This study has several noteworthy strengths, including longitudinal assessment and a large, diverse sample. However, the following limitations must be acknowledged. First, attrition among the study sample was significant at each time point, although we found no difference between participants who did and did not complete follow up assessments on variables of primary interest. A second notable limitation is the use of self-report measures of emotion dysregulation. This method presents challenges as compared to observational or lab-based measures, such as asking adolescents to report on their typical response to emotional experiences. Because of biases associated with recalling affective experiences, such as a tendency to remember the most recent experience (Fredrickson 2000; Stone et al. 1998), adolescents may have provided responses based only on their most recent or most intense emotional experiences. Shared method variance is also a limitation, given that psychopathology was also assessed using self-report. However, the use of self-report measures is common in studies examining adolescent emotion regulation (Sim and Zeman 2005; Southam-Gerow and Kendall 2000; Zeman et al. 2002), and in the current school-based study was the
