幼兒執行功能、社會能力與情緒能力之關係──中介效果檢驗

Relationships among Executive Function, Social Competence, and Emotional Competence of Young Children: A Mediating Effect Analysis

本研究旨在探討臺灣的幼兒執行功能及其情緒能力和社會能力之間的關聯。依據文獻分析結果，提出情緒能力和社會能力作為中介變量的假設模型。資料來自「臺灣幼兒發展調查資料庫」（Kids in Taiwan: National Longitudinal Study of Child Development & Care, KIT）中48月齡的幼兒為研究對象。KIT資料庫中共蒐集1,931位幼兒，當中男孩980名（50.8%）、女孩951名（49.2%）。由學齡前幼兒之主要照顧者接受結構式晤談，即依據「執行功能」、「情緒能力」、「社會能力」等三份量表的題目逐一回答。三份量表之數據採SPSS 24.0進行變量之相關分析；由於資料 是非常態分配，所以用Smart PLS 3.0統計軟體，分析所提出的二個假設中介模型。分析結果發現H_1：幼兒的情緒能力在執行功能與社會能力之間具有中介效果；H_2：幼兒的社會能力在執行功能與情緒能力之間具有中介效果等二條假設都獲支持，並且也發現：社會能力的中介效果的解釋力較高。就此發現，建議：1.家長應注重幼兒情緒能力、社會能力。2.應從整體性觀點深入分析影響幼兒心理發展因素的關聯模型。3.情緒能力與社會能力因果關係再確認。

Executive function, social competence, and emotional competence are critical factors that affect the physical and mental development of young children. Executive function refers to the cognitive process that consists of a set of interrelated cognitive activities that are responsible for completing a set of purposeful and targeted thoughts and behavior, including preplanning, the flexible transformation of thoughts and behaviors, and inhibition of external interference and inappropriate reactions. Executive function can improve the emotional competence of preschoolers to control behavior, maintain attention, remember rules, regulate emotions, and restrain impulsive behavior. Executive function also affects aspects of social competence, such as cognitive flexibility and working memory, which are components of executive function. Ciairano et al. (2007) observed that children with high cognitive resilience have higher cooperative behavior. McQuade et al. (2013) found that working memory is positively associated with social competence. Emotional competence refers to the self-efficacy of developing and manifesting emotions through interactions in a social situation (Saarni, 1999). Social competence in young children refers to the use of cognition and behavior to adapt to social culture, interact with others, and meet social expectations. Emotional competence affects social competence, and vice versa. To be socially competent, preschoolers must balance the needs of themselves and others in a social environment by employing four traits: independence, assertiveness, sociability, and compliance. Executive function can improve emotional competence and social competence. The emotional competence of young children can mediate the relationship between executive function and social competence, and social competence can mediate the relationship between executive function and emotional competence. In the present study, we searched for the keywords 'executive function,' 'social competence,' and 'emotional competence' in databases such as Airiti Library and Google Scholar. We found that no study has analyzed the mediating effect of social competence and emotional competence. We proposed two hypotheses. First, the emotional competence of young children would mediate the relationship between executive function and social competence (H_1). Second, the social competence of young children would mediate the relationship between executive function and emotional competence (H_2). The present study sought to verify the mediating effect of emotional competence and social competence and address research gaps. To verify the hypotheses, we selected and revised three scales (N = 100) to conduct a questionnaire survey. Confirmatory factor analysis results indicated that the scales for executive function, social competence, and emotional competence had a satisfactory composite reliability, average variance extracted, and discriminant validity. Data were collected from 1,931 young children aged 48 months by using the Kids in Taiwan: National Longitudinal Study of Child Development & Care database (Chang, 2021). Structured interviews were conducted with the primary caregivers of the young children, and responses were collected using the three scales. Among the children, 980 (50.8%) were boys and 951 (49.2%) were girls. Data were analyzed using structural equation modeling (SEM) to verify the hypothetical mediating models. First, we tested the common method variance (CMV), which is the bias caused by a research method that inflates or deflates the correlation between two variables. The result indicated that the CMV was marginal. Second, Amos (International Business Machines, New York, NY, USA) was adopted for SEM, but the model fit indices indicated poor goodness-of-fit. The multivariate normality test results were significant, which suggested that the collected data had a nonnormal distribution. Therefore, we used partial least squares SEM (PLS-SEM), a variance-based SEM that is not limited to normal distributions. Third, the mediating effect process was analyzed using the method of Hair et al. (2016), which consists of three steps. Step 1: We verified whether the path coefficient of the independent variable that predicts the dependent variable was significant. Step 2: We confirmed that the path coefficients of the independent variable that predicts the mediator and those of the mediator that predict the dependent variable were significant. Step 3: If the variables in Steps 1 and 2 were significantly correlated, we analyzed the mediation model. In the present study, the variables in Steps 1 and 2 were significantly correlated; thus, the mediation model was analyzed. To analyze the mediation model, the model fitness was first verified. The results indicated that the factor loading, composite reliability, average variance extracted, and discriminant validity all met the relevant criteria and that the measurement model (called the 'outer model' in PLS-SEM) was appropriate. The values for collinearity, R^2, f^2, and Q^2 all conformed to the standard of the structural model (called the 'inner model' in PLS-SEM); that is, the structural model was appropriate. Because the measurement and structural model fit, we further analyzed all the paths in the two mediating models and the direct, indirect, and total effects. The results supported H_1 and H_2. Therefore, the emotional competence of young children mediates the relationship between executive function and social competence, and the social competence of young child mediates the relationship between executive function and emotional competence. We further tested whether the mediating effect was partial or full by evaluating the variance accounted for (VAF), which is calculated by dividing the indirect effect by the total effect. According to Hair et al. (2016), a percentage of VAF of > 80%, 20%-80%, and < 20% represents full mediation, partial mediation, and no mediation, respectively. Our results demonstrated that the VAF of emotional competence was 47.11% and that of social competence was 61.83%. Thus, emotional and social competence in the mediating model both have a partial mediating effect. We subsequently verified whether this partial mediating effect was significant. We used the built-in bootstrapping function of Smart PLS and set the number of bootstrap samples to 5,000. The results indicated that social and emotional competence significantly mediated the relationship. According to our findings, we first propose that parents, educators, and policymakers should pay attention to the psychological characteristics of children's executive function, social competence, and emotional competence and strengthen the executive function of young children to enable them to develop their emotional and social competence. Second, the model of factors influencing the social and emotional competence of young children must be analyzed holistically. Data on language development, social and emotional development, physical movement development, and family are investigated in the database; thus, our findings and additional data can be combined and analyzed. Third, the causal model between emotional and social competence must be verified. This study sampled 48-month-old children and found that emotional and social competence both have a mediating effect and were predictors of one another. Finally, future studies can include data from a variety of information providers (e.g., teachers) to evaluate children's executive function and emotional and social abilities in various contexts.