The role of job satisfaction and organizational culture match in the relationship between pay satisfaction and turnover intention: A moderated mediation model

Abstract

High turnover among social workers is a long-standing concern. This article explores whether pay satisfaction (PS) influences turnover intention (TI) and whether job satisfaction (JS) mediates this relationship. We also examine whether organizational culture match (OCM) moderates this mediation model. Based on multiple regression model of 5525 Chinese social workers, we found that PS was negatively related to TI and JS mediated this relationship. The effects of such a mediation model were significantly larger when OCM was at a higher level. Findings highlight the need to increase material rewards for social workers and to build organizational culture supporting personal psychological needs.

Keywords

China job satisfaction organizational culture salaries and fringe benefits social work turnover

Introduction

Since the economic reform in 1978, China has experienced rapid industrialization, urbanization, and modernization, alongside a complex amalgam of social issues like unemployment, income disparity, migration history, mental illness, juvenile delinquency, and an aging population (Leung, 2007; Meng et al., 2018). Facing the urgent needs of diverse social groups, the Chinese government adopted multiple strategies for social work education, workforce construction, and service purchase, as part of the expansion of social services (Lei et al., 2018; Yuan et al., 2021). In 2018, 6.11 billion RMB (around US$923.3 million) were invested in the development of social work (19.6% more than that in 2017), 51,000 social work stations were established, 9793 social work agencies were registered, and 383,000 job positions for social work were created (Ministry of Civil Affairs, 2019).

Despite this remarkable increase, the workforce in China’s social sector encountered difficulties in recruitment and retention. Although nationwide data are unavailable, even in cities leading in social work development such as Shanghai, Beijing, and Guangzhou, the annual turnover rate among social workers was nearly 20% in recent years (Jiang et al., 2019; Sun and Zhou, 2019). High workforce turnover in the social service sector has remained a long-standing concern the world over. A high social work turnover rate affects the quality, consistency, and stability of the services for clients (Barak et al., 2001). It also increases the cost of staff training, and professional skills and knowledge accumulation, and affects team morale and professional reputation negatively (Carmeli and Weisberg, 2006).

Owing to its serious consequences, turnover has attracted academic attention across a range of disciplines. Scholars in the fields of psychology, sociology, and economics have proposed theories to explain employee turnover. Psychological theories have emphasized individual perceptions and attitudes toward work conditions; sociological theories have highlighted work-related factors; and economic theories have emphasized an individual’s rational responses to economic and organizational conditions (Barak et al., 2001). Empirical studies on turnover antecedents among human service workers have identified three main categories of contributors: (1) demographic and work-related factors; (2) professional perceptions, including organizational commitment and job satisfaction; and (3) organizational conditions, such as the compensation system and organizational culture (Barak et al., 2001; Tham, 2006). However, the empirical findings are inconsistent. Few empirical studies have explored the relationships among employees’ economic issues, professional perceptions, organizational conditions, and work attitudes in an integrated manner.

Informed by the self-determination theory (SDT), this study explored the relationships among pay satisfaction (PS) and job satisfaction (JS), organizational cultural match (OCM), and turnover intention (TI) of Chinese social workers. It investigated whether PS reduces TI among Chinese social workers and whether JS mediates the relationship between PS and TI. It examined the role of OCM as a moderator in this mediation model.

Pay satisfaction (PS), job satisfaction (JS), and turnover intention (TI)

Turnover, defined as an individual’s act of leaving an organization, is a process rather than a sudden action (Tham, 2006). Among all cognitions that form the turnover process, TI, which is defined as an individual’s conscious and deliberate willingness to leave an organization (Tett and Meyer, 1993), is reported as the single strongest and best predictor of actual turnover (Alexander et al., 1998). Actual employee turnover data are often inaccessible to researchers in cross-sectional studies. Therefore, most empirical studies have used TI rather than actual turnover as the outcome variable (Wang et al., 2020). Thus, this study also uses TI as the outcome variable.

As a theory of work motivation, SDT distinguishes between amotivation (which indicates having no intention to act) and motivation (which involves intentionality). SDT highlights the distinction between controlled and autonomous motivation (Deci et al., 2017; Gagné and Deci, 2005). The controlled motivation comprises external regulation and introjected extrinsic motivation, reflecting the fact that the employee seeks to be coerced or seduced by external contingencies or their introjected counterparts (Gagné and Deci, 2005). The autonomous motivation includes intrinsic (which refers to activities for which the motivation lies in the behavior itself; Deci et al., 2017) and well-internalized extrinsic motivation (i.e. integrated regulation, which indicates that the employee has fully integrated the behavior into the self; Gagné and Deci, 2005). SDT suggests that workplace motivations affect employees’ performance and wellbeing, and highlights that different types of motivations play different roles in the process (Deci et al., 2017).

