Considerations beyond salary: study of job satisfaction among Chinese social work practitioners in different positions

ABSTRACT

In the context of the brain drain of social work practitioners in China, studying their job satisfaction and the influencing factors is vital. This study, focusing on non-material factors, surveyed 981 frontline social workers and 390 agency managers in Beijing, Shanghai, Guangzhou, and Shenzhen. Findings show differences in work independence and evaluation mechanisms between frontline workers and managers. The OLS model reveals that public recognition, career advancement, and interpersonal relationships significantly affect frontline worker job satisfaction while interpersonal relationships do not impact managers’ job satisfaction. Emphasising the value of the social work profession is crucial to retain the talent.

KEYWORDS:

• Job satisfaction
• public recognition
• work independence
• frontline social workers
• agency managers

1. Introduction

Over the past few decades, China has witnessed significant economic and social transformation since the reform and opening up policies initiated in 1978. As a result, the landscape of social needs has evolved, necessitating a robust social work sector. While government-led initiatives were the primary problem-solving mechanism historically, they have proven insufficient to address the dynamic challenges of today. Since 2010, China has strategically prioritises the growth of a skilled social work workforce through policy formulations aimed at professionalisation and vocationalisation. Despite these efforts, critical issues persist. Notably, there is a substantial loss of social work talent and practitioners often grapple with low self-efficacy and job satisfaction. According to data provided by the Shenzhen Social Work Association, the rate of turnover among social workers in Shenzhen, known to have the most advanced development within the social work industry in China, has increased annually from 2008 to 2014, reaching 22.2% in 2014. Although turnover decreased between 2015–2016, it rose again to 18.1% in 2017 (Xu, 2019). The brain drain phenomenon remains a pressing concern.

While existing studies predominantly focus on the influence of salary and benefits (Xi & Lei, 2014), we recognise that broader factors impact job satisfaction. Therefore, this study aims to explore what other factors affect the job satisfaction of social work practitioners and whether these factors differ in their impact on different positions. In this way, we aim to propose targeted interventions that extend beyond financial incentives, ultimately enhancing job satisfaction and ensuring the stability of China’s social work industry.

2. Literature review and research hypotheses

2.1. Definitions of job satisfaction

The study of job satisfaction originated with the Hawthorne experiment, which suggested that worker attitudes towards work directly affected productivity and that the key to job attitudes was workers’ satisfaction with their jobs. In 1935, Robert Hoppock first clearly proposed the definition of job satisfaction as the employees’ feelings about both the physical and psychological aspects of work and a subjective response to the work situation (Kamdron, 2010). Currently, there are three main academic definitions of job satisfaction. The first is the dimension definition: the focus is on the employees’ emotional responses to their work contents, that is, employee perceptions or attitudes towards factors such as the job, organisation, work environment, leadership style, benefits and promotions, and interpersonal relationships (Carbonel & Claudette, 2007), while less attention is paid to the processes that shape job satisfaction. The second is the reference-type definition: job satisfaction depends not only on how satisfied employees are with job-related elements but also on the measurement of the input and output in the job (Tong & Zhang, 2010). If the results are shown to be balanced, the employees will feel fair and satisfied; on the contrary, if the employees think that their input in the job is far beyond the reward, it reduces their job satisfaction. The third is the expectation-based definition: job satisfaction is the gap between actual and expected value in the particular work environment people specialise in (Wyrwa & Kaźmierczyk, 2020). A smaller gap indicates higher employee satisfaction; conversely, a larger gap between value and expectation suggests lower satisfaction.

This study aligns with the first definition, emphasising employees’ subjective perceptions and attitudes towards the content of their jobs. Utilising this definition for measurement enables a more direct observation of the impact of both salary and non-salary factors on employees’ job satisfaction.

2.2. Factors influencing job satisfaction and research hypotheses

Research on job satisfaction often intertwines with employee motivation. As early as the 1920s, Frederick Winslow Taylor, known as the ‘father of scientific management’, championed the implementation of stimulating compensation systems to encourage diligent work among employees (He & Lin, 2006; M. Zeng & Qin, 2003). While piece-rate systems have indeed enhanced labour productivity within enterprises, the prevailing notion that money alone serves as the sole motivator for employees and productivity improvement remains challenging to substantiate in practical settings (Li et al., 2001). Numerous studies reveal that the relationship between compensation and job satisfaction is far from linear, and scientific management theories often overlook other critical factors influencing workers’ overall job contentment. Subsequently, research on employee motivation has shifted from material incentives to non-material aspects. The Hawthorne experiments conducted during the 1930s underscored the ‘social’ dimensions of employees, demonstrating that emotions and attitudes are profoundly shaped by the work environment. These emotional states, in turn, significantly impact employee productivity. However, the initial findings were somewhat limited due to oversimplified assumptions about employee behaviour within organisations. Organisational behaviour theories have since contributed to the evolution and refinement of employee motivation models. Notable examples include Maslow’s Hierarchy of Needs Theory, McGregor’s Theory X and Theory Y, and Herzberg’s Dual Factor Theory of Motivation-Health. These theories propose valuable hypotheses regarding how non-physical factors can exert motivational effects beyond mere financial rewards. Since the 1970s, the theory of organisational behaviour was further developed, and contemporary motivational theories such as ERG theory, equity theory, expectancy theory, motivational self-determination theory (Deci & Ryan, 2000), school ethic theory (Cai, 2016), transformational leadership theory, and ethical leadership theory (Bass, 1995; Benevene et al., 2018; Ogunfowora, 2014) have gained valid support from empirical research, suggesting that in addition to material incentives, non-material incentives are acting on employees’ work motivation through complex and diverse mechanisms.

