De Laet Steven, Colpin Hilde, Vervoort Eleonora, Doumen Sarah, Van Leeuwen Karla, Goossens Luc, Verschueren Karine
University of Leuven.
Dev Psychol. 2015 Sep;51(9):1292-306. doi: 10.1037/a0039478. Epub 2015 Jul 20.
The present longitudinal study examined how relationships with teachers and peers jointly shape the development of children's behavioral engagement in late elementary school. A sample of 586 children (46% boys; Mage = 9.26 years at Wave 1) was followed throughout Grades 4, 5, and 6. A multidimensional approach was adopted, distinguishing support and conflict as teacher-child relationship dimensions, and acceptance and popularity as peer relationship dimensions. Additive, moderation, and mediation models were tested. Latent growth curve modeling showed evidence for an additive model in which high initial and increasing levels of teacher support, and high initial levels of peer acceptance, independently reduce the normative declines in children's behavioral engagement. This implies that targeting only 1 relationship in intervention cannot compensate for negative aspects of the other relationship. Teacher conflict only predicted initial levels of behavioral engagement, whereas peer popularity did not predict behavioral engagement (not even in a subsample of children with relatively high levels of relational or physical aggression). However, cross-lagged panel mediation analyses revealed that children who were perceived as more popular in Grade 5 were less engaged in school in Grade 6. Practical implications of these findings are discussed. (PsycINFO Database Record
本纵向研究考察了与教师和同伴的关系如何共同塑造小学高年级儿童行为参与度的发展。对586名儿童（46%为男孩；第1波时的年龄中位数为9.26岁）进行了四年级、五年级和六年级的全程跟踪研究。采用了多维方法，将支持和冲突作为师生关系维度，将接纳和受欢迎程度作为同伴关系维度。对累加模型、调节模型和中介模型进行了检验。潜在增长曲线模型显示了累加模型的证据，即高水平的初始教师支持以及不断增加的教师支持，和高水平的初始同伴接纳，能独立减少儿童行为参与度的正常下降。这意味着在干预中只针对一种关系无法弥补另一种关系的负面影响。教师冲突仅能预测行为参与度的初始水平，而同伴受欢迎程度则无法预测行为参与度（即使在关系攻击或身体攻击水平相对较高的儿童子样本中也是如此）。然而，交叉滞后面板中介分析显示，在五年级被认为更受欢迎的儿童在六年级时对学校活动的参与度较低。讨论了这些发现的实际意义。（《心理学文摘数据库记录》）
