Assessing the Contribution of Nonverbal Behaviors in Displays of Confirmation During Parent-Adolescent Interactions: An Actor-Partner Interdependence Model

Abstract

The current study examined the role of nonverbal behaviors in displays of confirmation (i.e., validation, acceptance) in parent-adolescent interactions. An advantage of this study is that actor-partner-interdependence models were used to examine how participants' own behaviors as well as their partners' behaviors predicted confirmation. Analyses showed that parents were more confirming than adolescents, but both actor and partner nonverbal behaviors were associated with confirmation levels. In general, involvement and pleasantness were associated with greater levels of confirmation, whereas expressiveness was associated with less confirmation. Participant sex and member status did not largely moderate the relationship between the nonverbal dimensions and confirmation; however, specific findings suggest that nonverbal behaviors associated with confirmation by parents may be somewhat different than nonverbal behaviors associated with confirmation by adolescents. Overall, the nonverbal dimensions uniquely predicted a quarter of the variance in confirmation suggesting that nonverbal behaviors play an important role, but are not the sole indicators, in this type of relational communication.

Notes

¹The reliability for the friendliness item for one coder across time points was .513, which was largely based on the difference in rating for the fifth minute. Hence, only the first four time points (α = .76) were included to form the aggregate score across time points.

²In APIMs, the data is set up so that each individual is a separate case. Additionally, each individual's score on an independent variable is entered on his/her own row as an actor score and on his/her partner's row as a partner effect. The following table includes example data to exemplify the way the data is set up in APIMs. For dyad #1, as noted by the arrows, the adolescent's vocalic pleasantness score is entered on the adolescent's row to assess the actor effect (how his/her own vocalic pleasantness is related to his/her own confirmation) and also entered on their parent's row to assess the partner effect (how the adolescent's vocal pleasantness is related to their parent's confirmation). An example of an interaction term for RQ1c is also included in the last column. Note. For Member Status, Adolescent = −1 and Parent = 1. Voc = Vocalic and Pleas = Pleasantness.

³Equations for calculating pseudo R² use the level-1 (τ₀₀) and level-2 (σ²) error terms. Unrestricted error terms are those from the baseline model and the restricted error terms are those from models including predictors. n = number of individuals in each level-2 unit, which in this study is 2.

Level-1 R² = 1 – [(level-1 restricted error+level-2 restricted error)/(level-1 unrestricted error+level-2 unrestricted error)];

Level-2 R² = 1 – [(level-1 restricted error/ n+level-2 restricted error)/(level-1 unrestricted error/ n+level-2 unrestricted error)].

⁴The tables for this and subsequent models can be obtained from the author.