Exercise-related tendencies, internet addiction, academic stress, and subjective sleep quality among junior high school students

Anli Ai¹, Huijuan Wang², Yang Wu^3,*, Kelei Guo^2,*, Chao Ma⁴, Dong Li²
1 School of Physical Education, Shaoguan University, Shaoguan, China
2 School of Physical Education and Health, Zhaoqing University, Zhaoqing, China
3 School of Recreational Sports and Tourism, Beijing Sport University, Beijing, China
4 Department of Physical Education, Zhanjiang Preschool Teachers College, Zhanjiang, China
* Corresponding Author: Yang Wu. Email: email ; Kelei Guo. Email: email

Journal of Psychology in Africa https://doi.org/10.32604/jpa.2026.082754

Received 22 March 2026; Accepted 09 June 2026; Published online 18 August 2026

Download PDF

Abstract

This study examined the associations between physical exercise-related tendencies and subjective sleep quality and the roles of Internet addiction and academic stress in this relationship. A self-report survey was conducted with 740 Chinese junior high school students (54.5% female; mean age = 13.58 years, SD = 1.05). Path analysis indicated that physical exercise-related tendencies were positively associated with subjective sleep quality and negatively associated with Internet addiction and academic stress. Academic stress was negatively associated with subjective sleep quality, whereas Internet addiction was not independently associated with subjective sleep quality after accounting for academic stress. Bootstrapped indirect-effect analysis showed a significant indirect association through academic stress and a significant sequential indirect association through Internet addiction and academic stress, but the independent indirect association through Internet addiction was not significant. From a stress-related behavioural perspective, these findings suggest that adolescent sleep-related research and school health practice may benefit from considering exercise-related tendencies, problematic Internet use, and academic stress together. Findings should be interpreted as cross-sectional associations rather than causal effects.