Role Stress,Burnout,and Workplace Support Among Newly Recruited Social Workers

Abstract

Purpose:

This study examines the mediation and moderation roles of workplace support in the effect of role stress on burnout among newly recruited social workers in mainland China.

Method:

A total of 1,638 newly recruited social workers, those in their first year of professional employment, are invited as participants to complete a questionnaire package, which includes their demographic information, role stress, workplace support, and Maslach’s Burnout Inventory.

Results:

Role stress significantly correlates with emotional exhaustion, depersonalization, and decreased personal accomplishment of burnout. Workplace support moderates the effects of role stress on three syndromes of burnout. In addition, role stress can lead to burnout by reducing perceptions of workplace support.

Conclusions:

Workplace support is important in buffering the effect of role stress on burnout. Results suggest interventions on improving workplace support to alleviate role stress and burnout of new social workers would be beneficial.

Keywords

newly recruited social worker role stress burnout workplace support

With its historical and social background, social work in mainland China experienced a course of emergence, suspension, and revival in the last century. Since 2003, the Ministry of Civil Affairs, who is responsible for social welfare and service provisions, began to recruit social work personnel into its subordinate departments and bureaus across the country to meet the rising needs of society; in addition, this step provides impetus to promote formal social work training in higher education and vocational standard assessment for social workers (Yan & Cheung, 2006). Given the late development of social work in China, the workforce quality and stability need improvement. In this profession, China experiences workforce shortages in terms of recruitment and retention of social workers due to low salary, unstable career prospects, and lack of public awareness and acceptance (Gao & Yan, 2015). Majority of practitioners fall into the age range between 18 and 30, but a considerable proportion of social workers only serve in the profession for less than 2 years (Li, 2016). Determining how to improve the work experience and sustain the employment stability of new social workers is key for the long-term development of this field.

As a psychological response to work condition, burnout has been identified as a key risk factor that may lead to various negative consequences, including poor service quality and intention to leave (Kim et al., 2011; Travis et al., 2016). The term “burnout” was first applied to describe a work-related syndrome wherein a practitioner becomes increasingly “inoperative” (Freudenberger, 1975). Maslach et al. (1986) conceptualized burnout into three components as follows: emotional exhaustion (EE; feelings of emotional overextension and depleted emotional and physical resources), depersonalization (DP) or cynicism (negative or excessively detached responses toward the recipients of one’s service or care and various work aspects), and decreased personal accomplishment (PA; decline in one’s feelings of competence and successful achievement at work). People in professions such as social work have been proven to frequently have greater exposure to burnout than other occupations (Collins et al., 2010; Leiter & Maslach, 1988; Shen, 2008). Burnout has serious consequences and may harm the physical and mental health of social workers, affect the service quality and efficiency, and prompt workers to leave the profession. Recently, new recruits in human service professions, such as nurses and teachers, are at particularly high risk of suffering burnout in unsupportive work environments (Høigaard et al., 2012; Rudman & Gustavsson, 2011). Younger social workers likewise report higher burnout than their older counterparts (Schwartz et al., 2007). However, little is known about the situations, factors, and mechanisms of burnout among new social work practitioners, which is crucial for their long-term career development and retention in this profession.

The most recognized theoretical model regarding job-related factors that contribute to burnout among social workers is the job demands–resources (JD-R) model (Demerouti et al., 2001). As an important job demand, role stress is consistently found related to burnout in various occupations (Azeem et al., 2014; Garrosa et al., 2011; Richards et al., 2016). Role stress is characterized by role ambiguity (uncertainty about aspects of the job, including objectives, priorities, and expectations) and role conflict (degree to which the job or task is characterized by conflicting requirements or competing values). High role stress exhausts individual mental and physical resources and therefore leads to depleted energy and compassion and reduced work accomplishment. Experience and adaptation offset the work stress that can otherwise lead to burnout (Prosser et al., 1999). New social workers are often young, and their first year of recruitment is very challenging in terms of adapting to work conditions, gaining the skills and confidence required to establish their careers, and reinforcing effective practice. Moreover, social workers operate under high pressure and very demanding environments. For social workers, the first years are challenging because they are embarking on careers that are highly client-based and becoming involved in complex social and emotional contexts. Accordingly, during the transition of their initial year, they are more likely to experience high levels of role stress and may lead to burnout. Therefore, role stress is associated with burnout among social work newcomers.

In the unique context of China, the discrepancy between formal education and field requirement, diverse education background of practitioners, and strong government involvement are macro-level factors that may create role stress for newly recruited social workers. A long-standing debate discusses the application of Western social work values and ethics to social work education in China (Hutchings & Taylor, 2007). Although indigenization is believed to be a primary task for social work development in the unique social–political–cultural context of China (Yan & Tsang, 2005), the theories, techniques, value system, practice models, and teaching pedagogy are still largely rooted in Western practice. The importance paid to theoretical learning rather than training practical skills possibly creates role stress for new social work graduates, who need to collaborate with a considerable proportion of practitioners with little formal social work training or practice credentials (Jiang et al., 2019). Such collaborations are stress-producing because diversity in values and knowledge may cause disputes on goals, methods, and priority regarding practice (Law & Gu, 2008; Li et al., 2019). Moreover, the majority of social service projects, within or independent of political infrastructure, are believed to be under strong government influence due to funding regulations (Cortis et al., 2018). Therefore, the tension between professionalism and administrative influence can potentially cause practitioners to struggle with the fundamental question on how to define and carry out social work in China and what their professional role entails (Leung, 2012).