Among the factors that affect TI, pay is a typical extrinsic motivator that induces controlled motivation, as few people find their jobs interesting and important enough to stay if they are unpaid (Barak et al., 2001; Deci et al., 2017). Greater performance is observed when employees consider the compensation system fair (Rigby and Ryan, 2018). PS is the degree of overall positive or negative feelings that individuals have toward their pay (Miceli and Lane, 1991). It describes the discrepancy between actual salary and personal standards of comparison, such as what they deserve, want, or see others receiving (Rice et al., 1990). Accumulated findings have shown that low PS is directly associated with high TI (Chan and Ao, 2018; Luo and Chui, 2018). A study of 819 Chinese social workers found that PS was significantly inversely related to TI (Wang et al., 2021).

JS is determined by comparing actual job outcomes to the outcomes desired by the employee (Hulin and Judge, 2003). It reflects an affective or attitudinal reaction to a job (Spector, 1985). Locke (1976) proposed that job attitudes were derived from three conditions, including discrepancies between what the job offers and what the person expects, the degree to which the job fulfills individual needs, and/or the degree to which individual values are fulfilled. The attitudinal nature of JS shows that employees will tend to hold onto a satisfying job and quit a dissatisfying one (Spector, 1985). Employees who are highly satisfied with their jobs are expected to develop relatively low levels of TIs, as both their intrinsic and extrinsic needs are achieved (Carmeli and Freund, 2009). This is supported by previous research, which has shown that JS is a consistent predictor of turnover behavior (Barak et al., 2001). For example, Carmeli and Freund (2009) collected two primary datasets from social workers in nonprofit social service organizations in Israel and found that JS was negatively related to TI. Chan and Ao (2018) conducted a study among 303 casino employees in Macau and found that JS was negatively related to TI. Therefore, the first hypothesis was framed as follows:

Hypothesis 1. PS is negatively associated with TI; JS is negatively associated with TI.

According to SDT, JS is a typical dependent variable among motivation variables (Gagné and Deci, 2005), that is, employees assess JS based on intrinsic (such as feelings of purpose at work) and extrinsic motivation (such as compensation; Egan et al., 2004). Most literature has shown that PS significantly influences JS (Chan and Ao, 2018; Singh and Loncar, 2010). In addition, PS can affect TI directly (Rathakrishnan et al., 2016) or indirectly through JS or organizational commitment (Heavey et al., 2013). Thus, it is reasonable to assume that JS mediates the relationship between PS and TI.

Hypothesis 2. JS mediates the relationship between PS and TI.

The role of organizational culture match (OCM)

Studies show that autonomous motivation (i.e. intrinsic motivation and integrated regulation) maximizes commitment and satisfaction in the workplace (Gagné and Deci, 2005). According to SDT, when employees can identify with the value and importance of their jobs, they will demonstrate an enhanced sense of work motivation. Integration takes place when contextual regulations have been evaluated and brought into congruence with one’s other values and needs (Ryan and Deci, 2002). With integrated regulation, employees may believe that their individual values and identities match the organizational culture.

Organizational culture represents a set of values, beliefs, and behavior patterns that form the core identity of an organization (Denison, 1984). An organization’s culture can make the organization more or less attractive to individuals based on each person’s value structure (Emerson, 2013), as people tend to seek out organizations that epitomize their personal values and beliefs (Bosomtwe and Obeng, 2018). Empirical evidence highlighted that the match between the values of the hospital and nurses’ preferred values predicted nurses’ turnover (Vandenberghe, 1999). Silverthorne (2004) found that high-level person-organization fit was statistically related to low turnover rates. A meta-analysis emphasized that the subjective measure of person-organization fit had the strongest average correlations with TI, with a negative coefficient of −0.58 (Verquer et al., 2003). Therefore, the third hypothesis in this study is as follows:

Hypothesis 3. OCM is negatively related to TI.

Organizational culture has an effect on employee behavior and is proposed as a potential moderator of voluntary turnover (Kessler, 2014). A previous study found that organizational culture moderates the relationship between JS and TI (Medina, 2012). However, more research is needed to examine the role of OCM on the relationship between JS and TI. Moreover, the role of OCM on the mediated relationship between PS and TI through JS has not been examined. According to SDT, employees who perceive a good fit between themselves and the organizational culture, will feel that their inner values and beliefs are supported, validated, shared, and emphasized by the organization and their colleagues; as a result, they are more satisfied with their work and compensation, and have greater trust in and loyalty to the organization (Rigby and Ryan, 2018). Therefore, the magnitude of the relationship between JS and TI may be based on the level of OCM. Similarly, the magnitude of the indirect effect of PS on TI may be based on the level of OCM. We thus propose the following hypotheses:

Hypothesis 4. OCM moderates the relationship between JS and TI. This relationship is weaker among individuals with low levels of OCM than among those with high levels of OCM.

Hypothesis 5. OCM moderates the magnitude of the mediated relationship between PS and TI through JS, such that the mediated relationship is expected to be stronger with high levels of OCM.

The full theoretical model in this study is presented in Figure 1.

Figure 1.

The hypothesized moderated mediation model of pay satisfaction (PS), job satisfaction (JS), organizational cultural match (OCM), and turnover intention (TI).

