Wang et al.

BMC Psychiatry (2017) 17:364

DOI 10.1186/s12888-017-1533-6

RESEARCH ARTICLE Open Access

Associations between occupational

stress, burnout and well-being among
manufacturing workers: mediating roles
of psychological capital and self-esteem
Ziyue Wang1, Hongbo Liu2, Haijian Yu1, Yanwen Wu1, Shuai Chang1 and Lie Wang1*

Abstract
Background: Occupational stress is an important risk factor for mental health among occupational population.
Exploring related mediators of workers’ mental health are important to improve their health and performance.
Our study aims to explore the relationships between work stress, positive psychological resources, burnout and well-being.
Methods: The study was performed during the period of June and July in 2015. A questionnaire that consisted of the
Effort-reward Imbalance Scale, the Maslach Burnout Inventory-General Survey, the Psychological Capital Questionnaire, the
Rosenberg Self-esteem Scale, the Flourishing Scale, as well as demographic and working factors.
Results: Psychological capital mediated the associations of effort-reward imbalance and emotional exhaustion, cynicism,
professional efficacy and well-being. Self-esteem mediated the associations of effort-reward imbalance ratio with cynicism,
professional efficacy and well-being, and mediated the associations of overcommitment and cynicism, professional efficacy
and well-being.
Conclusions: The findings indicated that enhancing positive psychological resources could be considered in developing
intervention strategies for psychological health among manufacturing employees.
Keywords: Occupational stress, Burnout, Well-being, Positive psychology

Background Meanwhile, prior studies of manufacturing workers have

A growing number of companies start to pay attention mostly concentrated on the health hazards of occupa-
to not only the physical health of their employees, but tional exposure [6] and neglected the research of their
also their psychological health. Occupational stress exists psychological health. To fill this gap, this study aimed to
in all occupations. However, excessive occupational explore the level of occupational stress in manufacturing
stress has been considered “toxic” parts of job environ- workers and its potential mechanism whereby their psy-
ment, and is often associated with psychological and chological health is affected.
physical health. Empirical studies demonstrated that The well-known model of effort-reward imbalance
occupational stress was a risk factor for burnout, poor (ERI), proposed by Siegrist, was adopted to understand
wellbeing, cardiovascular disorders [1–3]. Research on the potential mechanism of stress [7, 8]. This model
the relationships between occupational stress and mental consisted of three core parts: efforts, rewards and over-
health has predominantly focused on medical staff, commitment, and emphasized the lack of reciprocal
teachers, and information technology employees [4, 5]. social exchange between costs (efforts, such as time
pressure to fulfill a task) and gains (rewards, such as
money, respect/support received from colleagues, job se-
* Correspondence: liewang@[Link]
1
Department of Social Medicine, School of Public Health, China Medical
curity, and career opportunities) at work. The imbalance
University, No.77 Puhe Road, Shenyang North NewArea, Shenyang, Liaoning, between effort and reward often lead to emotional dis-
People’s Republic of China tress. The model also introduced a specific pattern of
Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License ([Link] which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
([Link] applies to the data made available in this article, unless otherwise stated.
Wang et al. BMC Psychiatry (2017) 17:364 Page 2 of 10

