(PS4-A7) Comparing Conventional Retrospective Self-report vs. Ecological Momentary Assessments in Measuring Rumination in Adolescents

Background: Rumination is a cognitive risk factor associated with depression. Theories have suggested that rumination can contribute to the onset and maintenance of depression. Past research has largely relied on conventional retrospective self-report measures to assess ruminative thoughts in adolescents. Specifically, the Child’s Response Style Questionnaire (CRSQ) is a widely used measure to assess rumination in youth. More recently, ecological momentary assessment (EMA) has also been used to examine ruminative thought characteristics and consists of daily smartphone surveys that can assess thought and mood in the daily lives of individuals. Previous studies have found gender differences in rumination. However, these studies have relied on conventional, retrospective self-report measures. In the current study, we compared rumination levels between adolescent boys and girls using conventional self-report (CRSQ) vs. EMA. 

Methods: Participants (n=71 girls and n=40 boys) aged 12-18 years old (M_age= 15.9) completed EMA smartphone surveys two to three times a day for one week assessing two items of rumination. These smartphone surveys (total 1,037 surveys) also assessed characteristics of spontaneous thought and mood. Additionally, participants were administered the CRSQ at baseline, originally developed by Abela et al. (2000). The survey consisted of 25-items on a 4-point Likert scale.  

Results: The correlation between the conventional self-report (CRSQ) and EMA measures of rumination was r = 0.63, p < .001. For the self-report measure (CRSQ), and consistent with prior research, girls reported significantly higher rumination than boys, t(109) = -2.09, p = .039; Cohen’s d = 0.42). In contrast, for the EMA measure of rumination, there were no significant differences between genders in rumination (t(110.14) = -1.32, p = .189; Cohen’s d = 0.23).

Conclusion: The traditional self-report and EMA measures yielded contradictory results, with gender differences only emerging on the former measure, which is the one most commonly used in the literature. These findings may reflect a retrospective bias in the self-report questionnaire (not present in EMA), with girls overreporting and/or boys underreporting when reporting rumination from memory. In contrast, the EMA measure may be more biased (e.g., random sampling of rumination in the daily lives of teens may not adequately capture gender differences in rumination in response to specific stressors). Future research is needed to adjudicate between these (and other) competing interpretations of this discrepancy in findings between measures.