The need for organisations to enhance employee productivity during the industrialisation process has spurred academic research on employee motivation. While much of this research has traditionally focused on for-profit enterprises, non-profit organisations present a distinct context. Unlike their for-profit counterparts, the primary purpose of non-profit organisations is to pursue social benefits (Mao et al., 2004). Consequently, alongside material incentives, there is a heightened need to reinforce non-material motivators, such as fostering a sense of mission and responsibility. Funds obtained by non-profit organisations cannot be freely distributed among members, resulting in comparatively weaker material incentives (X. C. Chen & Li, 2006). However, empirical evidence suggests that neither internal nor external benefits significantly weaken the link between the daily stresses faced by employees and their burnout (Savaya et al., 2020). The role of non-material incentives remains multifaceted and has not been uniformly confirmed. On the other hand, the governance of non-profit organisations revolves around supervision and incentives. Different types of non-profit organisations – whether institutions, privately-run non-profits, social groups, or foundations – possess distinct supervisory bodies, constituting diverse governance structures. Consequently, the incentives relevant to each type of non-profit organisation warrant separate examination. In China, the surge of marketisation and state-owned enterprise reforms during the 1990s, coupled with the country’s 21st-century accession to the WTO, has intensified the competition for talent. Consequently, enhancing employee satisfaction, curbing talent turnover, and fostering work initiative have become critical objectives in organisational management. Over the past three decades, Chinese scholars have made significant strides in researching employee satisfaction within enterprises and institutions. The scope has evolved from an initial emphasis on improving economic treatment to a later exploration of external environmental and non-compensation factors.However, despite this breadth of research, a relative dearth of studies exists concerning social organisations, particularly regarding the job satisfaction of social workers.

In recent years, research on the social work workforce has gained momentum, with scholars addressing critical issues such as social worker burnout, turnover, and job loss. Notably, compensation and benefits consistently emerge as primary reasons for the high turnover tendency, poor career stability, and diminished professional identity among social workers (Tang & Luo, 2022). While the economic perspective provides a robust explanatory framework, an exclusive focus on economics risks overlooking the nuanced qualities of social workers as ‘social beings’. This tension arises from the ethically charged nature of social work. Consequently, scholarly attention has shifted towards non-economic factors. Recent studies delve into social workers’ work initiative (Qin et al., 2022) and emotional stress (Guo & Han, 2021). However, despite these advancements, a comprehensive systemic perspective remains elusive. To fully understand and address the multifaceted challenges faced by social workers, future research must embrace a holistic view-one that recognises both the economic realities and the intrinsic complexities of their roles within society. Previous research by Western scholars has underscored the impact of three critical factors on job satisfaction among employees in non-profit organisations: personal interpersonal relationships, the organisational context, and the broader social environment (Savaya et al., 2020). However, despite this recognition, empirical findings regarding the precise role of these three aspects in shaping social workers’ job satisfaction remain inconclusive.

2.2.1. Interpersonal relationships

Interpersonal relationships encompass the intricate web of connections between individuals, reflecting psychological proximity and behavioural tendencies (Lin et al., 2016). Influenced by diverse schools of thought, Western scholars employ explicit concepts – such as interaction, social exchange, communication, interpersonal conflict, and attraction – to explore these relationships (Zhai, 1993). A good working atmosphere and organisational trust have a significant impact on employees’ emotions, work status, teamwork, and overall organisational satisfaction. Cui et al. (2012) study confirmed that a comfortable work atmosphere and close teamwork can support individuals’ emotions and alleviate their anxiety and burnout at work. Cummings and Bromiley defined the work atmosphere formed by how employees perceive and work with others at work as organisational trust, and Shaw argued that organisational trust affects cooperation among members and team effectiveness (Yu et al., 2007), and is an important variable in predicting employees’ overall satisfaction with the organisation as a whole (Driscoll, 1978).

In contemporary Western societies, organisational dynamics thrive on principles of honesty, trust, and fair competition, often grounded in contractual agreements. However, the landscape of interpersonal relationships in China diverges significantly due to cultural factors such as fate, familialism, and hierarchy. These relationships place substantial emphasis on emotional investment, extending beyond formal rules governing resource exchange (Chen & Chen, 2004; Lin & Si, 2010). While such human connections effectively reduce transaction costs and enhance opportunities, they also exert a profound influence within social work contexts.

In the field of social work, the establishment of professional trust often begins with gestures such as gift-giving, favours, and the delicate art of ‘saving face’ during client interactions (Q. Zeng, 2009). The emotional interactions intricately intertwine with instrumental cooperation among colleagues. Notably, the depth of emotional engagement correlates with subordinate obedience within superior-subordinate relationships. In the Chinese cultural context, maintaining working relationships involves a delicate balance between humanistic and instrumental components. While this blend of motivations fosters commitment, it also poses potential challenges to organisational health. To address this dynamic we propose the first hypothesis of this paper:

H1:

Interpersonal relationships significantly influence social work practitioners’ job satisfaction, with positive interpersonal relationships enhancing social work practitioners’ job satisfaction.

Considering our local context and existing data, we will assess three dimensions: ‘relationship with colleagues’, ‘relationship with agency managers’, and ‘relationship with service users’. This examination aims to understand how interpersonal relationships impact employees’ job satisfaction.