coping with job demands, termed as overcommitment, self-acceptance, competence, optimism, meaning and
referring to “a set of attitudes, behaviors and emotions purpose, supportive and rewarding relationships, per-
reflecting excessive striving in association with a strong ception of the personal contribution to the well-being
desire of being approved and esteem”. The overcom- of others, and being respect. The current study used
mitted individuals may make great efforts, even under burnout and psychological well-being to explore the
low-gain circumstances, and thus were more likely to associations between occupational stress and psycho-
experience burnout [9]. logical health in terms of both negative and positive
Burnout is regarded a syndrome that can have signifi- outcomes. Thus, we hypothesize:
cant effects on physical and psychosocial well-being
[10–12]. The popular theory of job burnout was devel- H1 There is a positive association between job stress
oped by Maslach, Schaufeli and Leiter [13], who defined and job burnout among workers.
burnout as “a psychological syndrome in response to H2 There is a negative association between job stress
chronic interpersonal stressors on the job” or a set of and well-being among workers.
negative consequences of prolonged work stress [12].
The three key dimensions of this syndrome are an over- Although the associations between ERI and psycho-
whelming exhaustion, a sense of cynicism and detach- logical outcomes have been examined in certain occupa-
ment from the work, and feelings of ineffectiveness or tions, the mechanisms behind the associations still
lack of accomplishment [13]. The general survey of remain largely unknown. Over the years, more and more
Marlsh Burnout Inventory (MBI-GS), which is based on researchers began to notice that stress reaction differs a
the three-dimensional of burnout, has been largely lot among different individuals, no matter in mental or
shapes and dominates burnout research [14]. On the physical health. Stress and coping theory considered that
level of organization, burnout is associated with turnover personality characteristics could influence individual’s
intention, lower productivity, and a decreased commit- appraisal and coping processes, and even psychological
ment [11, 15]. Wang et al. have been demonstrated that symptoms in stressful situations [23]. Based on the
teachers who experienced high level of occupational transaction theoretical framework, studies on the
stress would have more symptoms of burnout [16]. Job process of dealing with stress should consist of not only
burnout is a psychological syndrome arising in response independent and outcome variables, but also mediating
to chronic interpersonal stressors and job tediousness variables that determine the individual more or less suc-
[11]. Similarly, depression is one of the psychological cessful adaption to the stress [24]. With an increasing
stresses that may overlap with burnout. However, the recognition of the value of positive psychology, organiza-
distinction between burnout and depression is that tions may improve their employees’ physical and mental
burnout is supposedly work related, whereas depression health by enhancing their inner positive psychological
is considered to be more multifactorial in origin and resources. This is consistence with the transactional the-
pervasive in nature [17]. Futhermore, Shin et al. [12] ory, by which stress is the condition that results when
proposed that burnout was one of the strongest predic- person-environment transactions lead the individual to
tors of depression by a longitudinal study. Therefore, perceive a discrepancy between the demands of situation
this study used burnout to understand the negative men- and the resources of the person’s psychological,
tal status of occupational population. In addition, the biological or social systems [25]. Psychological capital
World Health Organization [18] argued that the focal (PsyCap) is a high-order construct extracted from posi-
point of mental well-being should not be limited to the tive organizational behavior, and has been demonstrated
absence of mental illness, but the positive character of as a positive resource for improving employees’ perform-
mental system and the attitudes that are inherent to it. ance, job satisfaction, well-being, and for decreasing
Employees with high efforts and low rewards had higher stress, turnover, and burnout [26–29]. PsyCap refers to a
risks of poor well-being [19]. Eudaimonic well-being, positive state of mind exhibited during the growth and
one of the main theoretical frameworks that has been development of an individual and consists of four state--
proposed, refers to “the dimensions that describe the like psychological resources: hope, self-efficacy, opti-
optimal psychological functioning of the individuals” mism and resilience, all of which are measurable,
[20]. Diener et al. [21, 22] developed the Flourishing developable and manageable [27]. In addition to some
Scale (FS) to provide a comprehensive and brief measure studies focused on the antecedents and outcomes of Psy-
of flourishing that summarized the current dimensions Cap, there have been other studies pay attentions on the
proposed by others authors, and an overview of the indi- underlying mechanisms through which PsyCap impacts
vidual’s perception of person’s own positive functioning. work stress at different level of analysis. Park et al. [30]
The items of the FS describe broad and important found the mediating role played by PsyCap in the associ-
facets of human functioning: engagement and interest, ation between empowering leadership and psychological
Wang et al. BMC Psychiatry (2017) 17:364 Page 3 of 10

