Maartje Boer

CHAPTER 5 142 .higher SMU intensity and more SMU problems than boys. Girls also experienced less impulsivity than boys. Pre-vocational educated adolescents reported higher SMU intensity, more SMU problems, and more ADHD- symptoms than intermediate or pre-university educated adolescents. Adolescents with two Dutch parents reported more SMU intensity and more hyperactivity than adolescents with at least one parent from another country. These observed mean differences in factor scores do not affect our longitudinal results, because the RI-CLPM controls for all possible time- invariant confounders, which makes adding between-person characteristics as covariates redundant (Hamaker et al., 2015). ADHD-Symptoms, SMU Intensity, and SMU Problems The overall model fit of the RI-CLPM was good (CFI = 0.998; TLI = 0.984; RMSEA = 0.042; χ ²(10) = 19.472, p = 0.035). Table 5.4 shows the correlations between the random intercepts. Adolescents with high averages of SMU intensity and with high averages of SMU problems also reported high averages in attention deficits, impulsivity, and hyperactivity (correlations varying from r = 0.23 to 0.29, p = < 0.001 to 0.032). Adolescents who reported high averages of SMU intensity also reported high averages of SMU problems ( r = 0.40, p < 0.001). Table 5.5 depicts the auto-regressive and cross-lagged effects at the within-person level. Results for Hypotheses 1 and 2 are denoted by the light gray cells in the table and are all non-significant. Specifically, adolescents whose ADHD-symptoms increased did not report increases in SMU intensity one year later, nor did they report increased SMU problems one year later. These findings refute Hypotheses 1 and 2. The dark gray cells in Table 5.5 depict results for Hypotheses 3 and 4. Adolescents whose SMU intensity increased did not report increases in ADHD-symptoms one year later, because we did not find cross-lagged effects between SMU intensity and ADHD-symptoms. This finding fails to support Hypothesis 3. However, adolescents whose SMU problems increased, also experienced increased attention deficits one year later, both from T1 to T2 ( β = 0.31, p = 0.004) and from T2 to T3 ( β = 0.50, p = 0.016). Comparison of unstandardized effect sizes using a Wald-test indicated that the strength of these found relations were not significantly different ( χ ²(1) = 0.03, p = 0.870). Also, adolescents who experienced increased SMU problems at T2, reported