Abstract
To address a key gap in theorizing and research, we consider relational uncertainty as an explanation for why individuals experiencing depressive symptoms may be reluctant to talk about sensitive issues. We report closed-ended online survey data from 126 romantic couples in which one or both partners had been professionally diagnosed with depression. Results indicated that relational uncertainty mediated the positive association between depressive symptoms and topic avoidance. Partner effects from men to women also emerged. These findings are valuable for identifying relational uncertainty as a mechanism that may explain people's unwillingness to discuss challenging topics when they are experiencing depressive symptoms.
Acknowledgements
This research was supported by funds from the Campus Research Board of the University of Illinois. The authors are grateful to Corey Bandur, Anthony Carlsen, Bethany Carmien, Kathryn Clausing, Ben Ganellen, Amanda De Matteo, and Geena Wakeley for their assistance.
Notes
[1] We evaluated this assumption by surveying our participants about a broad range of avoided topics that have surfaced in the literature (e.g., Afifi & Schrodt, Citation2003; Caughlin & Afifi, Citation2004), including: (1) friendships with other people, (2) money, (3) family members, (4) everyday events, (5) household chores, and (6) drinking, smoking, and the use of drugs. This general measure of topic avoidance was unidimensional according to CFA results, but as we anticipated, participants reported less topic avoidance about these issues (M = 2.31, SD = 1.22, α = .81) than about depression (M = 3.59, SD = 1.27) or the relationship (M = 3.15, SD = 1.59). The results for the general measure were relatively similar to the results for topic avoidance about depression and the relationship, so we focus on the two more salient issues for parsimony.
[2] Although the study was open to both heterosexual and homosexual couples, the small number of homosexual couples who participated precluded meaningful comparisons. Hence, data from six lesbian couples and three gay couples were eliminated from the sample.
[3] Please contact the first author for the CFA results.
[4] The four measures of relational uncertainty shared strong positive bivariate correlations (see ), but subsidiary CFA results indicated that three models attempting to account for the overlap did not fit the data: (1) a single-factor model, (2) a four-factor model, and (3) a second-order model. We considered the four measures separately in light of these results coupled with theorizing about the nature of relational uncertainty (Knobloch, Citation2010).
[5] Covariances of the error terms between partners were: (1) self uncertainty = .50, p < .001, (2) partner uncertainty = −.02, ns, (3) relationship uncertainty = .06, ns, (4) topic avoidance about depression = .48, p < .01, and (5) topic avoidance about the relationship = .71, p < .001.