2.2.2. Organisational development

Organisational development, as a systematic endeavour, aims to enhance an organisation’s effectiveness by bolstering its problem-solving capabilities while concurrently advancing its human resources (Rothwell & Sullivan, 2001). Extensive research underscores the critical role of organisational factors in attracting and retaining talent. Corporate job seekers meticulously evaluate aspects such as the organisation’s structure, compensation and benefits, job roles, and qualifications during their search (Hu et al., 2008). Notably, an organisation’s image significantly influences talent acquisition and retention, as a negative perception can lead to employee stress and frustration. Factors such as organisational size, international presence, and geographical location also impact talent retention to varying extents. Importantly, the symbiotic relationship between organisational development and personal growth serves as a compelling motivator for career seekers to prioritise organisational prospects (Hu et al., 2008). Stronger organisational profiles often correlate with more adaptable hiring practices, ample opportunities for individuals to leverage their talents, and the potential for higher compensation. In sum, competitive salaries and ample room for advancement significantly contribute to job satisfaction.Similarly, social work agencies play a pivotal role in shaping the professional growth of social workers (Li, 2014). In China’s planned economy, the advancement of social organisations faced constraints. However, recent innovations in social governance have spotlighted the cultivation and expansion of social organisations, drawing governmental attention. At all administrative levels, governments have extended support to different types of organisations, fostering a ‘top-down’ model of social work development imbued with Chinese characteristics. This strategic approach aims to enhance internal job assessment, position selection, and mobility mechanisms for social workers, and thus to optimise the development environment for the social work talents. Nonetheless, this distinctive model presents two critical considerations. Firstly, it intertwines social work professionalism with political dynamics, potentially necessitating a delicate balance between state leadership, professional autonomy, and service effectiveness (Zheng et al., 2021). Secondly, despite policy directives aimed at enhancing social workers’ remuneration and benefits, salary caps inherent in non-profit organisations inherently limit earning potential. Consequently, the developmental trajectory of Chinese social work organisations may not necessarily correlate positively with the exercise of individual professional autonomy and increased compensation. Thus, a fundamental question arises: How does organisational development impact employees’ work attitudes, and in what manner does it influence job satisfaction? Drawing from the context of China’s transitional phase, and building upon existing research and data, we operationalise organisational development along two dimensions: a robust promotion mechanism and ample opportunities for advancement. This leads us to propose the second hypothesis of this study:

H2:

The development of social work organisations in China significantly influences social work practitioners’ job satisfaction, as more robust promotion mechanisms and ample opportunities for advancement enhance social work practitioners’ job satisfaction.

2.2.3. The social environment

The organisational environment encompasses various macro-level factors that influence all types of organisations, including the legal, economic, socio-cultural, scientific and technological, and natural domains. Notably, the socio-cultural environment directly shapes community engagement and the level of support and participation in social service agencies and their activities (Shi, 2015). In China, the unique model of top-down administrative development drives social work advancement, with the government leading through policy formulation, resource allocation, and oversight (Xu, 2017). While China’s social work sector has achieved significant milestones in professionalisation, job creation, talent development, agency establishment, service expansion, and benefits (Xu, 2017), challenges persist, as social organisations still face obstacles in gaining public support and integrating into the social governance structure (Huang, 2015). Citizen support hinges on perceptions of the social work profession, while external recognition and respect also influence social workers’ professional identity. Studies indicate that increased social policy support and public recognition of workers’ efforts, especially within the service sector, correlate with greater job satisfaction and professional stability (Zhu, 2012). This underscores the importance of recognition and respect in shaping the job satisfaction of social work practitioners in Chinese society. Therefore, we aim to explore how social workers’ status, perceived recognition, and societal appreciation collectively impact job satisfaction, framed within the concept of the ‘social environment’, forming the basis for our third hypothesis.

H3:

The social environment significantly influences social workers’ job satisfaction. Higher professional status, social recognition, respect and appreciation for the profession enhance social workers’ job satisfaction.

Additionally, previous studies have found that there are differences in demand preferences and satisfaction with various motivational indicators among people in different positions (Lin et al., 2016) and that managers are more satisfied than frontline employees (Zhang & Liao, 2007). This may be related to the fact that managers possess higher occupational skills and a higher sense of job value due to the greater responsibility inherent in their role. Explorations of employee and managerial job satisfaction have typically been conducted independently, considering that there are significant differences between the two roles in terms of personality traits, knowledge, job stress, rewards, autonomy, and challenges, all of which can significantly affect an individual’s job satisfaction. However, current research in China has focused primarily on the psychological state of frontline social workers, resulting in the relative neglect of job satisfaction at the management level. The sources of job satisfaction for management in non-profit organisations differ significantly from those of frontline social workers, with far-reaching implications for employee satisfaction, caseworker satisfaction, and even the overall functioning of the organisation. Existing research in China tends to marginalise this management perspective, and this study attempts to bridge this gap by isolating and exploring the job satisfaction of non-profit managers. Therefore, the fourth hypothesis of this study is proposed:

H4:

Interpersonal relationships, the development of social work, and the broader social environment significantly correlate with agency managers’ job satisfaction.

2.3. Significance of the study

In non-profit organisations, the need for non-material incentives has grown more pronounced, yet research on the non-economic impact of job satisfaction among employees in Chinese social work agencies remains relatively sparse. Moreover, there’s a dearth of detailed examination on job satisfaction across different positions within these agencies. Therefore, our aim is to contribute to the existing body of research on social workers’ job satisfaction in China by drawing from discussions in related disciplines both domestically and internationally. This study seeks to elucidate how various key factors influence the job satisfaction of social work practitioners within the Chinese context, offering valuable insights for optimising manpower management strategies within social work organisations. By doing so, it can serve as a valuable tool in reducing turnover rates among professional social workers and fostering the development of a skilled workforce. Additionally, while international empirical research on employee job satisfaction within social work organisations remains contentious, this study endeavours to provide data support by examining the Chinese case. Furthermore, although our focus is on non-profit entities, it’s worth noting that for-profit enterprises are increasingly recognising the limitations of traditional material incentives. In today’s globalised economic landscape, innovation is paramount for enterprises to bolster their market position. Knowledge-based employees, who are central to driving innovation within enterprises, often exhibit strong intrinsic motivation, a desire for self-fulfilment, and a profound need for recognition. These characteristics pose challenges to the efficacy of traditional incentives. Thus, our exploration of non-material incentives may offer valuable insights for enhancing employee motivation within for-profit enterprises as well (Bai & Xu, 2019).

