Journal of Psychological Science ›› 2024, Vol. 47 ›› Issue (4): 819-828.DOI: 10.16719/j.cnki.1671-6981.20240407

• Developmental & Educational Psychology • Previous Articles     Next Articles

Bidirectional Associations Between Friendship Quality, Subjective Well-Being and Loneliness Among Children: A Longitudinal Study

Liu Xu1,2, Liu Yuxiao1, Chen Qian1, Cao Min1, Peng Ji1, Zhou Zongkui1   

  1. 1School of Psychology, Central China Normal University, Wuhan, 430079;
    2Shen Zhen Technology University Affiliated High School, Shenzhen, 518118
  • Online:2024-07-20 Published:2024-07-17

儿童友谊质量与主观幸福感和孤独感的双向关系:一项纵向研究*

刘旭1,2, 刘宇潇1, 陈倩1, 曹敏1, 彭霁1, 周宗奎**1   

  1. 1华中师范大学心理学院,武汉,430079;
    2深圳技术大学附属中学,深圳,518118
  • 通讯作者: ** 周宗奎,E-mail: zhouzk@mail.ccnu.edu.cn
  • 基金资助:
    *本研究得到中国基础教育质量监测协同创新中心项目(2022-04-031-BZPK01)和湖北省教育科学规划重点课题(2022GA012)的资助

Abstract: The establishment of close friendships begins in childhood and is an important milestone in development. High-quality friendship is an essential force in promoting children's mental health development, which can improve children's subjective well-being and reduce loneliness. According to development contextualism, friendship is an important developmental context for children, which may have a bidirectional association with subjective well-being and loneliness. This bidirectional association has a long-term impact on children's development. To explore children's developmental trajectory, this study employed a longitudinal design spanning three years to examine the bidirectional relationship between children's friendship quality, subjective well-being, and loneliness.
A three-wave longitudinal study with one-year interval was conducted. A total of 212 third and fourth graders (Mage = 8.77 years, SD = .73; 59.4% male) in the central city completed questionnaires concerning friendship quality, subjective well-being, and loneliness. All the measures had good reliability and validity, and there was no serious common method bias. SPSS 25.0 was used to conduct descriptive statistics and correlation analysis. After completing item parceling, we constructed the random intercepts cross-lagged panel model (RI-CLPM) in Mplus 8.3 and employed the full information maximum likelihood estimation (FIML) to deal with missing data. The RI-CLPM allows to separate between-person effects from within-person effects. It was used to investigate whether children with higher friendship quality might subsequently have higher subjective well-being and lower loneliness, and vice versa.
The results of the correlation analysis showed that children's friendship quality was positively associated with subjective well-being, and negatively associated with loneliness, both concurrently and longitudinally (ps < .001). The results of RI-CLPM indicated that, at the between-person level, the random-intercept factor for friendship quality was positively related to that for subjective well-being (r = .55, p < .05), but negatively related to that for loneliness (r = -.74, p < .01). After controlling for between-person stability, the within-person paths showed that the relationship between friendship quality and loneliness was reciprocal. More specifically, initial friendship quality significantly negatively predicted loneliness at Time 2 (β = -.24, 95%CI [-.422, -.053]), which in turn negatively predicted friendship quality at Time 3 (β = -.22, 95%CI [-.426, -.004]). Additionally, loneliness at Time 1 also negatively predicted friendship quality at Time 2 (β = -.26, 95%CI [-.498, -.015]). However, between friendship quality and subjective well-being, subjective well-being at Time 1 can significantly predict friendship quality at Time 2 (β = .34, 95%CI [.059, .620]), whereas cross-lagged paths of the reversed direction were not significant (ps > .05).
This study reveals the bidirectional associations between friendship quality and loneliness at the within-person level by adopting a longitudinal design and constructing the RI-CLPM. This contributes to methodological innovation for the circular effects research orientation emphasized by development contextualism. Furthermore, it suggests that we can simultaneously intervene in both friendship quality and loneliness to construct a positive cycle. Meanwhile, the different relationships between friendship quality, subjective well-being, and loneliness indicate the importance of time in understanding individual development. In general, this study broadens the understanding of the specific connections and dynamic changes between friendship and the development of children's mental health.

Key words: children, friendship, loneliness, subjective well-being, random intercept cross-lagged panel model

摘要: 基于发展情境理论,对212名儿童进行为期三年的纵向研究,采用随机截距的交叉滞后模型考察友谊质量与主观幸福感和孤独感之间的双向关系。结果发现:(1)在个体间水平上,友谊质量分别与主观幸福感和孤独感之间存在显著相关;(2)在个体内水平上,T1的友谊质量显著预测T2的孤独感,T1的孤独感显著预测T2的友谊质量,T2的孤独感也显著预测T3的友谊质量;(3)在个体内水平上,T1的主观幸福感显著预测T2的友谊质量。综上,儿童的友谊质量与孤独感之间存在循环作用关系,而与主观幸福感之间仅存在T1主观幸福感预测T2友谊质量的单向关系。本研究拓展了友谊质量与心理健康相互作用关系的理论和干预思路。

关键词: 儿童, 友谊质量, 主观幸福感, 孤独感, 随机截距的交叉滞后模型