Methods

Source of data and study population

We used the 2019 baseline survey data from China Social Work Longitudinal Study (CSWLS), which is the first comprehensive national survey in the field of social work in China. It is conducted jointly by the School of Social and Public Administration and the Institute of Social Work and Social Policy. A priori power analysis was conducted to compute a required sample size. Eligiblity criteria for potential participants were that social work agency should be registered by 1 June 2019, and social worker should be full-time employee who performed social work services (not supportive tasks) and had been working in the current agency for at least 3 months. The study adopted a multistage sampling procedure to compose a national representative sample. There were three major steps for sample units selection: (1) the primary sampling units of 56 cities from the five stratums, which were selected based on the number of social work agencies, administrative units, and social economic development status; (2) the second sampling units of social work agencies were randomly selected from the sampling frame; and (3) the third sampling units of individual social workers were randomly selected from the employee roster of each agency (Yuan et al., 2021). The implementation of the baseline survey relied on the partnership with local institutions, which fulfilled the standard training sessions conducted by the research team. The team conducted close supervision, questionnaire inspection, and data entry inspection to ensure a satisfactory data quality (Yuan et al., 2021). Data were collected between June and October 2019 through personal interviews with structured paper–pencil questionnaires (Yuan et al., 2021). All participants gave their written consents and the study design was approved by the Institutional Review Board of the East China University of Science and Technology. After the initial data entry, all identifying details were removed. Only the principal investigator and a full-time manager have access to the contact details. The data were maintained and managed by the research team at the School of Social and Public Administration and the Institute of Social Work and Social Policy.

Our study population was drawn from the social worker dataset, which originally included 5965 social workers. We focused on those who were at least 16 years of age, and had been working in the current agency for over 3 months at the time of study. This left us with 5741 observations. We removed those with missing data on TI and other key explanatory variables. Our final sample included 5525 social workers.

Measures

TI

TI was measured using a 3-item scale developed by Nissly et al. (2005). Participants responded to the following items: ‘I intend to leave my current organization within the next six months’, ‘I may leave my current organization within the next three years’, and ‘I occasionally have the idea of leaving my current organization’. Each item was rated using a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The total score was summed across the 3-item scale. Higher scores indicated stronger TI. The Cronbach’s alpha for internal consistency and reliability was 0.78.

PS

According to discrepancy theories, predictions of PS are more accurate when they are based on discrepancies rather than on salary received (Rice et al., 1990). Following this theory, the independent variable PS was defined as the discrepancy between the actual and expected salary. It was measured based on the following three questions: ‘What is your monthly salary?’ ‘Are you satisfied with your salary?’ ‘If the answer is no, what is your expected salary?’ If the respondent was satisfied with their salary, the value of PS was 1. If the respondent was not satisfied, the value of PS was calculated by the ratio of actual to the expected salary. The higher the value, the higher the individual’s satisfaction with the present salary.

JS

JS was measured using a revised 5-item scale based on a study of licensed social workers in the United States conducted by the NASW (U.S. Department of Health and Human Services, 2004). Sample items included ‘I love my job more than the average person’, ‘I am passionate about my job most of the time’, and ‘I am quite satisfied with my job’. Each item was rated using a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The total score was summed across the 5-item scale. The higher the score, the greater the satisfaction she or he feels about the job. The Cronbach’s alpha for the scale was 0.86. The confirmatory factor analysis indicated good construct validity, χ²(5) = 149.24, RMSEA = 0.07, CFI = 0.99, TLI = 0.98.

OCM

OCM was measured using a 3-item scale developed based on Schein (1990) and Cameron and Quinn (2011). Participants responded to the following items: ‘I agree with the culture of our organization’, ‘I share the values of our organization’, and ‘I share the same values with most members of our organization’. Each item was rated using a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The total score was summed across the 3-item scale. Higher scores indicated higher OCM. The Cronbach’s alpha for the scale was 0.88.

Control variables

TI among workers can significantly differ based on age, gender, marital status, educational attainment, household registration, years working as social workers, and health status (Kim and Kao, 2014; Matz et al., 2014; Webb and Carpenter, 2012). Age and social work duration were measured in years. The data regarding gender (dummy coded male = 1, female = 0), marital status (dummy coded married = 1, single/divorced/widowed = 0), educational attainment (dummy coded college and above = 1, junior college or below = 0), household registration (dummy coded rural = 1, non-rural = 0), and self-reported health status (dummy coded fair/poor/very poor = 1, good/very good = 0) were recorded as dichotomous variables.