3. Methods

3.1. Sampling and data collection

The data used in this study were obtained from a survey on the social work profession in four cities: Beijing, Shanghai, Guangzhou, and Shenzhen in China. The survey was supported by social workers’ associations in each region, and the respondents were professional practitioners from social work agencies in the four cities. A total of 1,615 questionnaires were collected, which included information on the practitioners’ attributes, salary and benefits, career development, turnover intention, professional identity, and job satisfaction. The sampling approach integrates two methods – systematic sampling and simple random sampling – building upon an equal-proportional stratified sampling foundation to enhance sample representativeness while ensuring efficiency. The detailed process, using Beijing as an example, unfolds as follows: Initially an equal proportion stratified sampling strategy was employed, with 1/8^th of the 199 social work organisations in Beijing chosen to constitute the sample. Considering agency staff size, the agencies were categorised into five strata based on the number of individuals: 1–5, 6–10, 11–15, 16–50, and 51–100. Each stratum was then sampled in approximately 1/8 proportion, resulting in the following final sample sizes for each stratum: 4, 9, 5, 7, and 2, respectively. Subsequently, the systematic sampling method was applied to select a sample of agencies within each stratum. Thirdly, acknowledging the practical constraints on staff participation in interviews, simple random sampling was employed to select 3/4 of the staff in each organisation for the survey. The final sample size obtained represents approximately 10% of the total number of social work practitioners in Beijing. The sampling methods for the other three cities mirrored those in Beijing. Consequently, the final sample size comprised 1,615 individuals, distributed as follows: 313 participants in Beijing, 411 in Shanghai, 391 in Guangzhou, and 500 in Shenzhen. Based on the above sample, ‘middle management’ and ‘top management’in the ‘job category’ variable were merged into a single category: agency managers. The sample size of ‘frontline social worker’ and ‘agency manager’ was 1,387, including 994 frontline social workers and 393 agency managers. This study further cleaned the data according to the personal characteristics of 1,387 samples, namely gender, age, marital status, education level, household registration, and political status. After deleting the missing values, there were 1,371 observations, including 981 frontline social workers and 390 agency managers. Stata 16.0 was used to analyse the data.

3.2. Measurement

3.2.1. dependent variable

The dependent variable in this study is job satisfaction. Various scales have been developed to measure job satisfaction by incorporating influential factors. Among the most widely used scales are the Job Description Index (JDI), the Minnesota Satisfaction Questionnaire (MSQ), and the Job Satisfaction Survey (JSS). The JDI, in its final version, encompasses five categories: work, pay, supervision, promotion, and co-workers (Conway et al., 2013). The MSQ delineates job satisfaction across 20 dimensions, including personal ability to perform, fulfilment, motivation, company training and self-development, authority, company policies and implementation, compensation, teamwork, creativity, independence, ethical standards, rewards and punishments, personal responsibility, job security, social services, social status, relationship management, communication, technological development, diversified development, and working conditions (Weiss et al., 1967). The JSS focuses on facets like pay, promotion, supervision, fringe benefits, contingent rewards, operating conditions, co-workers, nature of work, and communication (Carbonel & Claudette, 2007).

Scholars have devised a job satisfaction scale that demonstrated a significant correlation in Chinese with the MSQ in empirical studies (Cui et al., 2012). This scale gauges employee job satisfaction through five secondary dimensions: leadership behaviour, management measures, job rewards, job collaboration, and the job itself. In the present study, eight dimensions from these measurement scales were identified for assessing the job satisfaction of social work practitioners in China. These dimensions include salary, welfare, vocational training, work independence, evaluation mechanism, agency management, work content, and work environment.

The total score was calculated by assigning satisfaction values to these eight dimensions to obtain the job satisfaction variable. The options for the eight questions were (1) very dissatisfied, (2) not very satisfied, (3) average, (4) basically satisfied, and (5) very satisfied, which were assigned the values of 1, 2, 3, 4, and 5, respectively. The total scores of the eight questions were then summed to obtain the job satisfaction variable.

We conducted a reliability analysis of the Social Worker Satisfaction Scale to validate the reliability of its eight dimensions. The results, presented in the table above, indicate a high level of consistency among the dimensions, with a Cronbach’s alpha coefficient of 0.843 and a KMO value of 0.829. These findings affirm that the scale possesses strong reliability in assessing social worker satisfaction. Furthermore, the examination of each question’s impact on Cronbach’s alpha revealed that the removal of any single question did not significantly improve alpha values. Thus, the scale demonstrates robust reliability in this study, as evidenced by Cronbach’s alpha, KMO, and Bartlett’s test, providing a dependable measurement tool for subsequent research. This enables us to explore the relationship more accurately between social worker satisfaction and other key variables.

