Maartje Boer

VALIDATION OF THE SMD-SCALE 45 2 of positive scores were lower than in the full sample. In Class 2 (34.75% of the sample), the proportions of positive scores were higher than in the full sample and Class 1 and ranged between 6.88% (‘conflict’) and 59.38% (‘escape’). In Class 3 (3.60%), the proportions of positive scores were higher than in Class 2 and varied between 66.11% (‘displacement’) and91.70% (‘problem’). Given that the proportions of positive scores on the nine criteria were highest in Class 3, followed by Class 2 and Class 1, respectively, we labeled the three classes as problematic SMU (Class 3), risky SMU (Class 2), and normative SMU (Class 1). Figure 2.3 Proportion of Positive Scores on the Nine Criteria, by Latent Class, n = 6,626 We found that differences in the proportions of endorsed criteria within classes often paralleled the full sample’s differences (e.g., ‘tolerance’ was one of the least endorsed criteria in the full sample and in the three class samples). In other words, we did not observe clear item patterns that distinguished between the three latent classes. Rather, the classes seemed to be distinguished by either high, medium, and low probability of endorsing any of the nine criteria. Therefore, we compared the three classes on adolescents’ number of endorsed criteria. Subsequently, we plotted these scores with the latent classes (Figure 2.4). In the problematic SMU class, most adolescents (87.07%) endorsed at least six criteria. In the risky problematic