Generalized Anxiety and Relational Uncertainty as Predictors of Topic Avoidance During Reintegration Following Military Deployment

Abstract

For military couples reunited following deployment, discussing or avoiding topics is a central dimension of communication. This paper theorizes about two predictors of topic avoidance that arise from a lack of confidence in social situations: generalized anxiety and relational uncertainty. In Study 1, 220 returning service members described issues they avoid discussing upon reunion. Content analytic findings indicated eight avoided topics. In Study 2, 118 military couples reported on topic avoidance for the first 3 months after homecoming. Multilevel modeling results revealed that the generalized anxiety and relational uncertainty of actors, but not partners, were consistent predictors of topic avoidance. The findings illuminate the complexities of communicating following a tour of duty.

Keywords:

• Generalized Anxiety
• Military Couples
• Relational Uncertainty
• Topic Avoidance

Acknowledgments

Study 2 was supported by a grant from the Family Resiliency Center of the University of Illinois. Data collection procedures were approved by the Institutional Review Boards of the University of Illinois and Rutgers University. The authors are grateful to Jennifer Bert, Michael Golaszewski, Emmelyn Joy, Yuri Kleban, Hani Kuttab, Amy Lindgren, Brandi Marinko, Virginia Murray, Kimberly Pusateri, Larissa Tosi, and Lauren Whalley for their help with recruitment and coding. They also thank John Caughlin for his insightful feedback on the paper.

Notes

[1] Knobloch and Theiss (2011a) reported closed-ended data from this sample.

[2] We conducted subsidiary analyses on the demographic variables to compare participants who reported at least one avoided topic (n=113) to participants who reported that they do not avoid any topics with their romantic partner (n=98). Age was the only variable that distinguished the two groups. Participants who reported at least one avoided topic (M=34.59 years, SD=8.12 years) were older than participants who did not (M=30.71 years, SD=8.16 years), t (209) = 3.37, p<.001.

[3] A second study drawn from the sample is described by Knobloch, Ebata, McGlaughlin, and Ogolsky (in press). Generalized anxiety and all of the dependent variables are unique to this paper.

[4] The fit statistics for the CFA models are available from the first author.

[5] Subsidiary CFA findings demonstrated that (a) self, partner, and relationship uncertainty were not unidimensional with the 12 items loaded onto 1 factor, and (b) self, partner, relationship, and reunion uncertainty were not unidimensional with the 19 items loaded onto 1 factor. Thus, we evaluated the measures in separate models (following Knobloch & Theiss, 2011a).

[6] Findings were similar for the subsample of couples in which the man was deployed and the woman was the at-home partner (n=114 couples).