AUTHOR=Zheng Zhanyue , Kang Cheng , Wang Chengqiang , Li You , Sun Yan 

TITLE=Machine learning-based multidimensional evaluation of the effectiveness of course civics teaching: a case study of the occupational health and occupational medicine course

JOURNAL=Frontiers in Education

VOLUME=Volume 10 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/education/articles/10.3389/feduc.2025.1649013

DOI=10.3389/feduc.2025.1649013

ISSN=2504-284X

ABSTRACT=IntroductionThis study investigated the integration of Course-based Ideological and Political Education (CIPE) into Occupational Health and Occupational Medicine (OHOM) courses within the “Internet Plus” context. The aim was to evaluate its impact on students' academic performance, professional ethics, and social responsibility.MethodsA total of 230 questionnaires were distributed to senior undergraduate students at Guilin Medical University, with 220 valid responses collected (response rate: 95.6%). Data were analyzed using Python 3.11 and R 4.1.1. Statistical methods included descriptive statistics, t-tests/non-parametric tests, chi-square tests, and multiple linear regression with adjustment for gender and age. Machine learning methods (XGBoost, random forests, and support vector regression) were combined with five-fold cross-validation and SHAP analysis for model optimization and interpretation.ResultsStudents in the CIPE-integrated group achieved significantly higher composite scores (83.90 ± 3.08) than the traditional group (82.66 ± 4.28, p < 0.05). Regression analysis identified course returns (β = 0.54, 95% CI: 0.16–0.92), group participation (β = 0.44, 95% CI: 0.03–0.84), online resource utilization (β = 0.46, 95% CI: 0.05–0.87), and ethical benefits (β = 0.52, 95% CI: 0.12–0.92) as significant predictors of performance. SHAP analysis highlighted the critical roles of group participation, course motivation, ethical benefits, and note review, while also revealing individual differences in learning behaviors and value formation.DiscussionThe deep integration of CIPE into OHOM courses through blended teaching significantly enhanced students' academic outcomes, professional ethics, and social responsibility. The combined use of traditional statistical methods and interpretable machine learning provided robust evidence for evaluating educational interventions and offered methodological guidance for extending CIPE applications in preventive medicine and public health curricula.