3.2.2. Independent variables and control variables

Eight questions related to interpersonal relationships, organisational development, and social environment were selected from the questionnaire. In order to eliminate the collinearity among some variables, the principal component method and maximum variance orthogonal rotation method were used for factor analysis. In this study, the variables used above have been treated in the same direction before the factor analysis, that is, the scores of the related variables entering the factor analysis were: (1) totally disagree; (2) disagree; (3) neutrality; (4) basically agree; (5) totally agree. The results show that the KMO test value is 0.7121 and the KMO value of each item is greater than 0.5. The approximate chi-square of Bartlett’s sphericity test is 5418.081, p < 0.001 (Table 2). Therefore, the eight variables selected in this study were suitable for factor analysis. Three common factors were extracted according to the standard that the characteristic value was greater than 1, and the cumulative variance contribution rate of the three factors reached 79.91%. As shown in Table 1, the common factor 1 corresponded to three items, which reflected the degree of public recognition, including social status, social recognition, and social respect for social work. Factor 2 reflected relationships between social workers and colleagues, agency supervisors, and clients. Two items corresponding to the public factor 3 could measure the opportunities for social workers’ career development. Finally, this paper standardised these three factors, so that the score of each factor ranged from 0 to 100 and took the processed factors as the key independent variables, namely public recognition, interpersonal relationship, and career advancement and development.

Table 1. Cronbach’s alpha and KMO of job satisfaction scale.

Variables	Obs	Cronbach’s alpha	KMO
1. Salary and benefits	1371	0.8266	0.782
2. Welfare and security	1371	0.8159	0.7971
3. Vocational training	1371	0.8162	0.8833
4. Work independence	1371	0.8092	0.8849
5. Evaluation mechanism	1371	0.8196	0.8914
6. Agency management	1371	0.8156	0.8831
7. Work content	1371	0.8512	0.6703
8. Work environment	1371	0.8387	0.7507
Overall		0.8432	0.8291
Bartlett test of sphericity	Chi-square		4732.833
	Degrees of freedom		28
	p-value		<0.001

Table 2. Rotation factor load, KMO and Bartlett’s test of sphericity.

	Variables	Public recognition factor (Factor1)	Interpersonal relationship factor (Factor2)	Career advancement and development factor (Factor3)	KMO
Social environment	increased social status of social workers,	0.9286	0.109	0.0829	0.7087
	increased perceived recognition of social workers,	0.9179	0.1339	0.0781	0.7303
	respect and appreciation of the profession by the society,	0.8716	0.0817	0.085	0.8617
Interpersonal relationship	having good relationships with colleagues,	0.0675	0.9011	−0.0712	0.6658
	having good relationships with agency managers,	0.1465	0.865	0.0285	0.7106
	having good interactions with service users,	0.1718	0.7644	−0.1762	0.8225
Organizational development	sound promotion mechanism,	0.1412	−0.0537	0.8875	0.5806
	plenty of room for development	0.0606	−0.0725	0.9027	0.5527
Correlation test	KMO	overall			0.7121
	Bartlett	chi-square			5418.081
		df			28
		p-value			<0.001

The control variables include personal characteristics variables and job characteristics variables. Descriptive analysis results of them are shown in Table 3. In terms of variables related to personal characteristics, the sample was comprised of 380 males, or 27.72% of the total number, and 991 females, nearly three times as many as males. We can infer that in the social work profession in China, the gender ratio is very uneven, with many more female social work practitioners than males. Regarding age, the 26–35 age group accounted for most of the total sample. China is at an early stage in the development of the social work profession, and the nature of social work focuses on the output services that determine the generally low age of social work practitioners; thus, our sampling results are reasonable. Regarding marital status, the difference between the sample size of the unmarried and married was not significant, and the sample drawn was balanced. Regarding educational level, it showed that the majority of social work practitioners had a high educational level of college or bachelor’s degree. Overall, the sample and data used in this study were suitably representative to support this study. In terms of variables related to job characteristics, the monthly average salary, being licenced or not, the major, monthly supervision time, and monthly rest time were selected.

Table 3. Sample characteristics.

Variable definition and assignment	Observations	Percentage (%)		Mean	SD	Min	Max
Job satisfaction	1371	100	23.938		4.920	8	40
Public recognition	1371	100	59.644		18.247	0	100
Interpersonal relationship	1371	100	72.617		13.547	0	100
Career advancement and development	1371	100	34.690		17.029	0	100
Gender	1371	100	0.723		0.448	0	1
male = 0	380	27.72
female = 1	991	72.28
Age	1371	100	30.48		6.752	18	60
≤25	344	25.09
26–35	739	53.90
36–45	229	16.70
≥46	59	4.30
Marital status	1371	100	0.506		0.500	0	1
unmarried = 0	677	49.38
married = 1	694	50.62
Education level	1371	100	3.77		0.59	1	5
junior high school = 1	3	0.22
senior high school = 2	18	1.31
college = 3	355	25.89
bachelor = 4	906	66.08
postgraduate = 5	89	6.49
Household registration	1371	100	0.381		0.486	0	1
in this city = 0	848	61.85
not in this city = 1	523	38.15
Politics status	1371	100	0.241		0.428	0	1
not member of Communist Party of China (CPC) = 0	1040	75.86
member of CPC = 1	331	24.14
Salary	1371	100	1.78		0.77	1	3
≤4000 RMB = 1	590	43.03
4001–5500 RMB = 2	498	36,32
≥5501 RMB = 3	283	20.64
licensed	1371	100	0.82		0.39	0	1
no licenced = 0	252	18.38
licensed = 1	1119	81.62
Major	1371	100	0.52		0.50	0	1
other professions = 0	657	47.92
social work = 1	714	52.08
Monthly supervision time (hr)	1371	100	2.59		3.57	0	64
Monthly rest time (day)	1371	100	6.81		2.04	0	10

4. Results

4.1. Job satisfaction of social work practitioners in different positions

As show in Table 4, social work practitioners’ job satisfaction scores were not high, and the satisfaction with salary and benefit, welfare and security were the lowest, which reflected that the salary and welfare of social work professionals in China must be improved. However, there was generally high satisfaction with work contents, with a mean score of 3.88, and the second highest score was in the work environment dimension. The overall satisfaction scores of social work practitioners in each dimension showed salary and benefits (2.3115) < welfare and security (2.5529)< work independence (2.7002) <evaluation mechanism (2.9212) < vocational training (2.9336) < agency management (3.1014)< work environment (3.5361) < work contents (3.8811).

