Problem:
Our work also has limitations that should be recognized in future studies. First, adolescents attended different school tracks within school levels, which is due to the national school systems. In German-speaking countries, there exist several school tracks for specific age groups that are divided according to students' performance. Unfortunately, we did not collect information about specific schools and classes that the students attended. Thus, we could not control for the probably partly nested nature of our data in the analyses and, hence, for influences of students' reference group on the academic achievement measures, such as the big-fish-little-pond effect (for a review, see Marsh & Seaton, 2015). Future studies should gather more information about students' current school situation and control for clustering of students within schools and classrooms. Second, the predictors had unequal and, in some cases, only marginally sufficient reliability. Concerning the low reliability of intelligence in Study 1, however, age-separated analyses showed higher reliability for older children and the mean and standard deviation were distributed nearly the same as in the general population (see Table 1). Third, associations were almost completely cross-sectional and therefore no conclusions on causal relationships can be made. In future studies, growth curve designs using longitudinal data or experimental studies should be employed to investigate the directionality of the effects. Need Assignment Help?