The JD–R model specifies that job demands and resources interact with each other in work-related outcomes (Schaufeli & Taris, 2014). Workplace or organizational support is often identified as an important job resource that enables individuals to cope with stress and protect against burnout (AbuAlRub, 2004; Bakker et al., 2005; Hamama, 2012; Hombrados-Mendieta & Cosano-Rivas, 2013; Lloyd et al., 2002). Help, guidance, feedback, and appreciation from coworkers and supervisors create a supportive work environment conducive to providing high-quality services. In addition, these types of support enable new recruits to derive gratification, self-esteem, security, and positive feelings from their jobs; they also develop professional identity and an attachment to their organizations, possibly buffering the adverse effect of stress. However, existing literature show controversy on their underlying relations. For instance, Coady et al. (1990) found no significant relationship between social workers’ perception of team support on EE and DP in spite of the significant association between support and PA. Similarly, no significant association was found between workplace support and burnout among social workers (Kim & Stoner, 2008). Other studies (Jenkins & Elliott, 2004; Mutkins et al., 2011; Park et al., 2004) found no significant buffering effect of workplace support on alleviating work stress or burnout. Therefore, further clarifications are necessary concerning the impact of workplace support, such as the specific effects of support on the three components of burnout and its moderating effect on the relations between role stress and burnout. Findings can assist in the development of intervention programs that aim to enhance the supportive networks of new recruits.

The negative consequences of role stress on burnout and the importance of workplace support as a moderator have been extensively studied. By contrast, little is known about the mediation mechanism of how role stress generates burnout. Perceived support refers to the perceptions on the availability and adequacy of support and satisfaction with its amount or quality (Lin, 1986). According to the social support deterioration model proposed by Barrera (1986), stress erodes the perceived availability or effectiveness of support, which leads to psychological distress and impairment. Therefore, to expect that role stress results in burnout by reducing the perception of workplace support is reasonable. That is, individuals who experience stress from role conflict and unclear job tasks are prone to perceive others in their job settings as unwilling or unavailable to help in times of stress, resulting in a lack of a critical buffer against burnout.

Considering the workplace support for social workers in China may be beneficial in understanding its significance as a key factor worthy of notice. Supervision is the most typical form of support in the field. In China, many social service organizations hire experts from Hong Kong and Taiwan or social work teachers from higher education institutions to provide supervision services for frontline social workers. Administrators, managers, and peers are also encouraged to become internal supervisors for younger and less experience colleagues (Qi & Shen, 2012). In theory, supervisors are responsible for providing administrative, educational, and professional support: Administrative support includes guidance of practice protocol and procedures of specific organizations; educational support refers to discussions on using theories, techniques, and interventions in practice; and emotional support involves efforts to address the work-related psychological issues for practitioners (Kadushin & Harkness, 2014). Notably, local scholars are particularly interested in how to ease the tension between Western social work practice model and the infrastructural and institutional requirements in Chinese reality (Tong et al., 2018; Xv, 2019). To a certain extent, new practitioners can better adapt to the work environment and understand the profession when they receive support from administrators, managers, supervisors, and peers.

In summary, social work in China faces the unsatisfactory quality of personnel, recruitment, and retention challenges alongside increasing stress and burnout issues. Hence, creating work environments that retain newcomers to the profession and sustain the future workforce is important. In this study, we focus on role stress and burnout among new recruits during their initial year in social work. The unique context of social work development in China also calls for explorations of whether, how, and why support from organizations help social workers to adapt to the unique challenges associated with role stress and to ease burnout. This study aims to utilize the JD–R and social support deterioration models to examine the mediation and moderation effects of perceived workplace support on the relationship between role stress and burnout.

Method

Participants and Procedures

Considering social work profession development trends, China Social Work Longitudinal Study (CSWLS) is a large-scale national survey carried out from June to October 2019 among social work institutions and social workers. This project was approved by East China University of Science and Technology (Yuan et al., 2020). A total of 6,776 valid questionnaires are obtained from across 57 cities in China through random sampling design. The surveyed full-time social workers include (1) those who offer professional services and have been employed in a social work institution for over 3 months (interns, volunteers, and administrative staff were excluded), (2) those affiliated with civil affairs sectors or quasi-government welfare organizations, and (3) medical social workers employed in hospitals. Participants filled out the questionnaire via pencil-and-paper within 45–60 min.

This study focused on newly recruited social workers. A total of 1,638 participants (525 males and 1,313 females) in their first year of social work services were included for data analysis. Age ranged between 18 and 68 years and the mean age was 27.56 (standard deviation [SD] = 6.98). Among the participants, only 9% (n = 147) had a high school degree or less and 17% (n = 278) had social work professional qualifications (i.e., social worker assistant, social worker, and senior social worker). The average length of service in social work is 6.97 months (SD = 3.76). Table 1 presents further details on the sample profile.

Table 1.

Sample Profile.

Variable	Frequency	Percentage
Gender
Male	325	19.8
Female	1,313	80.2
Ethnic background
Minority	106	6.5
Han	1,531	93.5
Missing	1	0.1
Religion
Yes	86	5.3
No	1,550	94.6
Missing	2	0.1
Marital status
Single	1,083	66.1
Married	530	32.4
Divorce	21	1.3
Widow	3	0.2
Missing	1	0.1
Education
Middle school and below	11	0.7
High school	136	8.3
Diploma college	549	33.5
Bachelor	830	50.7
Postgraduate	104	6.3
Missing	8	0.5
Health
Very good	684	41.8
Good	680	41.5
Average	271	16.5
Bad	2	0.1
Missing	1	0.1
Age cohort (years)
25 and below	914	55.8
26–35	514	31.4
36–45	153	9.3
46–55	51	3.1
56 and above	6	0.4
Qualification of social work
No	1,354	82.7
Yes	278	17.0
Missing	6	0.3

Note. N = 1,638.

Measures

Role stress

The measure of role stress is adapted from the shortened role conflict/role ambiguity questionnaire devised by Rizzo et al. (1970) and role clarity measure developed by Haynes et al. (1999). Participants are asked to respond to Role Conflict (nine items, e.g., “I often receive conflicting instructions from two or more persons”) and Role Ambiguity (five items, e.g., “I have clear planned goals for my job”) subscales. Participant responses indicate the degree to which the condition applied to them on a 5-point scale ranging from strongly disagree (1) to strongly agree (5). Exploratory factor analyses (with principle component analysis and varimax rotation method) are carried out to confirm the two-factor structure of the role stress instrument (Kaiser-Meyer-Olkin = .859; Bartlett p < .001; factor loadings = .517–.838; 48.3% of the explained variance). The internal reliabilities are .832 for role conflict, .798 for role ambiguity, and .712 for the total scale. The Role Ambiguity subscale is inverted and recoded hereafter and then total role stress is the average score of all the items, with a high score representing greater role stress.