Table 4. Average job satisfaction of each dimension.

Variables	All			Frontline social workers			Agency managers			T test
	Obser-vations	Mean	SD	Obser- vations	Mean	SD	Obser- vations	Mean	SD	T-value	Sig
1.Salary and benefits	1371	2.3115	0.9202	981	2.2671	0.9080	390	2.4231	0.9420	−2.8394	0.005
2. Welfare and security	1371	2.5529	0.9715	981	2.5250	0.9818	390	2.6231	0.9426	−1.6881	0.092
3. Vocational training	1371	2.9336	0.8436	981	2.9286	0.8554	390	2.9462	0.8142	−0.3466	0.729
4. Work independence	1371	2.7002	0.8729	981	2.6453	0.8751	390	2.8385	0.8527	−3.7148	<0.001
5. Evaluation mechanism	1371	2.9212	0.8727	981	2.9888	0.8541	390	2.7513	0.8968	4.5791	<0.001
6. Agency management	1371	3.1014	0.9029	981	3.0979	0.9206	390	3.1103	0.8578	−0.2293	0.819
7. Work content	1371	3.8811	0.8308	981	3.8950	0.8632	390	3.8462	0.7430	1.0474	0.295
8. Work environment	1371	3.5361	0.9037	981	3.5372	0.9258	390	3.5333	0.8470	0.0744	0.941
All	1371	23.938	4.9203	981	23.885	4.99	390	24.072	4.7443	−0.6347	0.526

Because the total satisfaction score did not pass the independent sample t test, there was no significant difference in the overall score of satisfaction between the frontline social workers and agency managers, but there were differences in some dimensions. The satisfaction scores of frontline social workers showed salary and benefits (2.2671) < welfare and security (2.525) < work independence (2.6453) < vocational training (2.9286) < evaluation mechanism (2.9888) < agency management (3.0979) < work environment (3.5372) < work contents (3.8950). For this group of people, in addition to the salary and welfare, the acquisition of job independence was the most important dimension of satisfaction. The satisfaction scores of agency managers showed salary and benefits (2.4231) < welfare and security (2.6231) < evaluation mechanism (2.7513) <work independence (2.8385) < vocational training (2.9462) < agency management (3.1103) < work environment (3.5333) < work contents (3.8462). For this group of people, apart from the salary and welfare, what made this group most dissatisfied was the evaluation mechanism from government departments, funders or other third-party departments. This reflected that the two groups had corresponding differences in job satisfaction according to their own job characteristics. The specific work of front-line social workers may often be influenced by uncertain factors from relevant government departments, agency managers, clients, and agency supervisors, which would lead to low satisfaction with work independence. The evaluation of social work agencies includes several aspects such as service quality, operation and management, purchaser satisfaction, client satisfaction, and financial evaluation. Agency managers were more aware of problems with the current social work evaluation mechanism than frontline social workers, such as the low reliability and validity of the evaluation and inadequate use of the evaluation results (Ma, 2016), which resulted in a low satisfaction of agency managers with the evaluation mechanism of the industry.

4.2. Regression analysis

This study constructed three models (Table 5) to test the influence of public recognition， interpersonal relationships, career advancement, and development on the job satisfaction of frontline social workers, agency managers and all of them. In this process, the relevant variables of personal characteristics and job characteristics were controlled.

Table 5. OLS model of job satisfaction of social workers in different positions.

	Model 1 (overall)	Model 2 (frontline social workers)	Model 3 (agency managers)
	Job Satisfaction	Job Satisfaction	Job Satisfaction
Public recognition	0.0681***	0.0644***	0.0745***
	(0.0071)	(0.0084)	(0.0132)
Interpersonal relationships	0.0468***	0.0542***	0.0209
	(0.0107)	(0.0115)	(0.0241)
Career advancement and development	0.1151***	0.1185***	0.1096***
	(0.0083)	(0.0101)	(0.0142)
Female	0.1432	0.0362	0.5050
	(0.2769)	(0.3391)	(0.5189)
Age	0.0267	0.0458*	−0.0184
	(0.0216)	(0.0256)	(0.0470)
Married	0.0289	−0.3754	0.8664*
	(0.2925)	(0.3702)	(0.5044)
Household registration not in this city	0.1812	0.1107	0.2984
	(0.2648)	(0.3410)	(0.4592)
Member of CPC	−0.0421	0.1232	−0.2846
	(0.2747)	(0.3583)	(0.4593)
College diploma above	−0.0323	−0.0938	0.1346
	(0.3023)	(0.3570)	(0.5768)
Monthly average salary 4001–5500 RMB	0.1976	0.1718	0.3367
	(0.2976)	(0.3556)	(0.6728)
Monthly average salary ≥5501 RMB	0.5863	0.2720	0.8533
	(0.3885)	(0.5223)	(0.7683)
Social work major	0.3333	0.4283	−0.0714
	(0.2586)	(0.3118)	(0.4803)
Monthly supervision time(hr)	0.1155***	0.1096**	0.1080**
	(0.0338)	(0.0449)	(0.0545)
Monthly rest time(day)	0.2159***	0.2337***	0.1594*
	(0.0622)	(0.0826)	(0.0941)
Licensed	−0.6509*	−0.6662*	−0.1817
	(0.3575)	(0.3909)	(0.8770)
Constant	9.9123***	9.0825***	12.2522***
	(1.1883)	(1.4117)	(2.3781)
Observations	1371	981	390
R2	0.2688	0.2740	0.2752

1.*p < 0.1, ** p < 0.05, *** p < 0.01; The brackets are robust standard errors.