Statistical analysis

First, descriptive statistics were used to evaluate sample characteristics. Pearson’s correlation analysis was conducted to test the overall relationship among study variables. Next, we performed mediation analysis to investigate the mediating effect of JS on the relationship between PS and TI. Finally, we performed moderated mediation analysis to examine whether OCM moderates the indirect effect of PS on TI through JS. Since traditional mediation tests often assume normality in the sampling distribution, a condition that may not always be met in practice. Bootstrap resampling, being a non-parametric approach, circumvents this assumption, making it more robust in situations where the normality assumption is questionable (Aroian, 1947; Sobel, 1982). Therefore, in this article, a bootstrapping procedure was used to test the significance of the indirect effects and the differences in these effects across levels of the moderator variable (conditional indirect effects) with 5000 bootstrap samples (Preacher and Hayes, 2008). Since the moderator (OCM) is a continuous variable, to test statistical significance of the conditional indirect effect, three values of OCM, that is, one standard deviation below the mean, the mean itself, and one standard deviation above the mean, were chosen following previous literature (Edwards and Konold, 2020; Preacher et al., 2007). The 95% confidence intervals for the coefficients calculated by bootstrapping were considered statistically significant if the confidence intervals excluded 0. All analyses were conducted using STATA 15 (Stata Corporation, College Station, TX, the United States).

Results

Table 1 presents the descriptive statistics for the control variables used in the study. Of the 5525 respondents, 79.0% were women. The mean age of the respondents was 30.53 (SD = 7.83), and 48.5% were married. The highest educational level of over 60% of the respondents was college or above (63.6%). About 60% of the respondents held non-rural Hukou, and the mean working duration as social workers was 3.66 years (SD = 3.44). Most respondents self-rated their health status as very good or good (80.0%).

Table 1.

Descriptive statistics of Chinese social workers (N = 5525). Data collected from structured questionnaires between June and October 2019.

Variables	Proportion in percentages or mean (SD)
Age in years	30.53 (7.83)
Gender
Male	21.0
Female (ref)	79.0
Marital status
Married	48.5
Single/Divorced/Widowed (ref)	51.5
Household registration
Rural	39.9
Non-rural (ref)	60.1
Working duration in years	3.66 (3.44)
Education
College or above	63.6
Junior college or below (ref)	36.4
Self-reported health status
Fair/poor/very poor	19.7
Good/very good (ref)	80.3

SD = Standard Deviation.

Table 2 presents the means, standard deviations, and correlations of TI, PS, JS, and OCM. The average score of TI among social workers was 7.67 (SD = 2.68). PS was significantly correlated to JS (r = 0.145, p < 0.05), TI (r = –0.192, p < 0.05) and OCM (r = 0.117, p < 0.05). JS was significantly associated with OCM (r = 0.435, p < 0.05) and TI (r = –0.386, p < 0.05). OCM was negatively related to TI (r = –0.335, p < 0.05).

Table 2.

Summary of mean, SD, and correlation coefficients among turnover intention (TI), pay satisfaction (PS) and job satisfaction (JS), and organizational culture match (OCM) among Chinese social workers (N = 5525). Data collected from structured questionnaires between June and October 2019.

Variables (range)	Mean	SD	(1)	(2)	(3)	(4)
(1) TI (3–15)	7.67	2.68	1.00
(2) PS	0.80	0.20	−0.192*	1.00
(3) JS (5–25)	18.07	3.04	−0.386*	0.145*	1.00
(4) OCM (3–15)	11.69	1.88	−0.335*	0.117*	0.435*	1.00

p < 0.05.

This study examined the mediating effect of JS on the relationship between PS and TI. As can be seen in Table 3, PS was positively related to JS (B = 2.208, p < 0.001) and negatively to TI (B = –2.045, p < 0.001), thus supporting Hypothesis 1. JS was negatively associated with TI (B = –0.282, p < 0.001), thus supporting Hypothesis 1. This study used bootstrapping (Shrout and Bolger, 2002) to verify the significance of the mediating effect of JS on the relationship between PS and TI. 95% confidence intervals (CI) of parameter estimates were calculated by using 5000 data samples from the raw ones (N = 5525). The indirect effect of PS on TI through JS was significant (B = –0.622, SE = 0.064, 95% CI = [–0.747, –0.496]). Thus, the study confirmed that JS mediated the association between PS and TI (supporting Hypothesis 2).

Table 3.

Summary of regression results for mediation effect of job satisfaction (JS) on the relationship between pay satisfaction (PS) and turnover intention (TI) among Chinese social workers (N = 5525). Data collected from structured questionnaires between June and October 2019.

	Model 1		Model 2		Model 3
	TI		JS		TI
Variables	B^a	SE	B^a	SE	B^a	SE
Control variables:
Age (years)	−0.061***	0.006	0.064***	0.007	−0.043***	0.005
Gender (Male = 1)	−0.266**	0.083	0.407***	0.097	−0.152	0.078
Marital status (Married = 1)	−0.574***	0.081	0.275**	0.095	−0.496***	0.077
Education (College or above = 1)	0.322***	0.073	−0.132	0.085	0.285***	0.069
Health status (Fair/poor/very poor = 1)	0.308***	0.085	−0.846***	0.099	0.070	0.081
Household registration (Rural = 1)	0.273***	0.073	0.123	0.085	0.308***	0.069
Years of working as social worker (yrs)	−0.029*	0.011	0.043**	0.013	−0.017	0.011
Independent variables
PS	−2.667***	0.172	2.208***	0.200	−2.045***	0.164
JS					−0.282***	0.011
Bootstrap indirect effect on TI (through JS)^b	B (SE)		LL 95% CI		UL 95% CI
	–0.622* (0.064)		–0.747		–0.496

Unstandardized coefficients are reported.

