Predicting academic achievement from the collaborative influences of executive function, physical fitness, and demographic factors among primary school students in China: ensemble learning methods

Abstract

Background: Elevated levels of executive function and physical fitness play a pivotal role in shaping future quality of life. However, few studies have examined the collaborative influences of physical and mental health on academic achievement. This study aims to investigate the key factors that collaboratively influence primary school students' academic achievement from executive function, physical fitness, and demographic factors. Additionally, ensemble learning methods are employed to predict academic achievement, and their predictive performance is compared with individual learners.

Methods: A cluster sampling method was utilized to select 353 primary school students from Huai'an, China, who underwent assessments for executive function, physical fitness, and academic achievement. The recursive feature elimination cross-validation method was employed to identify key factors that collaboratively influence academic achievement. Ensemble learning models, utilizing eXtreme Gradient Boosting and Random Forest algorithms, were constructed based on Bagging and Boosting methods. Individual learners were developed using Support Vector Machine, Decision Tree, Logistic Regression, and Linear Discriminant Analysis algorithms, followed by the establishment of a Stacking ensemble learning model.

Results: Our findings revealed that sex, body mass index, muscle strength, cardiorespiratory function, inhibition, working memory, and shifting were key factors influencing the academic achievement of primary school students. Moreover, ensemble learning models demonstrated superior predictive performance compared to individual learners in predicting academic achievement among primary school students.

Conclusions: Our results suggest that recognizing sex differences and emphasizing the simultaneous development of cognition and physical well-being can positively impact the academic development of primary school students. Ensemble learning methods warrant further attention, as they enable the establishment of an accurate academic early warning system for primary school students.

Keywords: Academic achievement; Collaborative influences; Demographic information; Ensemble learning methods; Executive function; Physical fitness.

Fig. 1

The principle of Stacking ensemble learning model Note. SVM is Support Vector Machine; DT is Decision Tree; LDA is Linear Discriminant Analysis; LR is Logistic Regression; MC is Meta-Classifier

Fig. 2

The results of correlation analysis for factors Note. Incongruent_RT and Incongruent_ACC are the mean reaction time and accuracy in incongruent trials, respectively. Congruent_RT and Congruent_ACC are the mean reaction time and accuracy in congruent trials. Homogeneous_RT and Homogeneous_ACC are the mean reaction time and accuracy in homogeneous trials. Heterogeneous_RT and Heterogeneous_ACC are the mean reaction time and accuracy in heterogeneous trials. 1-back_RT and 1-back_ACC are the mean reaction time and accuracy in 1-back task

Fig. 3

The best factor subset and feature importance ranking of each algorithm Note. a represents the eXtreme Gradient Boosting algorithm; b is the Random Forest algorithm; c is the Support Vector Machine algorithm; d is the Decision Tree algorithm; e is the Logistic Regression algorithm; f is the Linear Discriminant Analysis algorithm. Blue color indicates the factors selected as key, while gray indicates exclusion. Due to space constraints, all factor names couldn't be labeled in the diagram. The factors in each subfigure are in the same order: sex, grade, body mass index, sit and reach, standing long-jump, push-up, sit-up, 50-m sprint, vital capacity, 50-m × 8 shuttle run, mean reaction time in congruent trials, mean accuracy in congruent trials, mean reaction time in homogeneous trials, mean reaction time in heterogeneous trials, mean accuracy in homogeneous trials, mean accuracy in heterogeneous trials, mean reaction time in 1-back task, and mean accuracy in 1-back task

Fig. 4

The performance of machine learning models in test samples Note. XGB is eXtreme Gradient Boosting model; RF is Random Forest model; SVM is Support Vector Machine model; DT is Decision Tree model; LR is Logistic Regression model; LDA is Linear Discriminant Analysis model; Stacking is Stacking ensemble learning model