Work-and learning-related physical activity configurations and their associations with psychological well-being and mobile phone dependence among Chinese university students: a latent profile analysis
- 1School of Physical Education, Yancheng Institute of Technology, Yancheng, Jiangsu China
- 2Postdoctoral Research Station, College of Physical Education, Yangzhou University, Yangzhou, Jiangsu China
Abstract
Background: Physical activity accumulated in work or learning contexts may differ from leisure-time exercise in its purpose, autonomy, and psychological meaning. University students may also combine vigorous activity, moderate-intensity activity, and walking in heterogeneous ways that are obscured by total-volume measures. This study identified work- and learning-related physical activity configurations and examined their cross-sectional associations with psychological well-being and mobile phone dependence.
Methods: The available de-identified dataset contained 10,846 records from 47 normalized universities in China. The primary latent profile analysis included 8,507 students with complete vigorous-activity, moderate-intensity-activity, and walking indicators within the adapted work/learning domain. MET-min/week values were calculated using audited International Physical Activity Questionnaire scoring procedures, transformed as ln(x + 1), and standardized. One- through eight-profile tied-diagonal Gaussian mixture models were compared using information criteria, classification quality, random-start replication, parametric bootstrap likelihood-ratio tests, and resampling stability. Psychological well-being and mobile phone dependence were compared using the Bolck–Croon–Hagenaars method and 500 posterior pseudo-class draws, with adjustment for sex, age, grade, major, university type, ethnicity, and university clustering (N = 8,430; 46 clusters).
Results: The eight-profile solution was retained, with high classification accuracy (entropy = 0.977; minimum average posterior probability = 0.944) and strong bootstrap and subsample stability. The profiles comprised all three indicators zero (13.39%), moderate-intensity activity only (1.58%), vigorous activity present with walking zero (2.20%), walking only (20.74%), moderate-intensity activity plus walking without vigorous activity (21.98%), vigorous activity plus walking without moderate-intensity activity (3.20%), all three indicators present at lower relative volumes (18.40%), and all three indicators present at higher relative volumes (18.53%). Adjusted overall differences in psychological well-being were statistically detectable, χ²(7) = 19.40, p = 0.007, but the maximum standardized difference was very small (|d|=0.162). Mobile phone dependence also differed overall, χ²(7) = 22.39, p = 0.002; the largest adjusted contrast was 3.42 points between the vigorous-plus-walking/no-moderate profile and the all-three-higher-volume profile (95% CI 1.69 to 5.16; d = 0.236; Holm-adjusted p = 0.003).
Conclusions: Work- and learning-related physical activity configurations were heterogeneous and reproducible, but their associations with psychological well-being and mobile phone dependence were small. The profiles should be interpreted as descriptive combinations of domain-specific activity rather than overall physical activity levels, diagnostic groups, or evidence of causal effects. Their classification alone does not provide a sufficient basis for allocating individualized psychological or behavioural interventions.
Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s40359-026-05352-1.
Related
- Nonlinear associations between physical activity, sleep quality, and psychological distress in university students: a cross-sectional interpretable machine learning study
- Associations of psychological safety and perceptions of organizational environment among healthcare workers: a cross-sectional study in a university hospital
- Associations Between Psychological Factors, Routine Hematological and Biochemical Indicators, and Non-Suicidal Self-Injury Among University Students: An Exploratory Cross-Sectional Study.
- Perceived academic demands and psychological distress among Chinese university students: structural associations involving chronic time pressure and leisure-time physical activity.
- Evaluating the associations of learning environment with student engagement and learning gains among undergraduate students at a medical university in western China
BibTeX
@article{luo2026workand,
title = {{Work-and learning-related physical activity configurations and their associations with psychological well-being and mobile phone dependence among Chinese university students: a latent profile analysis}},
author = {Wei Luo and Kunyan Wang},
journal = {BMC Psychology},
year = {2026},
month = aug,
doi = {10.1186/s40359-026-05352-1},
url = {https://doi.org/10.1186/s40359-026-05352-1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Work-and learning-related physical activity configurations and their associations with psychological well-being and mobile phone dependence among Chinese university students: a latent profile analysis
AU - Wei Luo
AU - Kunyan Wang
PY - 2026
DA - 2026/08/12
JO - BMC Psychology
DO - 10.1186/s40359-026-05352-1
UR - https://doi.org/10.1186/s40359-026-05352-1
AB - Background: Physical activity accumulated in work or learning contexts may differ from leisure-time exercise in its purpose, autonomy, and psychological meaning. University students may also combine vigorous activity, moderate-intensity activity, and walking in heterogeneous ways that are obscured by total-volume measures. This study identified work- and learning-related physical activity configurations and examined their cross-sectional associations with psychological well-being and mobile phone dependence. Methods: The available de-identified dataset contained 10,846 records from 47 normalized universities in China. The primary latent profile analysis included 8,507 students with complete vigorous-activity, moderate-intensity-activity, and walking indicators within the adapted work/learning domain. MET-min/week values were calculated using audited International Physical Activity Questionnaire scoring procedures, transformed as ln(x + 1), and standardized. One- through eight-profile tied-diagonal Gaussian mixture models were compared using information criteria, classification quality, random-start replication, parametric bootstrap likelihood-ratio tests, and resampling stability. Psychological well-being and mobile phone dependence were compared using the Bolck–Croon–Hagenaars method and 500 posterior pseudo-class draws, with adjustment for sex, age, grade, major, university type, ethnicity, and university clustering (N = 8,430; 46 clusters). Results: The eight-profile solution was retained, with high classification accuracy (entropy = 0.977; minimum average posterior probability = 0.944) and strong bootstrap and subsample stability. The profiles comprised all three indicators zero (13.39%), moderate-intensity activity only (1.58%), vigorous activity present with walking zero (2.20%), walking only (20.74%), moderate-intensity activity plus walking without vigorous activity (21.98%), vigorous activity plus walking without moderate-intensity activity (3.20%), all three indicators present at lower relative volumes (18.40%), and all three indicators present at higher relative volumes (18.53%). Adjusted overall differences in psychological well-being were statistically detectable, χ²(7) = 19.40, p = 0.007, but the maximum standardized difference was very small (|d|=0.162). Mobile phone dependence also differed overall, χ²(7) = 22.39, p = 0.002; the largest adjusted contrast was 3.42 points between the vigorous-plus-walking/no-moderate profile and the all-three-higher-volume profile (95% CI 1.69 to 5.16; d = 0.236; Holm-adjusted p = 0.003). Conclusions: Work- and learning-related physical activity configurations were heterogeneous and reproducible, but their associations with psychological well-being and mobile phone dependence were small. The profiles should be interpreted as descriptive combinations of domain-specific activity rather than overall physical activity levels, diagnostic groups, or evidence of causal effects. Their classification alone does not provide a sufficient basis for allocating individualized psychological or behavioural interventions. Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s40359-026-05352-1.
SN - 2050-7283
ER -