Workplace support

This measure is adapted from that of House and Wells’ (1978) social support, which is widely used in work settings in terms of the variety of sources and types of social support (Deeter-Schmelz & Ramsey, 1997). In most Chinese social work organizations, the administrative level leads to four kinds of sources of workplace support: top manager (or administrator), immediate manager (or director), mentor (or professional supervisor), and coworkers. Participants are asked to rate the extent to which these people provided them with support. All items were rated on a 5-point Likert-type scale ranging from not at all (0) to very much (4). The overall score was computed by averaging all 24 questions (six questions for each source). Cronbach’s α for this measure is .956.

Burnout

Burnout among new recruited social workers was assessed using Maslach’s Burnout Inventory–Human Service Survey (MBI–HSS), which is widely tested across countries and languages, including its internal reliability, test–retest reliability, and validities (Maslach et al., 1996). MBI–HSS is also translated and validated among Chinese samples (Pang et al., 1997). This inventory is a 22-item questionnaire with three subscales: EE (nine items), DP (five items), and PA (eight items). Using a 7-point Likert-type scale from never (0) to every day (6), participants are asked to rate the frequency with which they experience feelings related to the three aspects of burnout. Cronbach’s α values for this study are .919 for EE, .816 for DP, and .927 for PA.

Data Analysis

Data analyses were conducted in SPSS Version 24.0. First, independent sample t tests and Pearson correlations were performed to show the simple associations between sample characteristics (such as gender, age, and length of service) and the components of burnout. Then, descriptive statistics such as means and SDs as well as Pearson correlations of the main variables of interest were conducted. Hierarchical linear regression analyses were further performed to test the robustness of these associations and to determine the contributions of role stress and workplace support in the variances of burnout symptoms. Finally, Model 4 and Model 1 in the macro PROCESS on SPSS were, respectively, used to test the mediation and moderation effects of workplace support in the relationship between role stress and burnout.

Results

Descriptive Statistics and Correlation Analysis

Table 2 presents the analysis on simple associations between sample characteristics and burnout. Differences of burnout in gender, religion, ethnic background, marital status, and qualification are tested using independent sample t tests. For instance, males reported significantly lower EE and higher DP than females, but no significant gender differences are found in PA. People with professional qualifications for social work reported significantly higher levels of DP than those without qualifications. However, people with different religions and ethnic background show no significant difference in burnout levels. Correlation analyses are likewise carried out between age, health status, education, and length of service and burnout. As previously mentioned, newly recruited social workers are inclined to encounter burnout at an increasing trend in the 1st year of recruitment (Fong et al., 2016). Thus, the specific length of service in social work (months) was calculated for the correlation analysis with the characteristics of burnout. Table 2 shows that the length of service is significantly and positively associated with EE (r = .07, p < .01) and DP (r = .05, p < .01) but negatively associated with PA (r = −.09, p < .01). That is, as the months of service as social workers increased, the levels of EE and DP increased whereas the level of PA decreased. Table 2 presents further details.

Table 2.

Simple Associations Between Sample Characteristics and Components of Burnout.

Independent Sample t Tests of Gender, Religion, Ethnic Background, Marital Status, and Qualification
Independent Variable (IV)		EE				DP				PA
Independent Variable (IV)		M (SD)	t	p	Cohen’s d	M (SD)	t	p	Cohen’s d	M (SD)	t	p	Cohen’s d
Gender	Male	11.45 (9.83)	−2.29	.02	.14	2.03 (3.71)	2.21	.05	.13	28.91 (8.97)	. 51	.60	.03
Gender	Female	12.81 (9.48)				1.60 (3.04)				28.64 (8.35)
Religion	No	12.50 (9.58)	1.08	.28	.12	1.65 (3.14)	1.59	.12	.20	30.22 (8.55)	1.72	.08	.19
Religion	Yes	13.65 (9.06)				2.33 (3.86)				28.61 (8.47)
Ethnic	Han	12.45 (9.54)	1.54	.12	.02	1.66 (3.18)	1.37	.17	.13	28.80 (8.39)	−1.88	.06	.21
Background	Minority	13.94 (9.73)				2.09 (3.27)				27.01 (8.48)
Qualification	No	12.33 (9.62)	−1.77	.07	.12	1.60 (3.08)	−2.01	.04	.13	28.79 (8.55)	1.02	.31	.07
Qualification	Yes	13.45 (9.08)				2.01 (3.45)				28.21 (8.08)
Marital status	Married	9.80 (8.17)	8.08	.00	.45	1.21 (2.71)	4.24	.00	.23	28.74 (9.42)	−.14	.89	.01
	Single (including widow, divorce)	13.85 (9.90)				1.92 (3.37)				28.68 (7.99)
Pearson correlations between age, health, education, length of service and burnout (r)
IV			EE				DP				PA
Age			−.22**				−.12**				.03
Health			.16**				.04				−.09**
Education			.14**				.07**				−.01
Length of service			.07**				.05*				−.09**

Note. EE = emotional exhaustion; DP = depersonalization; PA = personal accomplishment.

*p < .05. **p < .01.

Table 3 presents the means and standard deviations of each variable of interest, the Pearson correlations among all variables, and the Cronbach’s α values of the measures used. Role stress was significantly associated with the three aspects of burnout, specifically, positively related to EE (r = .41, p < .01) and DP (r = .33, p < .01) and negatively related to PA (r = –.28, p < .01). Workplace support has a significantly negative correlation with role stress (r = –.39, p < .01), EE (r = –.27, p < .01), and DP (r = –.26, p < .01) and positively correlated with PA (r = .17, p < .01). Role stress and workplace support were negatively correlated (r = −.39, p < .01), that is, people feel greater role stress tend to report less workplace support.

Table 3.

Descriptive Statistics and Correlations of Main Variables.

Variable	1	2	3	4	5	Mean	SD	Cronbach’s α
1. Role stress	1					2.57	0.48	.71
2. Workplace support	−.39**	1				4.22	0.74	.96
3. Emotional exhaustion	.41**	−.29**	1			12.50	9.56	.92
4. Depersonalization	.33**	−.26**	.54**	1		1.68	3.19	.82
5. Personal accomplishment	−.28**	.17**	−.15**	−.21**	1	28.69	8.48	.93

*p < .05. **p < .01.

