| 其他摘要 | With the continuous development of society and the improvement of education levels, parents and educators are paying more attention to children's academic performance. Many scholars have found that children's academic achievements are not determined by a single factor but are influenced by multiple factors, including children's intelligence, non-intelligence, family, school, and society. In recent years, it has indeed become increasingly common for adolescents to be affected by negative emotions such as anxiety and depression. The emotion-cognition interaction theory emphasizes the mutual influence between emotion and cognition. Emotions can affect students' cognitive processing, learning attention, learning motivation, the use of learning strategies, and interest in learning. However, negative emotions, mainly anxiety and depression, have a broader negative impact on academics, while children's personal factors also play different roles in academic performance. Studies have shown that a sense of agency is the cornerstone of human mental health and may play a fundamental role in emotions such as anxiety and depression. Whether the strength of the sense of agency also affects academic performance is one of the focuses of this study.
In addition to emotional factors, a considerable amount of research has explored the impact of family system factors on students' academic performance, including parenting styles, economic and social status, family structure, and family relationships. Family system theory suggests that a family is a system composed of family members with kinship, and there is an interactive relationship between each member. This study selects research variables such as parents' education level, parents' work status, sibling count, and learning leaders from family system factors to explore the relationship between family system factors and academics, as well as the combined effects of negative emotions and family system factors on academic performance. The paper conducts research from the following three aspects.
Study 1: The impact of negative emotions on academic performance was investigated. Firstly, 580 children from grades 4 to 6 in a Beijing primary school were selected as research subjects, and their academic performance, including Chinese, math, English, and the average scores of the three subjects, was used as the dependent variable of the study. Secondly, the subjects' depression scores, anxiety scores, and sense of agency scores were measured using the Children's Depression Inventory-Short Form (CES-DC), the Spence Children's Anxiety Scale-Short Version (SCAS-S), and the Chinese version of the Sense of Agency Scale (C-SoAS), respectively, supplemented by children's grade and gender as independent variables of the study. Thirdly, after preprocessing and descriptive statistics on the sample data, Pearson correlation analysis was used to explore the correlation between children's academic performance and various independent variables. The results showed that children's academic performance was positively correlated with sense of agency scores and grade level, negatively correlated with depression scores, and the correlation with anxiety scores was not significant. Finally, using the average score as a representative of academic performance, the decision tree method was used to identify the emotional factors that have a greater impact on academic performance, and the portrait characteristics of children with better academic performance were derived: higher grade level, lower depression level, and stronger sense of agency.
Study 2: The impact of family system factors on academic performance was investigated. Firstly, 179 fifth-grade children were selected from the subjects of Study 1 as the research objects, and their academic achievements, including Chinese, mathematics, English, and the average scores of the three courses, were taken as the dependent variables of the study. Data on parents' education level, number of siblings, learning leaders, parents' work status, and commuting time to school were collected as independent variables through the school's own database and questionnaire surveys. Secondly, after preprocessing and descriptive statistics on the samples, Pearson correlation analysis was used to explore the correlation between children's academic performance and various independent variables of family system factors. The results showed that children's academic performance was positively correlated with parents' education level and negatively correlated with the number of children in the family, while the correlation with parents' work status and commuting time to school was not significant. Subsequently, using the average score as a representative of academic performance, the decision tree method was used to identify the family system factors that have a major impact on children's academic performance, and the portrait characteristics of children with better academic performance were derived: higher parental education level and the father being the child's learning leader.
Study 3: The combined effects of children's negative emotional factors and family system factors on academic performance were investigated. Firstly, a decision tree model based on negative emotional factors and multiple family system factors was established to jointly predict academic performance, with the average score as a representative of academic performance. The results showed that the ten-fold cross-validation accuracy (89.3%) of the decision tree model combining negative emotions and family system factors was higher than the validation accuracy considering only emotional factors (82.8%) and only family system factors (82.6%). Secondly, using partial correlation analysis to control for family system factors, it was found that the correlation between emotional factors and academic performance remained basically unchanged, but the correlation coefficients with anxiety scores and depression scores increased relatively, while the correlation coefficients with grade level and sense of agency scores decreased relatively. These results indicate that after excluding family system factors, the impact of negative emotions on academic performance becomes more prominent, while the influence of grade level and sense of agency is relatively weakened. Again, through principal component analysis, it was further found that the variables in children's emotional factors and family system factors are independent of each other and cannot be arbitrarily eliminated. In summary, both children's emotional factors and family system factors have an important impact on children's academic performance.
This study provides an in-depth analysis of the relationship between academic performance and children's negative emotions and family system factors. The results show that academic performance is negatively correlated with depression scores and the number of children in the family, positively correlated with sense of agency, parental education level, and whether the child's learning is led by parents, while negating the correlation between academic performance and anxiety. The innovation of this study lies in the use of the decision tree method in machine learning to jointly analyze the impact of multiple variables on children's academic performance and comprehensively explore the combined effects of negative emotions and family system factors on academic performance. These findings provide important references for improving education quality and student academic performance. |
修改评论