well-being. Wang et al. [31] looked at PsyCap as a medi- consents. Each worker was given a questionnaire to read
ator in linking work-family conflict to job burnout. declaration of private information secrecy and complete
Drawing on prior researches, this study suggests that voluntarily. Investigators were responsible for distribut-
PsyCap may mediate the associations of occupational ing, retrieving and managing the questionnaires, and
stress with burnout and psychological well-being. they provided explanation without inducement for any
Besides PsyCap, self-esteem is also recognized as an im- unclear questionnaire items. The average questionnaire
portant internal resource. It is the evaluation which the spent 10 to 15 min. 79 workers refused to participate in
individual makes and customarily maintains with regard the study because of reading difficulty. The question-
to him or herself [32]. Rosse et al. [33] considered that naires with missing data that accounted for more than
low self-esteem individuals may be less effective in inter- 15% of the whole scale were regarded invalid. Finally,
personal relationships, and experience feeling of incom- effective responses were obtained from 1219 individuals
petence in their relationships with colleagues. Low (effective response rate: 81.3%), with 923 (75.7%) males
self-esteem individuals also may have fewer psychic and 296 (24.3%) females. The study was approved by
resources to cope with daily stressors, making them vul- the Committee on Human Experimentation of China
nerable to emotional exhaustion. Previous studies have Medical University (Additional file 1).
found that self-esteem is an important factor in alleviat-
ing symptoms of burnout [34]. Self-esteem is an import-
ant inner resource in dealing with stress, and is also one Demographic and working characteristics
of several mediating variables that influence the effects Demographic information regarding age, gender, marital
of stressors on health consequences. Based on the above, status, and education level were obtained. Three working
we formulate the following hypotheses: characteristics were also measured: Monthly income,
years of experience, and weekly working hours.
H3: PsyCap mediates association between job stress
and burnout.
H4: PsyCap mediates association between job stress Measurement of occupational stress
and well-being. Occupational stress was measured by the Chinese
H5: Self-esteem mediates association between job stress version of the effort-reward imbalance (ERI) model [35].
and burnout. The ERI questionnaire consists of three dimensions:
H6: Self-esteem mediates association between job stress extrinsic effort (6 items), reward (11 items), and over-
and well-being. commitment (6 items). The items of extrinsic effort and
reward are scored on a 5-point scale, and higher scores
As described above, the present study aims to investi- indicated higher demands of efforts and higher rewards.
gate the associations between occupational stress, For the overcommitment subscale, responses are scored
PsyCap, self-esteem, burnout and well-being, as well as from 1 to 4, with higher scores indicating higher de-
the potential mediating roles of positive psychological mands characterized by excessive work-related commit-
[Link] from earlier studies about work-related ment. Effort-reward ratio (ERR) was calculated using a
stress that forcus on the single intervening variable, the predefined algorithm that quantifies the degree of mis-
present study adopt multi- intermediary test to build a match between high cost and low gain, with a correction
systematic theoretical framework of stress and stress re- factor of 0.5454 [8]. The Cronbach’s coefficients for ex-
action from the perspective of positive psychology. It is trinsic effort, reward, and overcommitment scales were
helpful for management decision-making on manufac- 0.911, 0.936, and 0.771 in the present study.
turing workers, enriching the existing theoretical model
of positive psychology, and acting as preventive strat-
egies during working lifetime (Fig. 1). Measurement of psychological capital
PsyCap was evaluated by the Chinese version of the 24-
Methods item Psychological Capital Questionnaire (PCQ) [27].
Study design and sample The PCQ is comprised of four dimensions: self-efficacy
A cross-sectional study was conducted in Northern re- (6 items), hope (6 items), resilience (6 items), and opti-
gion of China in June and July, 2015. In total, 26 factor- mism (6 items). Each question is scored from 1 (strongly
ies (1500 workers) were randomly selected by cluster disagree) to 6 (strongly agree). Higher scores indicate
sampling in an industry which made component parts of higher levels of psychological capital. The PCQ has dem-
airplanes. Trainees were excluded in this study. Self- onstrated adequate reliability and validity in multiple
administered questionnaires were directly distributed to samples [27, 36]. For the total scale in our study, the
these subjects after obtaining their written informed Cronbach’s alpha coefficient was 0.953.
Wang et al. BMC Psychiatry (2017) 17:364 Page 4 of 10