Hierarchical Regression Analyses

Hierarchical regression analyses are carried out in SPSS to examine the relative contributions of role stress and workplace support on three components of burnout. Therefore, three sets of regression models with all the predictors are applied in three steps. For example, in the first set of regression model, EE is the dependent variable. In Step 1, the demographic variables that demonstrate significant correlation with EE (i.e., gender, age, education, and length of service) are included. Subsequently, role stress is included in Step 2 and workplace support is applied in Step 3. Table 4 shows that when predicting EE in the last step, workplace support showed significant contribution (B = −1.72, p < .001). In addition, the regression coefficient of role stress decreased from 7.78 in Step 2 but remained significant (B = 6.77, p < .001). Meanwhile, R ² increased when workplace support is added into the model (▵R ²= .01, p < .001). Results indicate workplace support explains greater variance of EE than demographic variables and role stress. The same patterns are found in the other two sets of regression models in predicting DP and PA. Tables 4 –6 present further details.

Table 4.

Hierarchical Regression Analyses of Role Stress and Workplace Support on Emotional Exhaustion (EE).

Independent Variable	Dependent Variable: EE
	Step 1				Step 2				Step 3
	B	SE	β	95% CI for B	B	SE	β	95% CI for B	B	SE	β	95% CI for B
Constant	8.64***	2.09		[4.58, 12.69]	−10.88**	2.24		[−15.27, −6.48]	−0.23	2.93		[−6.00, 5.53]
Gender	1.55**	0.58	.07	[0.42, 2.69]	1.73***	0.53	.07	[0.68, 2.77]	1.63**	0.53	.07	[0.59, 2.66]
Age	−0.21***	0.04	−.15	[−0.29, −0.12]	−0.17***	0.04	−.12	[−0.25, −0.09]	−0.16***	0.04	−.12	[−0.24, −0.09]
Health	2.10***	0.31	.16	[1.48, 2.71]	1.66***	0.29	.13	[1.09, 2.23]	1.66***	0.29	.13	[1.09, 2.22]
Education	0.81*	0.32	.07	[0.18, 1.42]	0.56	0.29	.05	[−0.01, 1.14]	0.43	0.29	.04	[−0.14, 1.01]
Marital status	−1.80**	0.65	−.09	[−3.07, −0.53]	−1.48*	0.60	−.07	[−2.66, −0.31]	−1.45*	0.59	−.07	[−2.61, −0.29]
Qualification	0.24	0.64	.01	[−1.01, 1.48]	0.09	0.59	.00	[−1.06, 1.24]	−0.09	0.58	−.00	[−1.23, 1.05]
Length of service	0.12**	0.06	.05	[0.00, 0.24]	0.07	0.06	.03	[−0.04, 0.18]	0.04	0.06	.01	[−0.08, 0.15]
Role stress					7.78***	0.47	.37	[6.86, 8.70]	6.77***	0.50	.32	[5.79, 7.75]
Workplace support									−1.72***	0.31	−.13	[−1.11, −0.29]
R ²			.09				.23				.24
ΔR²			—				.13***				.01***

*p < .05. **p < .01. ***p < .001.

Table 5.

Hierarchical Regression Analyses of Role Stress and Workplace Support on Depersonalization (DP).

Independent Variable	Dependent Variable: DP
	Step 1				Step 2				Step 3
	B	SE	β	95% CI for B	B	SE	β	95% CI for B	B	SE	β	95% CI for B
Constant	2.79***	.70		[1.41, 4.16]	−2.71***	.78		[−4.24, −1.17]	1.19	1.02		[−0.81, 3.19]
Gender	−0.36	.20	−.05	[−0.75, 0.02]	−0.31	.19	−.04	[−0.67, 0.06]	−0.35	0.18	−.04	[−0.71, 0.01]
Age	−0.05***	.02	−.11	[−0.08, −0.02]	−0.04**	.01	−.08	[−0.06, −0.01]	−0.04*	0.01	−.08	[−0.06, −0.01]
Health	0.20	.11	.05	[−0.01, 0.41]	0.08	.10	.02	[−0.12, 0.28]	0.07	0.10	.02	[−0.12, 0.27]
Education	0.07	.11	.02	[−0.15, 0.28]	0.01	.10	.00	[−0.19, 0.21]	−0.04	0.10	−.01	[−0.24, 0.16]
Marital status	−0.16	.22	−.02	[−0.59, 0.28]	−0.07	.21	−.01	[−0.48, 0.34]	−0.06	0.21	−.01	[−0.46, 0.35]
Qualification	0.34	.21	.04	[−0.08, 0.76]	0.28	.20	.03	[−0.12, 0.68]	0.21	0.20	.03	[−0.18, 0.61]
Length of service	0.03	.02	.04	[−0.01, 0.08]	0.02	.02	.02	[−0.02, 0.06]	0.01	0.02	.01	[−0.03, 0.05]
Role stress					2.18***	.16	.32	[1.86, 2.50]	1.80**	0.17	.26	[1.46, 2.14]
Workplace support									−0.63**	0.11	−.15	[−0.84, −0.42]
R ²			.03				.12				.14
▵R ²			—				.10***				.02***

*p < .05. **p < .01. ***p < .001.

Table 6.

Hierarchical Regression Analyses of Role Stress and Workplace Support on Personal Accomplishment (PA).

