Date of Award

Fall 12-21-2018

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Doctor of Philosophy (PhD)




Cynthia A. Erdley

Second Committee Member

Emily A. P. Haigh

Third Committee Member

Shannon McCoy

Additional Committee Members

Douglas W. Nangle

Rebecca Schwartz-Mette


The primary aim of this study was to investigate gender differences in problem content and dyadic problem talk duration as potential contributors to previously documented depressogenic effects of co-rumination in late adolescence. Participants (N = 176 undergraduate students) included pairs of same-gender female (n = 37), same-gender male (n = 15), and cross-gender (n = 36) friends who completed self-report measures assessing individual depressive symptom severity, as well as within-dyad co-rumination habits and friendship quality. Dyads also participated in an observational problem talk task, which asked each dyad member to identify a current personal problem and discuss it with their friend during a 16-minute videotaped session. Each participant’s identified problem was coded for inclusion of interpersonal and dependent content, and videotaped conversations were coded for the total time each dyad spent discussing problems and the total time each dyad member spent discussing their own problem (own-problem talk) and their friend’s problem (friend-problem talk).

Consistent with existing depression literature, results indicated that females reported greater depressive symptom severity than males. Female dyads also reported the most co-rumination and engaged in the longest total problem talk, and both male and female participants reported engaging in more co-rumination when their dyad partner was female. However, own- and friend-problem talk did not vary by gender, and neither co-rumination nor total, own-, or friend-problem talk duration were predictive of depressive symptoms. Although female gender did not predict problem content, and problem content was not associated with depressive symptoms, interpersonal problem content predicted increased own-problem talk.

These findings are in contrast to the overwhelming majority of research that has found co-rumination to be predictive of depressive symptoms, and provide no direct support suggesting that problem content and problem talk duration contribute to the depression gender gap. However, results do indicate that problem talk, a key component of co-rumination, is most likely to be prolonged when the problem being discussed has interpersonal content. The current results thus suggest that cumulative rather than interactive effects of gender and problem content may impact the co-rumination habits of late adolescents.