Bootstrap sample size = 5000. LL 95% CI = lower bound in 95% confidence interval; UL 95% CI = upper bound in 95% confidence interval.

p < 0.05; **p < 0.01; ***p < 0.001.

Table 4 presents the moderating effect of OCM on the relationship between JS and TI. Following Aiken and West (1991), the variables used as a component of the interaction term were mean-centered to deal with multicollinearity. As predicted in Hypothesis 3, OCM was negatively related to TI (B = –0.285, p < 0.001). The results revealed that OCM moderated the relationship between JS and TI (B = –0.046, p < 0.05), which supported Hypothesis 4. The relationship is graphically presented in Figure 2 at one standard deviation above and below the mean of the OCM. The slope of the relationship between JS and TI was relatively strong and weak for social workers with high and low levels of OCM, respectively.

Table 4.

Summary of regression results for moderated mediation model among pay satisfaction (PS), job satisfaction (JS), turnover intention (TI), and organizational cultural match (OCM) among Chinese social workers (N = 5525). Data collected from structured questionnaires between June and October 2019.

Variables	B^a	SE
	JS
Age	0.064***	0.007
Gender (Male = 1)	0.407***	0.097
Marital status (Married = 1)	0.275**	0.095
Education (College or above = 1)	−0.132	0.085
Health status (Fair/poor/very poor = 1)	−0.846***	0.010
Household registration (rural = 1)	0.123	0.085
Years working as social worker	0.043**	0.013
PS	2.208***	0.200
	TI
Age	−0.042***	0.005
Gender (Male = 1)	−0.126	0.077
Marital status (Married = 1)	−0.485***	0.075
Education (College or above = 1)	0.312***	0.068
Health status (Fair/poor/very poor = 1)	0.012	0.079
Household registration (rural = 1)	0.335***	0.068
Years working as social worker	−0.015	0.010
PS	−1.891***	0.161
JS	−0.209***	0.012
OCM	−0.285***	0.019
JS*OCM	−0.046*	0.025
Conditional indirect effect at OCM = M ± 1 SD^b
OCM	Effect (Boot SE)	Boot 95% CI
−1 SD (–1.881)	−1.463 (0.636)	(–2.709, –0.216)
Mean (0.000)	−1.655 (0.756)	(–3.137, –0.173)
+1 SD (1.881)	−1.847 (0.877)	(–3.565, –0.129)

Unstandardized coefficients are reported.

Bootstrap sample size = 5000.

p < 0.05; **p < 0.01; ***p < 0.001.

Figure 2.

Diagram depicting the interaction of job satisfaction (JS) and organizational cultural match (OCM) on turnover intention (TI) among Chinese social worker (N = 5525). Data collected from structured questionnaires between June and October 2019.

We examined the conditional indirect effect of PS on IT through JS for three values of OCM using the bootstrapping technique with 5000 repetitions: one standard deviation below the mean (–1.881), the mean, and one standard deviation above the mean (1.881). We found that the conditional indirect effects based on the three values of OCM were negative and significant (–1SD = –1.463, 95% CI = [–2.709, –0.216]; at mean = –1.655, 95% CI = [–3.137, –0.173]; +1SD = –1.847, 95% CI = [–3.565, –0.129]). Thus, consistent with our expectations, OCM moderated the indirect effect of PS on TI through JS. The indirect and negative effect of PS on TI through JS became stronger as the level of OCM increased. These results fully supported Hypothesis 5.

Discussion

High turnover among social workers has been a long-standing concern the world over because it has severe consequences for agencies, and their staff and clients. We examined the influence of PS on TI among social workers, and hypothesized that JS would mediate the relationship between PS and TI. We explored whether OCM could directly influence TI and strength or weaken the indirect relationship between PS and TI.

First, consistent with previous studies (Lambert et al., 2001; Mudor and Tooksoon, 2011; Tham, 2006; Wang et al., 2021), our results demonstrated that both PS and JS were significantly inversely related to TI, which supported Hypothesis 1. We investigated the role of JS in the relationship between PS and TI. As expected, the mediating effect of JS was significant, which supported Hypothesis 2. Our results show that a high level of PS can promote the JS of social workers who would thus be less likely to have the intention to leave. Given that JS is an affective reaction to a job, our finding that PS affects employees’ TI through JS emphasizes the core mechanism played by affective job reaction between material rewards and employees’ work attitudes. Similar findings were drawn in previous studies (Heavey et al., 2013; Lambert et al., 2001).