Independent Variable	Dependent Variable: PA
	Step 1				Step 2				Step 3
	B	SE	β	95% CI for B	B	SE	β	95% CI for B	B	SE	β	95% CI for B
Constant	29.92	1.90		[26.19, 33.64]	42.58	2.14		[38.38, 46.78]	38.35	2.83		[32.81, 43.90]
Gender	−0.01	0.53	−.00	[−1.05, 1.03]	−0.12	0.51	−.01	[−1.12, 0.88]	−0.07	0.51	−.00	[−1.07, 0.93]
Age	0.07	0.04	.06	[−0.01, 0.15]	0.05	0.04	.04	[−0.03, 0.12]	0.05	0.04	.04	[−0.03, 0.12]
Health	−1.07***	0.29	−.09	[−1.63, −0.50]	−0.77**	0.28	−.07	[−1.32, −0.22]	−0.77**	0.28	−.07	[−1.31, −0.22]
Education	0.09	0.29	.01	[−0.48, 0.66]	0.24	0.28	.02	[−0.32, 0.79]	0.29	0.28	.03	[−0.26, −0.84]
Marital status	−0.73	0.60	−.04	[−1.90, 0.44]	−0.93	0.58	−.05	[−2.06, 0.20]	−0.94	0.57	−.05	[−2.07, 0.19]
Qualification	−0.39	0.58	−.02	[−1.54, 0.75]	−0.24	0.56	−.01	[−1.34, 0.87]	−0.17	0.56	−.01	[−1.27, 0.93]
Length of service	−0.19**	0.06	−.08	[−0.30, −0.08]	−0.16**	0.05	−.07	[−0.26, −0.05]	−0.15***	0.06	−.07	[−0.25, −0.04]
Role stress					−5.06***	0.45	−.27	[−5.94, −4.18]	−4.65***	0.48	−.25	[−5.59, −3.71]
Workplace support									0.69	0.30	.06	[0.10, 1.27]
R ²	.02				.09				.10
▵R ²	—				.07***				.01***

*p < .05. **p < .01. ***p < .001.

Mediation Analyses

Based on previous analyses, the mediation and moderation effects of workplace support are examined using the SPSS macro PROCESS, a regression-based modeling tool with bootstrap approach (Hayes, 2013) that has received increasing interest and application in social science field in recent years. Specifically, the PROCESS Model 4 (Hayes, 2017) is carried out to examine the mediation effect of workplace support on the relationship between role stress and the three components of burnout. Significant demographic variables such as age, gender, health, and length of service in each model are controlled as covariates for possible influences. Three models are built with 5,000 bootstrap resamples, and the significant mediation effect is detected when the bias-corrected 95% confidence interval (CI) does not contain 0.

Table 7 presents the direct and indirect effects of role stress on the three components of burnout through workplace support. For instance, role stress has a significant direct effect of on EE (B = 6.89, 95% CI [5.91, 7.86]) and indirect effect through workplace support (B = 1.05, 95% CI [0.65, 1.48]). In addition, the indirect effects of role stress through workplace support are significant on DP (B = .38, 95% CI [0.21, 0.57]) and on PA (B = −.43, 95% CI [−0.83, −0.04]).

Table 7.

Estimated Coefficients for Mediation Models of Workplace Support.

	EE			DP			PA
	B	SE	95% CI	B	SE	95% CI	B	SE	95% CI
Constant	1.78	2.62	[−3.37, 6.92]	0.47	.81	[−1.12, 2.06]	39.64***	2.16	[35.40, 43.88]
Covariates
Gender	1.55**	0.53	[0.51, 2.58]	—	—	—	—	—	—
Age	−0.17***	0.04	[−0.25, −0.09]	−0.04**	.01	[−0.06, −0.02]	—	—	—
Health	1.64***	0.29	[1.08, 2.21]	—	—	—	−0.76**	0.28	[−1.31, −0.22]
Marital status	−1.55**	0.58	[−2.70, −0.41]	—	—	—	—	—	—
Length of service	—	—		—	—	—	−0.14**	0.05	[−0.25, −0.03]
Independent variable
Role stress	6.89***	0.50	[5.91, 7.86]	1.88***		[1.54, 2.22]	−4.52***	0.48	[−5.46, −3.58]
Mediator
Workplace support	−1.75***	0.31	[−2.35, −1.14]	−0.62***		[−0.83, −0.41]	0.71*	0.30	[0.13, 1.29]
Effect of role stress on burnout
Direct effect	6.89***	0.50	[5.91, 7.86]	1.88***	.17	[1.53, 2.21]	−4.52***	0.48	[−5.46, −3.58]

Note. EE = emotional exhaustion; DP = depersonalization; PA = personal accomplishment.

*p < .05. **p < .01. ***p < .001.

Moderation Analyses

The moderation effect analyses are examined using Model 1 in the SPSS macro PROCESS with significant demographic variables as covariates. Table 8 presents the direct effects of role stress, three components of burnout, and the mediation of workplace support. The direct effect from role stress on EE (B = 6.99, 95% CI [6.02, 7.97]) and the interaction effect of role stress and workplace support (B = −2.07, 95% CI [−3.25, −.85]) are both significant. The interaction effects of role stress and workplace support on DP (B = −1.10, 95% CI [−1.53, −0.68]) and PA (B = −1.73, 95% CI [−2.91, −0.56] were also significant.

Table 8.

Results of Workplace Support (WS) as Moderator Between Role Stress (RS) and Burnout.

	EE			DP			PA
	B	SE	95% CI	B	SE	95% CI	B	SE	95% CI
Constant	11.81***	1.46	[8.94, 14.68]	2.58***	.30	[1.99, 3.18]	30.81***	.64	[29.54, 32.07]
RS	6.99***	0.50	[6.02, 7.97]	1.94***	.17	[1.60, 2.28]	−4.42***	.48	[−5.35, −3.48]
WS	−1.41***	0.32	[−2.04, −0.77]	−0.44***	.11	[−0.66, −0.22]	0.10**	.31	[0.38, 1.61]
RS × WS	−2.07***	0.62	[−3.25, −0.85]	−1.10***	.22	[−1.53, −0.68]	−1.73**	.60	[−2.91, −0.56]
Gender	1.63**	0.53	[0.60, 2.66]	—	—	—	—	—	—
Age	−0.17***	0.04	[−0.25, −0.10]	−0.04***	.01	[−0.06, −0.02]	—	—	—
Health	1.63***	0.29	[1.06, 2.19]	—	—	—	−0.77**	.28	[−1.32, −0.23]
Marital status	−1.57**	0.58	[−2.72, −0.43]	—	—	—	—	—	—
Length of service	—	—	—	—	—	—	−0.14**	.05	[−0.25, −0.04]
R ²	.245***			.150***			.094***
▵R ²	.005**			.014***			.005**

Note. EE = emotional exhaustion; DP = depersonalization; PA = personal accomplishment.