OLS model:

$\mathit{Satisfactio}{n}_i={\beta }_0+{\beta }_1{f}_1+{\beta }_2{f}_2+{\beta }_3{f}_3+{\beta }_4{X}_i+{\epsilon }_i$

$f_1$, $f_2$, $f_3$ were the extracted three common factors, namely interpersonal relationship, public recognition, career advancement and development.$\mathit{\hspace{0.17em}}{X}_i$ represented the controlled variables, and ${\epsilon }_i$ was the error term to control the missing variables.

Model 1 is the results of the overall sample regression. While controlling for other variables, the three core independent variables were positively significant to the overall satisfaction of social work practitioners. Social work practitioners’ job satisfaction is influenced by the dimensions of interpersonal relationships, organisational development, and social environment. Some variables related to job characteristics (monthly supervision time, monthly rest time, being licenced) also had a significant impact on the overall satisfaction.

The results of model 2 showed that public recognition, interpersonal relationships, career advancement, and development had statistically significant and positive effects on the job satisfaction of frontline social workers. For every unit of public recognition, job satisfaction of frontline social workers increased by 0.0644 (p < 0.01). Interpersonal relationships significantly affect the job satisfaction of frontline social workers. The better interpersonal relationships with colleagues, managers, and clients, the higher their job satisfaction. For each additional unit of career promotion and development, the job satisfaction of frontline social workers increased by 0.1185 (p < 0.01). In model 2, we find that control variables such as age, monthly supervision time, monthly rest time, and licensure status significantly impact the job satisfaction of frontline social workers. Job satisfaction increased by 0.1096 (p < 0.05) for every additional hour of supervision. Ensuring adequate monthly rest time also improved their job satisfaction. Notably, at the 0.1 significance level, the job satisfaction of frontline social workers with professional certificates was 0.6662 points lower than those without certificates. Thus, hypotheses 1, 2, and 3 were tested and supported by the data.

Model 3 is a regression analysis of the job satisfaction of agency managers. Career advancement and development, and public recognition had significant and positive effects on the job satisfaction of agency managers, while interpersonal relationships had no significant effect. For every unit of public recognition, the job satisfaction of agency managers increased by 0.0745 (p < 0.01). For each additional unit of career advancement and development, the job satisfaction of this group of people increased by 0.1096 (p < 0.01). In Model 3, the control variables – marital status, monthly supervision time, and monthly rest time – had significant effects on the job satisfaction of agency managers. The job satisfaction score of agency managers increased by 0.1080 (p < 0.05) for every additional hour of professional supervision. Additionally, more monthly rest time was associated with higher job satisfaction among agency managers. There was no significant difference in job satisfaction between those who were licenced and those who were not. Hypothesis 4 was partially confirmed.

5. Discussion

The demand for social work services in China has surged due to ongoing social transitions and resulting uncertain situations. Ensuring a stable team of social work practitioners is needed. Against the backdrop of a brain drain within the social work profession, investigating job satisfaction and the influencing factors holds significant social relevance. This paper breaks beyond the traditional analysis of the causes of the brain drain in the social work industry through salary and relevant factors. Instead, taking a novel approach, drawing from insights from the perspective of content-based, process-based, and situational model theories influencing factors of job satisfaction.

According to the Herzberg’s Two-Factor Theory of Job Satisfaction, social work practitioners’ job satisfaction is closely related to‘motivators’ and ‘hygiene’. The recognition of social work as a profession is conducive to its access to relevant resources of ‘motivators’ in society. Different from findings in existing literature, in recent years, China has further defined the position and role of social work in social governance and community development in various policy documents, which has made the profession of social work enter the public’s vision and gain social recognition for its professional methods and good service results. As another type of ‘motivator’, the establishment of a predictable and achievable career promotion mechanism could stimulate social work practitioners’ positive emotions. In the individual’s emotional structure, gaining power and status in the organisation is conducive to promoting the awakening and process of job satisfaction. In the expected state, these emotional mechanisms will be gradually strengthened. Career advancement and development will bring interactive emotional experience between frontline social workers and agency managers and increase unity within the organisations. This also shows that the working mechanism of social work practitioners’ job satisfaction at the organisational level is in line with the Expectancy theory that focuses on human behaviour choices.

The quality of interpersonal relationships is one kind of ‘hygiene’, and the structure of organisational identity in China is special due to its emphasis on the harmony of relationships. The relational identity generated by communication with others may become an important phenomenon of Chinese organisational identity (Sun & Jiang, 2009). Professionally trained social workers are guided by humanitarian values and a series of working methods and techniques, such as respect and acceptance of others, empathy, and sincerity, which help them cultivate good interpersonal and working relationships. The work of frontline social workers is mainly involved in the development of service activities. Good interaction with clients can help social workers gain a sense of achievement in their work and enhance their identification with their profession. In addition, social work agencies are generally engaged in humanistic work, and service activities cannot be carried out without solidarity and cooperation among various departments and colleagues. Good collegial relationships can enhance the external support of frontline social workers, and when negative emotions arise at work, timely guidance from colleagues and friends can enhance their confidence and persistence to a certain extent. The effect of interpersonal relationships on the job satisfaction of agency managers is not as high as that of frontline social workers because there is comparatively less direct interaction between agency managers and clients. What’s more, agency managers need to master the degree of interaction with their staff: if they are too distant from their colleagues, it will lead to a backlash and make it difficult to carry out their work smoothly, but if they are too close, it will have an impact on their managerial authority.