Second, human service professionals are often defined as helping and compassionate employees who are primarily inspired by intrinsic motivations and are expected to be indifferent to financial matters (Wermeling and Smith, 2009). However, empirical evidence remains inconsistent with the impact of material rewards on desired employee behavior (Bimpong et al., 2020; Schweitzer et al., 2013; Wang et al., 2021). Even after controlling for the mediator of JS, PS significantly influenced TI. Only 23% of the effect of PS on TI was mediated by JS, which was about 0.3 times as large as the direct effect of PS on TI (data not shown). This indicates that social workers are concerned about their financial rewards, which suggests that money may not be the primary reason an individual chooses to enter the field of social work. However, material rewards are still important in predicting TI once on the job. Another interpretation is that even JS can be increased through various interventions, TI may still be high without an increase in financial rewards.

Third, Hypothesis 3 was confirmed with respect to the main effect of OCM on TI. TI was lower among social workers with high OCM than those with low OCM. This finding was in line with Shim (2010) who noted that organizational culture and climate were significant in explaining the intention to leave among child welfare employees. A possible explanation is that once employees appreciate that their organization shares their values, they develop affective commitment toward their organization and are less likely to seek employment elsewhere (Deci et al., 2017; Verquer et al., 2003).

Fourth, Hypotheses 4 and 5 – the moderating effect of OCM – were supported. As expected, compared to social workers with low OCM, JS had a stronger negative effect on TI for social workers with high OCM (Hypothesis 4). This suggests that if JS is obtained through work, individuals with high OCM are more likely to stay in their jobs than those with low OCM. Even if JS increases, TI among social workers with low OCM will not reduce significantly. We found that the magnitude of the conditional indirect effect of PS on TI was greater for social workers with high OCM (Hypothesis 5). This means that high OCM reinforces the mediating effect of JS on the relationship between PS and TI. A possible explanation is that an individual with high OCM will more actively process and integrate various job-related experiences, including pay, which will affect their overall affection in relation to the job, thus having a stronger impact on TI.

Our results make theoretical and empirical contributions to the literature. Theoretically, according to SDT, extrinsic and autonomous motivations affect employee performance independently (Gagné and Deci, 2005). Our moderated mediation model demonstrated a new relationship between extrinsic (PS) and autonomous (OCM) motivation. In addition to being independent predictors of behavioral outcomes, autonomous motivation moderated the relationship between extrinsic motivation and behavior outcomes. Empirically, less than 40% of social workers felt they were fairly paid globally (Ravalier et al., 2022). Our results emphasized the significant roles of pay and job satisfaction in reducing social workers’ turnover intention. Besides, our study highlighted the importance of individual and organizational culture match, as this factor has not only direct but also indirect effects on turnover among human service professionals.

This study has several limitations that should be acknowledged. First, it was a cross-sectional study that revealed the relationships among pay and JS, OCM, and TI. However, it was unable to determine the causal effects. Thus, longitudinal studies should validate our moderated mediation model. Second, all study variables were measured using a self-reported questionnaire, which means that participants could consistently respond to questions in a certain direction, which could, in turn, bias the results.

Despite these limitations, our findings have important practical implications. First, payment satisfaction is decisive to a social worker’s intention to leave. Social workers should tackle the inner struggle that material rewards are at odds with their professional calling (Healy and Meagher, 2004) and assertively advocate for the improvement of pay and material rewards (Dominelli, 1996; Luo and Chui, 2018; Wang et al., 2021). Second, job satisfaction mediates the relationship between pay satisfaction and intention to leave. Thus, if financial resources are limited, efforts to improve social workers’ job satisfaction can be possible options to reduce turnover intention. Third, our results show that understanding the importance of the motivation/culture match can help organizations make better hiring decisions. By communicating the organizational culture to prospective employees during recruitment, the organization may be able to decrease its future turnover rate. Fourth, adequate communication between the employers and employees on organizational culture should be valued. It can help employees better understand, adapt, and integrate into the organizational culture, thus reducing turnover rate. Fifth, the results of our moderated mediation model demonstrate that organizational culture should be built to support personal psychological needs in order to stimulate the positive effect of material rewards on organizational commitment and other desired work behavior among social service professionals. Last but not least, our findings may have reference values for the development of social work in regions where the profession is still in its establishment phase (Kaloga and Camara, 2023; Nikku and Pulla, 2014). If a certain level of material rewards cannot be provided due to limited financial resources, strategies to increase job satisfaction and organizational culture match can be adopted to reduce turnover intention among social workers. In general, researchers and practitioners should pay more attention to not only the material rewards but also the organizational culture that affects the well-being of social workers and the development of the profession.

Footnotes

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

All participants have provided written consent and the study design was approved by the Institutional Review Board of the East China University of Science and Technology.

ORCID iD

Xin Li

Yiqing Yuan

Author biographies

Xin Li is an Associate Professor whose research interests focus on productivity loss due to illness, occupational behavior, and work-related health.

Yiqing Yuan is a social work doctoral student. She got her MSW at NYU. Her research area covers professional development and social determinants of health.

Xuesong He is a full professor of social work whose research interests focus on social work theory and social work develoment.

References

Aiken

West

(1991) Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage.

Alexander

Lichtenstein

, et al (1998) A causal model of voluntary turnover among nursing personnel in long-term psychiatric settings. Research in Nursing & Health 21(5): 415–427.