**p < .01. ***p < .001.

Furthermore, three levels of workplace support (mean − 1 SD-lower support, mean-average support, and mean + 1 SD-higher support) are defined for simple slope analyses. The interaction effects are graphically depicted in three figures using the estimated coefficients from the models. Figures 1 and 2 show that among new recruited social workers with lower workplace support, the relationships of role stress with EE and DP were stronger, with greater role stress associated with greater EE and DP. In Figure 3, among new recruited social workers with lower workplace support, greater role stress is associated with lower PA.

Figure 1.

Visual representation of the moderation effect of workplace support on the relationship between role stress and emotional exhaustion.

Figure 2.

Visual representation of the moderation effect of workplace support on the relationship between role stress and depersonalization.

Figure 3.

Visual representation of the moderation effect of workplace support on the relationship between role stress and personal accomplishment.

Discussion

This study examines the relationship between role stress and burnout among newly recruited social workers, particularly the mediating and moderating roles of workplace support from supervisors and coworkers. Consistent with prior studies in various human service professions, role stress characterized by role conflict and role ambiguity is significantly related to burnout among newcomers in social work (Dewi & Riana, 2019; Richards et al., 2016). Specifically, individuals who report greater role stress through their experience of receiving incompatible or conflicting requirements (role conflict) and/or the lack of clear information to carry out their job (role ambiguity) indicate higher EE, DP, and lower PA at work. Meanwhile, role stress is negatively correlated with the perceived availability of workplace support. New social workers reporting greater workplace support also tend to report lower EE, DP, and higher PA.

On the basis of the JD–R model, we examine the moderation effect of workplace support on the relationship between role stress and the three components of burnout. As expected, the perception of workplace support did act as a moderator. Specifically, among new recruited social workers perceiving lower support from their supervisors and coworkers in job settings, the relations between role stress and burnout are stronger, with greater role stress associated with higher EE, DP, and lower PA. By contrast, higher workplace support buffers the effect of stress on the likelihood of burnout. However, the interaction effect of workplace and role stress on burnout is detected to be significant. Thus, due to small R ² change values (both were .005) in EE and PA models, workplace support as a moderator requires cautious interpretation, and further evidence is necessary to make a definitive conclusion.

In addition, we refer to the social support deterioration model and find that the effect of role stress on burnout among new social workers is mediated by workplace support. That is, individuals who feel greater stress from role conflict and ambiguity tend to perceive less support in their work environment. This result, in turn, leads to higher burnout, which is presented as higher EE and DP and lower PA.

Implications for Practice

To our knowledge, this study is the first to examine the moderating and mediating roles of workplace support in the relationship between role stress and burnout among newly recruited social workers. Although our data do not indicate that burnout, lack of support, and role stress are particularly alarming among new social workers in China, our main concern is to explore their interaction mechanisms to pave the way for early intervention.

Role stress, if unsolved, can result in burnout and even lead to one’s intent to leave the profession. Thus, the workforce quality and shortage can worsen. Measures that reduce role stress and prevent burnout are necessary for new social workers entering this demanding profession. First, from the management perspective, long-term care facilities and policies should be developed, including conducting regular assessments of levels of burnout and social support particularly when dealing with a role that includes clarifying goals, requirements, and expectations in many aspects of the job. New recruits should be encouraged to take advantage of the comprehensive support network in their organizations for the sake of their own role stress and psychological well-being related to work conditions, build social contacts at work, and seek or accept social support from the network.

With regard to practice in China, the major implication is that professional supervision training or mentoring for new recruits should be available, sustainable, and effective. The “Plan for Long-term Social Work Workforce Development 2011–2020” (Ministry of Civil Affairs, 2012) states that the government plans to train 80,000 supervisors with theoretical knowledge and practice experience. These personnel will become the foundation to provide support to new recruits and contribute to improve work experience, stability, and workforce quality for Chinese social work.

The findings also remind that different causes of role stress among new recruits should be identified timely and require appropriate support. Based on different educational and career experience background of green social workers and the unique context of social work development in China, the needs and causes of role stress may differ. For newcomers who are ambitious graduates prepared with professional values and knowledge, guidance on how to integrate theoretical learning into the specific tasks their organization may help them prioritize goal-setting and manage workload. For new recruits who are accustomed to top-down decision making, standardized work, and lack of professional training background, their supervisor and mentor should create teaching opportunities to expose them to social work knowledge, values, and techniques. In addition, creating an inclusive and free environment is important for social workers to discuss their understanding of their profession and its indigenization and the institutional and infrastructural constraints that may hinder social work development in China. Such discussion may be of particular benefit to address DP and lack of PA experienced by these new social workers. If managers or supervisors only teach passive or forceful adaptation to policies and practice the current model in institution or organizations, new practitioners may become mere subordinates to the organization instead of individuals with the competence to critically reflect on the profession and make their own contributions to shaping social work in China.

Limitations and Future Directions

This study provides a better understanding of the underlying mechanisms of role stress and burnout among newly recruited social workers and highlights the mediation and moderation effects of the perception of workplace support. However, its limitations also need discussion. First, the study is methodologically limited by cross-sectional data in assessing the related mediation and moderation effects. Fortunately, CSWLS is carried out every 3 years since 2019. Future research has the opportunity to utilize longitudinal data to identify causal relationships, do more follow-up tests, and even conduct interventions for easing role stress and burnout as well as increasing workplace support. Second, the scales used in this study and CSWLS have much room for improvement. For instance, despite its wide adoption, the Maslach Burnout Inventory has suffered increasing critiques for unclear items or questioned component when measuring burnout (Fong et al., 2014). In future study, more conceptually coherent measures can be considered. Moreover, when measuring workplace support, this study focuses on sources of support at job settings while ignoring nonwork sources, such as from friends and family members. The latter offers emotional support and active coping assistance, which are important in reducing job burnout (Thoits, 2011; Woodhead et al., 2016). The current study does not investigate the effects of specific sources and types of support (administrative, educational, and emotional), which can be considered for future research. Last but not least, the issue of reporting bias on role stress and burnout may exist in this survey, and the reality of working conditions of new recruits may require additional information. Observations and evaluations from supervisors and colleagues may also be collected in addition to the self-report of new recruits. More qualitative material such as cognitive interviews from the perspective of new workers may also help improve our understanding of the relationship between role stress, workplace support, and burnout.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by Shanghai Philosophy and Social Science Planning Program (2018BSH009).