As another‘hygiene’ factor, social work supervision has played an important role in promoting the professional development of social work agencies and the cultivation of professionals. It plays educational and supportive functions and guarantee of monthly supervision time can buffer the heavy work of social work practitioners. We further find that whether one holds a social work professional certificate has a significant effect on the job satisfaction of social work practitioners in general and frontline social workers in particular, with the latter who have certificates having lower job satisfaction than those who do not. This is contrary to our general perception that holding a relevant certificate will enhance the professional identity of social workers and thus create higher job satisfaction (Tang & Luo, 2022). Based on the Expectancy theory, this is because the frontline social workers who hold certificates have higher expectations of work professionalism. Nevertheless, as mentioned before, the professionalism of their direct social work practices has been weakened due to inadequate work independence. In addition, there is no obvious difference in the benefits and career advancement of employees with or without professional certificates.

Accordingly, more work should be done to increase the job satisfaction of frontline social workers and agency managers. The first is to guarantee the work independence of these social work practitioners by reducing the administrative intervention in direct social work practices in the government purchase of services. In China, embedding themselves into the administrative system has become the most rational and inevitable choice for social work agencies (Author, 2021), but this embedding should not cover social work professionalism. The second, which is related to the first one, is to reform the evaluation system of social work services and social work agencies by including more professional elements rather than administrative ones. The evaluation criteria from multi-actors should also be unified on the base of coordination and negotiation. The third is to improve the career advancement mechanism of social work agencies vertically, so that social work practitioners especially those with professional certificates, have room for promotion, which is the key to retaining talents. Last, but not least, is to strengthen the training and supervision of social work practitioners, to provide professional and emotional support for them, and to cultivate teamwork of high quality. These measures will also further enhance the public recognition of social work in China.

Future research can build on this study to enhance the effectiveness of social work practice. Firstly, it should be noted that the research samples of this study are all from social workers in four first-tier cities in China. Compared with other cities, the personal characteristics of social workers, the development of their institutions, and the social environment are different. Therefore, the job satisfaction of social workers in cities with different levels of economic development needs to be further compared. Second, professional value is a relatively subjective concept, and the understanding and evaluation of professional value are influenced by cultural, social, and historical backgrounds, so it may not fully reflect the diversity and complexity of professional value in different cultures and backgrounds. The research on the value characteristics of social work needs to comprehensively use a variety of research methods, including qualitative research and mixed method research, for more complicated analysis and in-depth research design. Thirdly, future research should aim to elucidate the intricate relationship between job satisfaction and factors such as burnout and the inclination to leave. This endeavour is essential for constructing a well-defined and comprehensive framework delineating the mechanisms governing the talent flow within the field of social work.

6. Conclusion

In addition to the general fact that the salary and welfare of social work practitioners in China need to be improved, this study find that this group of people are more satisfied with the work content and working environment. Social workers are often not satisfied with their work remuneration, but they are more satisfied with the work environment, which mainly includes organisational culture atmosphere and interpersonal relationships, and the work content that can reflect the three main psychological states of practitioners’ sense of value, sense of responsibility and sense of efficacy. This also shows that social work in China pays more attention to the professional practice content and the construction of social and organisational environment and has achieved results in recent years. Nevertheless, there is room for improvement in the dimensions of work independence, evaluation mechanism, vocational training, and agency management, especially the first two. The administrative development model of social work in China has made significant progress towards professionalisation. However, there are still unresolved issues such as the unreasonable organisational structure and inadequate coordination among multiple stakeholders. For frontline workers, their job independence lies in the dilemma between professional value practice and administrative guidance which may blur the boundary of professional autonomy. For the managers who take on the mission of agency development, there are many overlaps in the specific standards of agency performance and project evaluation of multi-departments, which reflect the superposition of many factors, such as a lack of control over the resources needed for decision-making and work. We think that compared with for-profit enterprises, the research on the problem of staff turnover in non-profit organisations should take into account broader immaterial factors besides salary and welfare, especially in the social work industry with strong value characteristics.

Acknowledgments

This research was supported by the Project of the National Social Science Foundation of China under Grant No. 19BSH176; the ‘Young Academic Team Leading Program 2022’ of Minzu University of China, with accordance ethical approval from the funding academy.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available. Data could be available upon request to the research team.

Additional information

Funding

The work was supported by the Major Project of Philosophy and Social Sciences Research, Ministry of Education [No. 2024JZDZ036].

Notes on contributors

Hui Yang

Hui Yang is a professor of the Department of Social Work and Social Policy, Nankai University. She has long been devoted to the construction of Chinese knowledge system of social work and social policy, and research on health care and anti-poverty social policies, social work for the old and young.

Zhezhen Lv

Zhezhen Lv is an assistant professor of Department of Social Work, School of Social Development, Tianjin University of Technology. She has been devoted to the research on professional identity of Chinese social workers and rural social work practice.

Ying Zhang

Ying Zhang is a doctoral candidate at Minzu University of China, with research interests in Social work education and geriatric social work.

Ting Cui

Ting Cui is a MSW graduate and Street Office staff with research interests in community social work. She is experienced in community building and the development of community social worker teams.

Honglin Chen

Honglin Chen is a professor of gerontological social work at the University of Eastern Finland. She also serves as member of the Committee on Geriatric Welfare Science and Technology of the Chinese Ageing Well Academic Association. She has long been devoted to research on positive ageing, non-pharmacological interventions for cognitive disorders, gerontological social work, and social capital.