Predicting employee turnover intention among healthcare and RMG workers: An integrated PLS-SEM, ANN, and fsQCA approach
Acta Psychologica · 2026-11 · 271 · 108126
DOI: 10.1016/j.actpsy.2026.108126 ↗Kurz gefasst
Fachartikel in der Zeitschrift Acta Psychologica, 2026-11. Verfasst von Hasin Mahtab Mahin und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Management Studies, Comilla University, Cumilla, Bangladesh. Electronic address: Mr.Mahin@stud.cou.ac.bd. Die Arbeit nennt 98 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study examines factors associated with employee turnover intention in Bangladesh's healthcare and ready-made garment (RMG) industries, two sectors vital to the national economy but facing labour instability linked to workplace stressors. Grounded in the health-impairment pathway of the Job Demands-Resources (JD-R) framework, it examines how job demands relate to burnout and turnover intention. We collected cross-sectional survey data from 311 full-time employees in the Dhaka and Chattogram divisions and analyzed them using partial least squares structural equation modeling (PLS-SEM), an artificial neural network (ANN), and fuzzy-set qualitative comparative analysis (fsQCA). Job stress, work-life imbalance, job insecurity, and burnout were each positively associated with turnover intention. In the pooled sample, burnout partially mediated the relationships between job stress, work-life imbalance, and job insecurity and turnover intention, although sector-level indirect effects differed. The three demands jointly explained only 8.9% of burnout variance, so most of it remains unexplained. The ANN identified burnout as the strongest predictor, with job stress and work-life imbalance similarly important and job insecurity the weakest; its predictive accuracy was comparable to, but not demonstrably better than, that of a linear benchmark. fsQCA identified three equifinal combinations of demands associated with high turnover intention, and no single condition was necessary. The results align with the JD-R health-impairment pathway; we did not examine job resources. Because the data are cross-sectional and measure turnover intention and not actual turnover, the findings offer tentative implications for workforce retention.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Management Studies, Comilla University, Cumilla, Bangladesh. Electronic address: Mr.Mahin@stud.cou.ac.bd.
- Department of Management Studies, Comilla University, Cumilla, Bangladesh. Electronic address: obaidulhoque88@stud.cou.ac.bd.
- Department of Management Studies, Comilla University, Cumilla, Bangladesh. Electronic address: mosharrafhossain@cou.ac.bd.
- Department of Public Administration, Comilla University, Cumilla, Bangladesh. Electronic address: mahin@cou.ac.bd.
- Department of Management, Jamalpur Science & Technology University, Jamalpur, Bangladesh. Electronic address: atiq.mgt@jstu.ac.bd.
- University of Manchester, UK. Electronic address: mds48@leicester.ac.uk.
Zitieren
APA
Mahin, H. M., Hoque, M. O., Hossain, M. M., Uddin, M. M., Rahaman, A., & Sarker, M. D. (2026). Predicting employee turnover intention among healthcare and RMG workers: An integrated PLS-SEM, ANN, and fsQCA approach. Acta Psychologica, 271, 108126. https://doi.org/10.1016/j.actpsy.2026.108126
Vancouver
Mahin HM, Hoque MO, Hossain MM, Uddin MM, Rahaman A, Sarker MD. Predicting employee turnover intention among healthcare and RMG workers: An integrated PLS-SEM, ANN, and fsQCA approach. Acta Psychologica. 2026;271:108126. doi:10.1016/j.actpsy.2026.108126
BibTeX
@article{Mahin2026,
author = {Mahin, Hasin Mahtab and Hoque, Md Obaidul and Hossain, Md Mosharraf and Uddin, Mohammed Mahin and Rahaman, Atiqur and Sarker, Md Dalim},
title = {Predicting employee turnover intention among healthcare and RMG workers: An integrated PLS-SEM, ANN, and fsQCA approach},
journal = {Acta Psychologica},
year = {2026},
volume = {271},
pages = {108126},
doi = {10.1016/j.actpsy.2026.108126}
}
EUVIMED