ORCID iD

Xiao Li

References

AbuAlRub

R. F.

(2004). Job stress, job performance, and social support among hospital nurses. Journal of Nursing Scholarship, 36(1), 73–78. https://doi.org/10.1111/j.1547-5069.2004.04016.x

Azeem

S. M.

Nazir

N. A.

Zaidi

Z. B. A.

Akhtar

(2014). Role of stress and burnout among nurses in the private hospitals. International Journal of Academic Research in Business and Social Sciences, 4(3), 420–428. https://doi.org/10.6007/IJARBSS/v4-i3/720

Bakker

A. B.

Demerouti

Euwema

M. C.

(2005). Job resources buffer the impact of job demands on burnout. Journal of Occupational Health Psychology, 10(2), 170–180. https://doi.org/10.1037/1076-8998.10.2.170

Barrera

(1986). Distinctions between social support concepts, measures, and models. American Journal of Community Psychology, 14(4), 413–445. https://doi.org/10.1007/BF00922627

Coady

C. A.

Kent

V. D.

Davis

P. W.

(1990). Burnout among social workers working with patients with cystic fibrosis. Health & Social Work, 15(2), 116–124. https://doi.org/10.1093/hsw/15.2.116

Collins

Coffey

Morris

(2010). Social work students: Stress, support and well-being. British Journal of Social Work, 40(3), 963–982. https://doi.org/10.1093/bjsw/bcn148

Cortis

Fang

Dou

(2018). Social service purchasing in China: Rationale, features, and risks. Asian Social Work and Policy Review, 12(3), 200–208. https://doi.org/10.1111/aswp.12151

Deeter-Schmelz

D. R.

Ramsey

R. P.

(1997). Considering sources and types of social support: A psychometric evaluation of the House and Wells (1978) instrument. The Journal of Personal Selling and Sales Management, 17(1), 49–61. https://psycnet.apa.org/record/1999-15456-004

Demerouti

Bakker

A. B.

Nachreiner

Schaufeli

W. B.

(2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86(3), 499–512. https://doi.org/10.1037//0021-9010863499

10.

Dewi

R. S.

Riana

I. G.

(2019). The effect of workload on role stress and burnout. Journal of Multidisciplinary Academic, 3(3), 1–5.

11.

Fong

T. C.

R. T.

Au-Yeung

F. S.

Sing

Law

Lee

(2016). The relationships of change in work climate with changes in burnout and depression: A 2-year longitudinal study of Chinese mental health care workers. Psychology, Health & Medicine, 21(4), 401–412. https://doi.org/10.1080/13548506.2015.1080849

12.

Fong

T. C.

R. T.

(2014). Psychometric properties of the Copenhagen burnout inventory—Chinese version. The Journal of Psychology, 148(3), 255–266. https://doi.org/10.1080/00223980.2013.781498

13.

Freudenberger

(1975). The staff burnout syndrome in alternative situation. Journal of Psychological Research Practice, 4, 1–5.

14.

Gao

J. G.

Yan

M. C.

(2015). Social work in the making: The state and social work development in China. International Journal of Social Welfare, 24(1), 93–101. https://doi.org/10.1111/ijsw.12089

15.

Garrosa

Moreno-Jiménez

Rodríguez-Muñoz

Rodríguez-Carvajal

(2011). Role stress and personal resources in nursing: A cross-sectional study of burnout and engagement. International Journal of Nursing Studies, 48(4), 479–489. https://doi.org/10.1016/j.ijnurstu.2010.08.004

16.

Hamama

(2012). Burnout in social workers treating children as related to demographic characteristics, work environment, and social support. Social Work Research, 36(2), 113–125. https://doi.org/10.1093/swr/svs003

17.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press.

18.

Hayes

A. F.

(2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford.

19.

Haynes

C. E.

Wall

T. D.

Bolden

R. I.

Stride

Rick

J. E.

(1999). Measures of perceived work characteristics for health services research: Test of a measurement model and normative data. British Journal of Health Psychology, 4(3), 257–275. https://doi.org/10.1348/135910799168614

20.

Høigaard

Giske

Sundsli

(2012). Newly qualified teachers’ work engagement and teacher efficacy influences on job satisfaction, burnout, and the intention to quit. European Journal of Teacher Education, 35(3), 347–357. https://doi.org/10.1080/02619768.2011.633993

21.

Hombrados-Mendieta

Cosano-Rivas

(2013). Burnout, workplace support, job satisfaction and life satisfaction among social workers in Spain: A structural equation model. International Social Work, 56(2), 228–246. https://doi.org/10.1177/0020872811421620

22.

House

J. S.

Wells

J. A.

(1978). Occupational stress, social support and health. In Mclean

Black

Colligan

(Eds.), Reducing occupational stress: Proceedings of a conference (pp. 8–29). U.S. Government Printing Office.

23.

Hutchings

Taylor

(2007). Defining the profession? Exploring an international definition of social work in the China context. International Journal of Social Welfare, 16(4), 382–390.

24.

Jenkins

Elliott

(2004). Stressors, burnout and social support: Nurses in acute mental health settings. Journal of Advanced Nursing, 48(6), 622–631. https://doi.org/10.1111/j.1365-2648.2004.03240.x

25.

Jiang

Wang

Chui

(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. https://doi.org/10.1177/0020872817712564

26.

Kadushin

Harkness

(2014). Supervision in social work. Columbia University Press.

27.

Kim

Kao

(2011). Burnout and physical health among social workers: A three-year longitudinal study. Social Work, 56(3), 258–268. https://doi.org/10.1093/sw/56.3.258

28.

Kim

Stoner

(2008). Burnout and turnover intention among social workers: Effects of role stress, job autonomy and social support. Administration in Social Work, 32(3), 5–25. https://doi.org/10.1080/03643100801922357

29.

