![Sample our Communication Studies journals, sign in here to start your access, 2013 & 2014 volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5927/TOC-Subject-Sample-2021-Information-Science2.jpg"})
Abstract
A model was constructed to test the argument that when the topic of the parents’ relationship is introduced in conversations between parents and ‘adolescents, adolescents from divorced families may be especially likely to feel caught between their parents due to a need for protection (of themselves, their parent, and their relationship), which should make them anxious (i.e., self-reported anxiety) and physiologically aroused (i.e., changes in skin conductance levels or SCL). When adolescents feel aroused, we argued that they should attempt to avoid talking about their parents’ relationship with their parent. Self-report and observational data, as well as physiological data, were collected from 112 parent-adolescents dyads. The results revealed that divorce predicted adolescents’ feelings of being caught, which influenced their need for protection. This need for protection, in turn, predicted adolescents’ self-reported anxiety and changes in SCL. Unlike what was hypothesized, SCL was not associated with adolescents’ avoidance tendencies. Nevertheless, self-reported anxiety was associated with adolescents’ self-reported topic avoidance. The implications of these results, and a new observational coding scheme for avoidance, are discussed.
Notes
1. The mediating paths in the model were examined in follow-up analyses to better understand the direct and indirect effects of the variables in the model. The paths in the model were deconstructed and the direct and indirect effects were observed. When there are multiple mediating paths in a model, each mediating path needs to be examined in isolation from the other mediating paths in the model to assess the specific indirect effects (see Brown, Citation1997). The breakdown of the direct and indirect effects is equivalent to Baron and Kenny's (Citation1986) four conditions necessary for mediation. The mediating effects were then verified with the Sobel test (Sobel, Citation1982).
The mediating effect of feeling caught with divorce and protection was examined first in isolation from the other mediating variables in the model. Divorce was not significantly associated with protection, β=.11, ns. When feeling caught was introduced as a mediating factor of divorce and protection, divorce was still not significantly associated with protection, β = −.11, ns, but was positively and significantly associated with feeling caught, β=.49, p<.001, which was positively and significantly associated with protection, β=.43, p<.01. Given that the association between divorce and protection was nonsignificant before and after the inclusion of the feeling caught variable, there was not a mediating effect for feeling caught. Rather, there were direct associations between divorce and feeling caught and feeling caught and protection. Divorce, alone, was not significantly associated with protection.
A second mediation path that was of interest was the extent to which protection mediated the association between feeling caught and self-reported anxiety. Feeling caught was positively and strongly associated with self-reported anxiety, β=.81, p<.001. When protection was introduced in the model, the association between feeling caught and anxiety was nonsignificant, β=.18, ns. Feeling caught was also associated with protection, β=.57, p<.001, and protection was associated with anxiety, β=.40, p<.05. The reduction in the path between feeling caught and anxiety when protection was included in the model was significant, z = 1.99, p<.05, providing support for mediation.
A final mediation path that was analyzed was the degree to which protection mediated the path from feeling caught to physiological arousal. Feeling caught was not significantly associated with physiological arousal, β=.13, ns, and was also not associated with physiological arousal, β = −.04, ns, when protection was included in the model. Feeling caught was associated with protection, β=.56, p<.001, but protection was only approaching significance with physiological arousal, β=.23, p<.06. Therefore, there was no evidence of mediation for protection and physiological arousal when protection, feeling caught, and physiological arousal were examined in isolation from the other variables in the model. Apparently, arousal or SCL is affected by the combined impact of divorce, feeling caught, and protection, with the only direct association to it being protection when the full model is considered.