Abstract: Academic self-efficacy plays an important role in children's academic success and has obtained considerable attention from educational researchers and practitioners to date. The development of academic self-efficacy is not only the outcomes of temporary experiences but rather stems from a sustained accumulation of mastery experiences. Nevertheless, few studies have adopted a longitudinal dynamic perspective to explore the influencing factors of academic self-efficacy. Particularly, during the school-age years, a critical period for the development of children's self-regulation and changes of teacher-student relationship, self-regulation such as executive function develops rapidly in order to help children to effectively cope with new challenges, and meanwhile enduring positive teacher-student relationship also helps to enhance children's confidence in problem-solving. Given this, focusing on both individual (e.g., executive function) and external relationship (e.g., teacher-student relationship) factors and deeply exploring their unique and interactive effects on academic self-efficacy from a longitudinal perspective was considered important and necessary. Therefore, we conducted a longitudinal study across six months (3 months apart) to examine the initial levels and growth rates of executive function and teacher-student relationship, and their unique and interactive effects on children's academic self-efficacy.
A total of 523 children in grades 3 to 5(51% boys, mean age 10.36 years) participated in this study to complete self-report questionnaires. All participants were tested on the Behavior Rating Inventory of Executive Function and Teacher-student relationship questionnaire for three times, and the academic Self-Efficacy Questionnaire fat T3. Data analyses were performed in three steps: (1) Descriptive analyses and repeated measures ANOVA were computed using the SPSS 21.0 to examine gender differences, differences between only children and those with siblings, associations among the measured variables, and initially evaluate the observed changes; (2) An unconditional multivariate linear latent growth model was estimated for the growth trajectories of executive function and teacher-student relationship; (3) The latent moderated structural equation models were used to examine the unique and interactive effects of executive function and teacher-student relationship on academic self-efficacy.
The results indicated that: (1) On average, children experienced an increase in executive function and a decrease in relationship with their teachers over six months; (2) Noth the intercepts and slopes of executive function (β= .43, β = .33, ps. < .01) and teacher-student relationship (β= .39, β = .27, ps. < .01) positively predicted subsequent academic self-efficacy; and (3) Noth the intercept and slope of the teacher-student relationship could moderate the relation between the intercept (not the slope) of executive function and subsequent academic self-efficacy(β= .11, β = .13, ps. < .01). Specifically, compared to those children with poor initial levels or a faster rate of decline in teacher-student relationship, the initial level of executive function could be more strongly positively predicted later academic self-efficacy for those children with better initial levels or a slower rate of decline or even an increasing rate in the teacher-student relationship.
One implication of this study is to extend the conventional interaction paradigm beyond looking at predictors as static variables. We employ a strategy for exploring the dynamic interplay between trajectories of executive function and the teacher-student relationship on later academic self-efficacy. This approach, in contrast to traditional cross-sectional analysis, captures the intricate and evolving processes involved. With respect to educational practices, the findings highlight the importance of regular evaluations and the value of nurturing high-quality teacher-student relationships and providing continuous executive functioning training to facilitate academic efficacy for children. Furthermore, the longitudinal interaction effect reveals that enhancing the quality of the teacher-student relationship can amplify the association between high levels of executive functioning and high levels of academic self-efficacy. This suggests that teachers' encouragement and support based on children's self-regulation could be beneficial for children's learning efficacy. The present research holds substantive theoretical and practical implications for school education.

Key words: academic self-efficacy, executive function, teacher-student relationship, longitudinal interactive effect

摘要： 为揭示儿童执行功能和师生关系的变化轨迹对学业自我效能感的独特效应和交互效应,对523名小学三至五年级儿童进行了6个月的追踪测量。结果表明：(1) 6个月内,儿童的执行功能有所增长,而师生关系质量有所下降;(2)执行功能和师生关系的初始水平和发展速度均能正向预测学业自我效能感;(3)师生关系初始水平和变化速度与执行功能初始水平而非发展速度之间存在交互效应。研究结果有助于揭示个体与环境因素随时间交互影响学业自我效能感的复杂机制,对提升学龄儿童学业自我效能感具有实践指导意义。

关键词: 学业自我效能感, 执行功能, 师生关系, 纵向交互效应