comparisons in problematic SMU. First, the CFA confirmed good structural validity of the scale across all countries. Second, the internal consistency of the items was good in all countries, suggesting that the scale provides reliable scores. Third, the factor structure of the scale was measurement invariant across countries. Also, gender and socioeconomic status invariance was established in all countries, and age invariance in all countries except Malta. Fourth, in line with previous research, in almost all countries, problematic SMU was negatively associated with mental wellbeing and positively with the intensity of online communication, confirming good criterion validity. All countries showed good structural validity by means of good model fit of a one-factor model and high factor loadings of the items. These findings suggest that all nine items substantially contribute to theunderlyingconstruct of problematic SMU. This implies that alongside the six items referring to the core criteria of addiction (Griffiths, 2005; Griffiths et al., 2014), the three additional items that distinguish the SMD-scale fromother problematic SMU- scales (Andreassen et al., 2012, 2016), including problems, displacement, and deception, further contribute to the conceptualization of problematic SMU. Hence, with their inclusion, the SMD-scale may advance the measurement of problematic SMU. To verify this suggestion, future studies comparing the psychometric properties of the SMD-scale with scales based on only the six core criteria of addiction are recommended. The finding that the factor model was measurement invariant across countries implies that adolescents from different countries interpret the questions from the scale in a similar manner and that the scale measures the same underlying construct across countries (Davidov, 2010). Hence, the scale is suited for measuring and comparing adolescents’ level of problematic SMU in international surveys. Furthermore, as a next step, future research examining the potential reasons for country-level differences in the prevalence of problematic SMU are considered promising. Moreover, the finding that gender, age, and socioeconomic invariance was observed in all countries (except for age invariance in one country), implies that the scale also measures the same underlying construct for boys, girls, 11-, 13-, 15-year-olds,