Maartje Boer

CHAPTER 5 148 possible confounding time-invariant traits. The findings of this study are therefore an important first step in answering the question of directionality. By distinguishing two types of social media behaviors and three symptoms of ADHD, we gained a better understanding of the relation between specific elements of both social media behaviors and of ADHD-symptoms. However, the self-report measures used in this study may deviate from observed ADHD- and social media behaviors (Orben & Przybylski, 2019a). Also, due to the use of long time-intervals, potential relations between daily fluctuations in ADHD-symptoms and social media behaviors could not be observed. Additionally, time-varying covariates that are not included in the study may have contributed to the found associations. For example, age may have played a role in the found relations over time, because during adolescence SMU intensity typically increases with age (Boer & Van den Eijnden, 2018). Furthermore, the convenience sample and the somewhat overrepresented native and pre-university adolescents relative to the general adolescent population in the Netherlands limit the generalizability of our findings. Taking these limitations into account, more longitudinal research on social media behaviors and ADHD-symptoms using more waves and shorter time-intervals, with larger and more representative samples is desired to confirm the unidirectional conclusion of the present study. More specifically, future research using smartphone applications that measure time spent on (specific) social media in combination with momentary assessments of ADHD-symptoms may provide more objective (and specific) insights into the relation between SMU intensity and ADHD-symptoms over time (Orben & Przybylski, 2019a). Another promising direction for future research would be the investigation of the longitudinal relations between social media behaviors and ADHD-symptoms for different subgroups separately, because particular groups (e.g., girls, low-educated) may be more susceptible to media effects (Valkenburg & Peter, 2013). Conclusion To conclude, findings from this longitudinal study suggest that SMU problems increase ADHD-symptoms among adolescents, but SMU intensity does not. Moreover, our findings indicate that the relation was unidirectional, because the reverse pattern was not observed. The present study extends