Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Mar 14;25(1):380.
doi: 10.1186/s12913-025-12552-9.

Analysis of the factors influencing the salary level and satisfaction of medical staff in hospitals in less developed areas of Western China based on machine learning algorithms: evidence from Guangxi Zhuang Autonomous Region

Xinyi Xu  1 , Qisheng Gao  2 , Tengyan Wu  3 , Wenhui Zhuang  1 , Yejia Mo  1 , Wenjun Zhang  1 , Bo Wei  3 , Zhong Tang  4 , Pinghua Zhu  5
Affiliations

Analysis of the factors influencing the salary level and satisfaction of medical staff in hospitals in less developed areas of Western China based on machine learning algorithms: evidence from Guangxi Zhuang Autonomous Region

Xinyi Xu et al. BMC Health Serv Res. .

Abstract

Background: Compensation plays a critical role in motivating staff and enhancing operational performance and human resource costs in hospitals. This study was aimed at investigating pay levels and the key factors influencing pay satisfaction in secondary and tertiary public hospitals in Guangxi.

Methods: Questionnaires were distributed to 48 hospitals across 14 prefecture-level cities in Guangxi. Information on personal characteristics, salary levels, work situations and perceptions of current salary conditions was provided by 10,343 staff in secondary and tertiary hospitals. Five machine learning models were employed to identify the most significant influencing factors of salary satisfaction in Guangxi public hospitals.

Results: Overall, the actual total after-tax income in secondary public hospitals in Guangxi ranged from $466.55-$744, while the income of staff in municipal-level tertiary public hospitals ranged from $5,001 to $1,041.75 per month. Among the five models, the support vector machine (SVM) demonstrated the best performance in analyzing the influencing factors of compensation satisfaction. The most influential factors for total compensation satisfaction included the extent to which compensation reflected labor value, salary increases since 2017 compared to peer hospitals, total after-tax income and the difference in compensation between staff within and outside the establishment of hospitals. Satisfaction with salary growth aligned closely with the factors influencing overall compensation satisfaction. Satisfaction with pay equity was also influenced by the ability of salary gaps between different positions to reflect differential effort.

Conclusions: A relatively low pay level in secondary hospitals in Guangxi was revealed. The factors influencing satisfaction with total pay, pay fairness and pay growth since 2017 varied. SVM outperformed other models in the analysis of the factors influencing pay satisfaction. To enhance pay satisfaction in secondary and tertiary hospitals in Guangxi, it is crucial to establish a salary distribution system aligned with the value of labor across different positions and tailored to the unique characteristics of each hospital.

Keywords: Machine learning; Personnel and salary reform; Public hospital; Salary satisfaction.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: This study was approved by the Research Ethics Committee of Guangxi Medical University (approval number: KY20240323). A statement explaining the informed consent to participate was included on the opening page of the survey. The participants were only able to complete the online questionnaire after accepting the terms of consent. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Variable importance of satisfaction with total pay, satisfaction with fairness of pay distribution, and satisfaction with pay growth since 2017 using SVM. Note: A represents the variable importance associated with satisfaction regarding total pay, B represents the variable importance associated with satisfaction regarding the fairness of pay distribution, and C represents the variable importance associated with satisfaction regarding pay growth since 2017, as determined using SVM
Fig. 2
Fig. 2
The six primary factors influencing total pay satisfaction, visualized using the iml package. Note: The variable “Salary Reflect Value” is divided into four levels:1 - Fully reflects work value, 2 - Partially reflects, 3 - Barely reflects, 4 - Does not reflect at all. The variable “The Salary Increased Since 2017” is divided into four levels:1 - Significant increase, 2 - Moderate increase, 3 - Slight increase, 4 - No increase. The variable “Total Income After Tax” is divided into five levels:1 - Below $466.40, 2 - 466.55 -$744, 3 - 744.15 - $1041.6, 4 - 1041.75 - $1339.2, 5 - Above $1339.35. The variable “Formation Influence” is divided into five levels:1 - Very significant, 2 - Quite significant, 3 - Moderate, 4 - Not very significant, 5 - No influence
Fig. 3
Fig. 3
The five primary factors influencing satisfaction with fairness of pay distribution. Note: The variable “Salary Reflect Value” is divided into four levels: 1 - Fully reflects work value, 2 - Partially reflects work value, 3 - Barely reflects work value, 4 - Does not reflect work value at all. The variable “Salary Gap Between Positions” is divided into four levels: 1- Fully reflects labor income, 2 - Mostly reflects labor income, 3 - Barely reflects labor income, 4 - Does not reflect labor income at all. The variable “The Salary Increased Since 2017” is divided into four levels: 1 - Significant increase, 2 - Moderate increase, 3 - Slight increase, 4 - No increase
Fig. 4
Fig. 4
The six primary factors influencing satisfaction with salary increases since 2017. Note: The variable “The Salary Increased Since 2017” is divided into four levels: 1 - Significant increase, 2 - Moderate increase, 3 - Slight increase, 4- No increase. The variable “Salary Reflect Value” is divided into four levels: 1 - Fully reflects work value, 2 - Partially reflects work value, 3 - Barely reflects work value, 4 - Does not reflect work value at all. The variable “Total Income After Tax” is divided into five levels: 1 - Below $446.40, 2 - 466.55 - $744, 3 - 744.15 - $1041.6, 4 - 1041.75- $1339.2, 5 - Above $1339.35
See this image and copyright information in PMC

References

    1. Lu Y, Hu X, Huang X, Zhuang X, Guo P, Feng LF, et al. The relationship between job satisfaction, work stress, work–family conflict, and turnover intention among physicians in Guangdong, China: a cross-sectional study. BMJ Open. 2017;7(5):e014894. - PMC - PubMed
    1. Wang Z, Xie Z, Dai J, Zhang L, Huang Y, Chen B. Physician burnout and its associated factors: A Cross-sectional study in Shanghai. J Occup Health. 2014;56(1):73–83. - PubMed
    1. Alam A, Asim M. Relationship between job satisfaction and turnover intention. Int J Hum Resource Stud. 2019;9(2):163.
    1. Li C, Wang Q, Ren Q, Lao X, Zhu P. Study on Spatial-temporal differentiation of health human resources in Guangxi. Health Econ Res. 2021;38(2):62–6.
    1. Niu A, Li P, Duan P, Ding L, Xu S, Yang Y, et al. Professional quality of life in nurses on the frontline against COVID-19. J Nurs Adm Manag. 2022;30(5):1115–24. - PMC - PubMed