Law

A. K. C.

J. X.

(2008). Social work education in mainland China: Development and issues. Asian Social Work and Policy Review, 2(1), 1–12. https://doi.org/10.1111/j.1753-1411.2008.00006.x

30.

Leiter

M. P.

Maslach

(1988). The impact of interpersonal environment on burnout and organizational commitment. Journal of Organizational Behavior, 9(4), 297–308. https://doi.org/10.1002/job.4030090402

31.

Leung

T. T.

(2012). The work sites as ground of contest: Professionalisation of social work in China. British Journal of Social Work, 42(2), 335–352. https://doi.org/10.1093/bjsw/bcr063

32.

(2016). Shehui gonzuozhe de zhiye liudong [Career mobility of social workers: Research status and expansion]. Social Work and Management, 2, 70–77.

33.

Zeng

(2019). Degree or examination: What is the foundation of the social work workforce in China? International Social Work, 62(1), 447–460. https://doi.org/10.1177/0020872817731145

34.

Lin

(1986). Conceptualizing social support. In N. Lin

A. Dean

Ensel

W. M.

(Eds.), Social support, life events, and depression (pp. 17–30). Elsevier.

35.

Lloyd

King

Chenoweth

(2002). Social work, stress and burnout: A review. Journal of Mental Health, 11(3), 255–265. https://doi.org/10.1080/09638230020023642

36.

Maslach

Jackson

S. E.

Leiter

(1996). The Maslach Burnout Inventory: Manual. Consulting Psychologists Press.

37.

Maslach

Jackson

S. E.

Leiter

M. P.

Schaufeli

W. B.

Schwab

R. L.

(1986). Maslach Burnout Inventory. Consulting psychologists press.

38.

Ministry of Civil Affairs. (2012). The plan for long-term social work workforce development (2011–2020). http://www.mca.gov.cn/article/zwgk/jhgh/201204/20120400302325.shtml

39.

Mutkins

Brown

Thorsteinsson

E. B.

(2011). Stress, depression, workplace and social supports and burnout in intellectual disability support staff. Journal of Intellectual Disability Research, 55(5), 500–510. https://doi.org/10.1111/j.1365-2788.2011.01406.x

40.

Pang

Chin

Choi

(1997). Burnout in front line nurses in caring of the terminally ill. Hospital Authority, Hong Kong.

41.

Park

K.-O.

Wilson

M. G.

Lee

M. S.

(2004). Effects of social support at work on depression and organizational productivity. American Journal of Health Behavior, 28(5), 444–455. https://doi.org/10.5993/AJHB.28.5.7

42.

Prosser

Johnson

Kuipers

Dunn

Szmukler

Reid

Bebbington

Thornicroft

(1999). Mental health, “burnout” and job satisfaction in a longitudinal study of mental health staff. Social Psychiatry and Psychiatric Epidemiology, 34(6), 295–300.

43.

Shen

(2012). Shehuigongzuo jigou dudao peiyang dujing xuanze [Supervisor training path in social work institutions]. Social Work, 8, 31–33.

44.

Richards

K. A. R.

Levesque-Bristol

Templin

T. J.

Graber

K. C.

(2016). The impact of resilience on role stressors and burnout in elementary and secondary teachers. Social Psychology of Education, 19(3), 511–536. https://doi.org/10.5993/AJHB.28.5.7

45.

Rizzo

J. R.

House

R. J.

Lirtzman

S. I.

(1970). Role conflict and ambiguity in complex organizations. Administrative Science Quarterly, 15, 150–163.

46.

Rudman

Gustavsson

J. P.

(2011). Early-career burnout among new graduate nurses: A prospective observational study of intra-individual change trajectories. International Journal of Nursing Studies, 48(3), 292–306. https://doi.org/10.1016/j.ijnurstu.2010.07.012

47.

Schaufeli

W. B.

Taris

T. W.

(2014). A critical review of the job demands-resources model: Implications for improving work and health. In Bauer

G. F.

Hämming

(Eds.), Bridging occupational, organizational and public health (pp. 43–68). Springer.

48.

Schwartz

R. H.

Tiamiyu

M. F.

Dwyer

D. J.

(2007). Social worker hope and perceived burnout: The effects of age, years in practice, and setting. Administration in Social Work, 31(4), 103–119. https://doi.org/10.1300/J147v31n04_08

49.

Shen

(2008). Shehuigongzuozhe de zhiyejuedai: Guoji yanjiu yu jingyan qishi [Burnout of social workers: International researches and experiences]. Shanghai Youth Management Officers Journal, 2, 31–33.

50.

Thoits

P. A.

(2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior, 52(2), 145–161. https://doi.org/10.1177/0022146510395592

51.

Tong

Xin

Luo

. (2018). Bentu shegong jigou dudao de chengci he gongneng [The levels and functions of local social work supervisors: An exploring framework]. Social Work and Management, 18(1), 5–10.

52.

Travis

D. J.

Lizano

E. L.

Mor Barak

M. E

. (2016). “I’m so stressed!”: A longitudinal model of stress, burnout and engagement among social workers in child welfare settings. The British Journal of Social Work, 46(4), 1076–1095. https://doi.org/10.1093/bjsw/bct205

53.

Woodhead

E. L.

Northrop

Edelstein

(2016). Stress, social support, and burnout among long-term care nursing staff. Journal of Applied Gerontology, 35(1), 84–105. https://doi.org/10.1177/0733464814542465

54.

(2019). Shehuigongzuo zhuanye jiangou [Construction of social work profession: Analysis from institutional embedness]. Lanzhou University Journal, 47(2), 24–33.

55.

Yan

M. C.

Cheung

K. W.

(2006). The politics of indigenization: A case study of development of social work in China. Journal of Sociology and Social Welfare, 33, 63–83.

56.

Yan

M. C.

Tsang

(2005). A snapshot on the development of social work education in China: A Delphi study. Social Work Education, 24(8), 883–901. https://doi.org/10.1080/02615470500342314

57.

Yuan

Liu

Sun

. (2020). Baseline survey of China Social Work Longitudinal Study 2019: Design and implementation. Research on Social Work Practice. https://doi.org/10.1177/1049731520984536