Aroian

(1947) The probability function of the product of two normally distributed variables. The Annals of Mathematical Statistics 18(2): 265–271.

Barak

MEM

Michàl

Nissly

, et al (2001) Antecedents to retention and turnover among child welfare, social work, and other human service employees: What can we learn from past research? A review and metanalysis. Social Service Review 75(4): 625–661.

Bimpong

KAA

Khan

Slight

, et al (2020) Relationship between labour force satisfaction, wages and retention within the UK national health service: A systematic review of the literature. BMJ Open 10(7): e034919.

Bosomtwe

Obeng

(2018) The link between organizational culture and turnover intention among employees in Ghana. International Journal of Contemporary Research and Review 9(8): 20951–20958.

Cameron

Quinn

(2011) Diagnosing and Changing Organizational Culture: Based on the Competing Values Framework (3rd edn). Hoboken, NJ: Jossey-Bass.

Carmeli

Freund

(2009) Linking perceived external prestige and intentions to leave the organization: The mediating role of job satisfaction and affective commitment. Journal of Social Service Research 35(3): 236–250.

Carmeli

Weisberg

(2006) Exploring turnover intentions among three professional groups of employees. Journal of Human Resource Development International 9(2): 191–206.

10.

Chan

(2018) The mediating effects of job satisfaction and organizational commitment on turnover intention, in the relationships between pay satisfaction and work–family conflict of casino employees. Journal of Quality Assurance in Hospitality & Tourism 20(2): 206–229.

11.

Deci

Olafsen

Ryan

(2017) Self-determination theory in work organizations: The state of a science. Annual Review of Organizational Psychology and Organizational Behavior 4(1): 19–43.

12.

Denison

(1984) Bringing corporate culture to the bottom line. Organizational Dynamics 13(2): 4–22.

13.

Dominelli

(1996) Deprofessionalizing social work: Anti-oppressive practice, competencies and postmodernism. British Journal of Social Work 26(2): 153–175.

14.

Edwards

Konold

(2020) Moderated mediation analysis: A review and application to school climate research. Practical Assessment, Research, and Evaluation 25(5): 5.

15.

Egan

Yang

Bartlett

(2004) The effects of organizational learning culture and job satisfaction on motivation to transfer learning and turnover intention. Human Resource Development Quarterly 15(3): 279–301.

16.

Emerson

(2013) Organizational culture, job satisfaction and turnover intentions: The mediating role of perceived organizational support. PhD Thesis, Virginia Commonwealth University, Richmond, VA.

17.

Gagné

Deci

(2005) Self-determination theory and work motivation. Journal of Organizational Behavior 26(4): 331–362.

18.

Healy

Meagher

(2004) The reprofessionalization of social work: Collaborative approaches for achieving professional recognition. British Journal of Social Work 34(2): 243–260.

19.

Heavey

Holwerda

Hausknecht

(2013) Causes and consequences of collective turnover: A meta-analytic review. Journal of Applied Psychology 98(3): 29–37.

20.

Hulin

Judge

(2003) Job attitudes. In: Weiner

(ed.) Handbook of Psychology. New York: John Wiley, pp. 255–276.

21.

Jiang

Wang

Chui

, et al (2019) Professional identity and turnover intentions of social workers in Beijing, China: The roles of job satisfaction and agency type. International Social Work 62(1): 146–160.

22.

Kaloga

Camara

(2023) A case study of the emerging social work sector in Guinea, West Africa. Social Work Education 42(2): 185–201.

23.

Kessler

(2014) The effect of organizational culture on employee’s turnover intention in Israel. Annals of the University of Oradea, Economics Science Series 23(1): 1019–1027.

24.

Kim

Kao

(2014) A meta-analysis of turnover intention predictors among U.S. child welfare workers. Children & Youth Services Review 47(P3): 214–223.

25.

Lambert

Hogan

Barton

(2001) The impact of job satisfaction on turnover intent: A test of a structural measurement model using a national sample of workers. Social Science Journal 38(2): 233–250.

26.

Lei

Luo

Chui

, et al (2018) Whether professional training matters: Attitudinal antecedents to the turnover intentions of social workers in Guangzhou, China. Journal of Social Service Research 45(3): 444–454.

27.

Leung

(2007) An international definition of social work for China. International Journal of Social Welfare 16(4): 391–397.

28.

Locke

(1976) The nature and causes of job satisfaction. In: Dunnette

(ed.) Handbook of Industrial and Organizational Psychology. Chicago, IL: Rand McNally, pp. 1297–1343.

29.

Luo

Chui

(2018) Will material interest make social workers quit their job? A meta-analysis. Journal of Social Work 20(3): 340–364.

30.

Matz

Woo

Kim

(2014) A meta-analysis of the correlates of turnover intent in criminal justice organizations: Does agency type matter? Journal of Criminal Justice 42(3): 233–243.

31.

Medina

(2012) Job satisfaction and employee turnover intention: What does organizational culture have to do with it? Master’s Thesis, Columbia University, New York.

32.

Meng

Gray

Bradt

, et al (2018) Emergence of social work practice in rural China: A way forward? International Social Work 62(2): 933–943.

