Poster #137 - Longitudinal Relations Between Adolescent Sadness/Worry Regulation and Friendship Quality: a Cross-Lagged Analysis

• In Event: Poster Session 14

Abstract

Adolescents who adaptively manage negative emotions appear to be more competent in building close friendships (Miller-Slough & Dunsmore, 2020). High quality friendships characterized by frequent discussion and sharing of experiences (Furman & Rose, 2015) can provide a safe environment for adolescents to practice and hone adaptive emotion regulation (ER, von Salisch & Zeman, 2018). However, research has not investigated whether the regulation of internalizing emotions, such as sadness and worry, and the quality of close friendships may influence each other over time. Moreover, the stability of friendship quality and ER over time during adolescence has scarcely been examined (Furman & Rose, 2015). Seeking to address these gaps, in the present study, we will use a cross-lagged panel design with 3 timepoints over 4 years to identify the longitudinal pathways between internalizing ER (i.e., regulation of sadness/worry) and positive friendship quality. We will also examine the stability of ER and friendship quality from early- to mid-adolescence.
Participants at T1 were 209 adolescents (Mage = 12.66 years, SD = 1; 51.7% girls; 77% White) with 169 of these adolescents (Mage = 14.70 years, SD = 1.05, 52.1% girls; 75% White) participating at T2, and 132 adolescents (Mage = 16.43, SD = 1.72, 53.7% girls, 77% White) participating at T3. We combined four subscales from the Children’s Sadness and Worry Management Scales (Zeman et al., 2002, 2010) to form one sadness/worry regulation scale, (see Figure 1 for the scale structure). Excluding its 3-item Conflict subscale, the total score of a 15-item, 5-subscale Friendship Quality Questionnaire (Parker & Asher, 1993) was used to assess positive friendship quality.
First, as illustrated by paths denoted as “a” in Figure 2, we hypothesize that the frequency of adaptive ER will increase slightly over time given that these skills may be honed with experience (Zeman et al., 2019). Second (b in Figure 2), we expect friendship quality to increase slightly over time (Way & Greene, 2006). Third (c), we posit that adolescents with better ER will be concurrently involved in higher quality friendships at all timepoints (c in Figure). Fourth (d in Figure), we expect that ER at T1 will predict increased friendship quality at T2, which will then predict stronger ER at T3 because high quality friendship may provide a supportive context for practicing regulation skills. Fifth (e in Figure), we anticipate that T1 friendship quality will be weakly positively associated withT2 ER, but T2 ER will strongly predict T3 friendship quality since stronger ER skills can increase friendship quality (Miller-Slough & Dunsmore, 2016). Gender will be a covariate in the model.
We will use Little’s MCAR test to examine whether the longitudinal data is missing at random, and will use full maximum likelihood estimation to impute missing data. Using Mplus 8.8 (Muthén & Muthén, 1998-2017), the cross-lagged panel model (see Figure 2) will examine the directional influence of sadness/worry regulation and friendship quality on each other over time.

Authors

• Rachel Gillian Li, William and Mary
  Presenting Author

• Janice L Zeman, William and Mary