33.

Miceli

Lane

(1991) Antecedents of pay satisfaction: A review and extension. In: Rowland

Ferris

(eds) Research in Personnel and Human Resources Management. Greenwich, CT: JAI Press, pp. 235–309.

34.

Ministry of Civil Affairs (2019) 2019 ‘Quantifying’ iteration of social work development. Available at: http://www.mca.gov.cn/article/gk/tjtb/201903/20190300015629.shtml (accessed 1 March 2023).

35.

Mudor

Tooksoon

(2011) Conceptual framework on the relationship between human resource management practices, job satisfaction and turnover. Journal of Economics and Behavioral Studies 2(2): 41–49.

36.

Nikku

Pulla

(2014) Global agenda for social work and social development: Voices of the social work educators from Asia. International Social Work 57(4): 373–385.

37.

Nissly

Barak

MEM

Levin

(2005) Stress, social support, and workers’ intentions to leave their jobs in public child welfare. Administration in Social Work 29(1): 79–100.

38.

Preacher

Hayes

(2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods 40(3): 879–891.

39.

Preacher

Rucker

Hayes

(2007) Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research 42(1): 185–227.

40.

Rathakrishnan

Imm

Kok

(2016) Turnover intentions of lecturers in private universities in Malaysia. Pertanika Journal of Social Sciences and Humanities 24(Suppl.): 129–145.

41.

Ravalier

Jones

Truell

, et al (2022) Global social work working conditions and wellbeing. International Social Work 65(6): 1078–1094.

42.

Rice

Phillips

McFarlin

(1990) Multiple discrepancies and pay satisfaction. Journal of Applied Psychology 75(4): 386–393.

43.

Rigby

Ryan

(2018) Self-determination theory in human resource development: New directions and practical considerations. Advances in Developing Human Resources 20(2): 133–147.

44.

Ryan

Deci

(2002) Overview of self-determination theory: An organismic-dialectical perspective. In: Deci

Ryan

(eds) Handbook of Self-Determination Research. Rochester, NY: University of Rochester Press, pp. 3–33.

45.

Schein

(1990) Organizational culture. American Psychologist 45(2): 109–119.

46.

Schweitzer

Chianello

Kothari

(2013) Compensation in social work: Critical for satisfaction and a sustainable profession. Administration in Social Work 37(2): 147–157.

47.

Shim

(2010) Factors influencing child welfare employee’s turnover: Focusing on organizational culture and climate. Children and Youth Services Review 32(6): 847–856.

48.

Shrout

Bolger

(2002) Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods 7(4): 422–425.

49.

Silverthorne

(2004) The impact of organizational culture and person-organization fit on organizational commitment and job satisfaction in Taiwan. Leadership Organization Development Journal 25(7): 592–599.

50.

Singh

Loncar

(2010) Pay satisfaction, job satisfaction and turnover intent. Industrial Relations 65(3): 470–490.

51.

Sobel

(1982) Asymptotic confidence intervals for indirect effects in structural equation models. Sociological Methodology 13: 290–312.

52.

Spector

(1985) Measurement of human service staff satisfaction: Development of the job satisfaction survey. American Journal of Community Psychology 13(6): 693–713.

53.

Sun

Zhou

(2019) Work conditions, family support and professional development: Multifactor analysis of Chinese social workers’ turnover intention. Social Work and Management 19(4): 29–36. (In Chinese)

54.

Tett

Meyer

(1993) Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology 46(2): 259–293.

55.

Tham

(2006) Why are they leaving? Factors affecting intention to leave among social workers in child welfare. British Journal of Social Work 37(7): 1225–1246.

56.

U.S. Department of Health and Human Services (2004) A study of the rules and use of licensed social workers in the United States. Available at: https://aspe.hhs.gov/system/files/pdf/125736/rulsw_0.pdf

57.

Vandenberghe

(1999) Organizational culture, person-culture fit, and turnover: A replication in the health care. Journal of Organizational Behavior 20(2): 175–184.

58.

Verquer

Beehr

Wagner

(2003) A meta-analysis of relations between person–organization fit and work attitudes. Journal of Vocational Behavior 63(3): 473–489.

59.

Wang

Zheng

Sui

, et al (2021) For money or identity – or both? Which could promote the retention of social workers in China? The British Journal of Social Work 51(2): 524–544.

60.

Wang

Jiang

Zhang

, et al (2020) Organizational justice, burnout, and turnover intention of social workers in China. Journal of Social Work 21(3): 456–475.

61.

Webb

Carpenter

(2012) What can be done to promote the retention of social workers? A systematic review of interventions. British Journal of Social Work 42(7): 1235–1255.

62.

Wermeling

Smith

(2009) Retention is not an abstract notion: The effect of wages and caretaking. Journal of Social Service Research 35(4): 380–388.

63.

Yuan

Liu

Sun

, et al (2021) Baseline survey of China social work longitudinal study 2019: Design and implementation. Research on Social Work Practice 31(5): 513–519.