PsyCap
Self-esteem

Burnout
Job stress
Well-being

Demographic Factors
(Control Variables)

Fig. 1 Conceptual framework of this study. PsyCap:psychological capital

Measurement of self-esteem correlations among major variables. Hierarchical regres-

Self-esteem was measured with Rosenberg self-esteem sion analysis was adopted to explore the effects of inde-
scale (RSES), developed by the sociologist Dr. Morris pendent variables on burnout and well-being. In step1, all
Rosenberg [32]. This scale is a ten-item questionnaire. demographic and working variables which were signifi-
Each item is scored on a 4-point Likert scale, which cantly related to dependent variables were entered.
ranged from “strongly agree” to “strongly disagree”. Five Dummy variables were set for the categorical variables; in
of the items are positively stated and five are negatively step 2, the ERR and overcommitment were entered; in
stated. The Chinese version was translated by Cheng et step 3, positive resources variables were added. Standard-
al. [37] and has been widely used in different popula- ized estimate (β), F, R2 and R2-changes (ΔR2) for each step
tions. In the current study, The Cronbach’s alpha coeffi- were displayed. Four different models were estimated in
cient was 0.879. this study, one for each dimension of burnout and one for
well-being. Asymptotic and resampling strategies, devel-
Measurement of burnout oped by Preacher and Hayes [43], were used to examine
This study used Maslach Burnout Inventory-General the mediating roles of positive psychological resources
Survey (MBI-GS) to measure burnout [38, 39]. The scale (a*b product) on the association between occupational
consists of three dimensions: emotional exhaustion (EE), stress and psychological outcomes (burnout and well-
cynicism (CY), and professional efficacy (PE). Each of being). In these regression equations, ERI and over-
the items is scored on a Likert scale from 0 to 6. Higher commitment were modeled as predictors, burnout and
scores on the EE and CY, and lower score of PE indicate well-being as dependent variables, PsyCap and self-esteem
higher level of burnout. The Chinese version of the as mediators, and demographic and working variables as
MBI-GS was revised and validated by Li et al. and has covariates. The bootstrap estimate was based on 5000
demonstrated good reliability and validity in the Chinese bootstrap samples. The bias-corrected and accelerated
population [40]. In this study, the Cronbach’s alpha coef- 95% confidence interval (BCa 95% CI) for each a*b prod-
ficients for EE, CY, and PE were 0.940, 0.940, and 0.928. uct was calculated, and a BCa 95% CI excluding 0
indicated a significant mediating role. All analyses were
Measurement of well-being performed using SPSS 17.0 for Windows, and the signifi-
Psychological well-being was measured with the brief cance level was set at p < 0.05(two-tailed). Missing values
8-item Flourishing Scale (FS). The range of scores is were addressed with mean substitution.
from 8 to 56, where higher scores mean a higher level of
psychological well-being [41]. The FS scale has demon- Results
strated good validation in different cultures [42]. The Descriptive statistics
Cronbach’s alpha coefficient was 0.950 in this study. Demographic and working characteristics of the workers
and univariate analysis of burnout and well-being scores
Statistical analyses were shown in Table 1. The mean age of the subjects
Frequency statistics and one-way ANOVA or t-test was was 35.44 years (SD = 9.47, ranged from 20 to 59).
used to describe distributions of burnout and well-being Univiariate analyses indicated that age, monthly income,
in categorical demographic and working characteristics weekly working hours, and working years were associ-
with mean, standard deviation (SD), number (N) and ated with emotional exhaustion and cynicism, respect-
percentage (%). Pearson’s correlation was used to examine ively (P < 0.05). Education level was related to cynicism
Wang et al. BMC Psychiatry (2017) 17:364 Page 5 of 10

Table 1 Distribution and univariate analyses of MBI-GS and well-being

N(%) EE CY PE WB
Mean ± SD Mean ± SD Mean ± SD Mean ± SD
Age
20–29 437(35.8) 10.46 ± 7.09 7.08 ± 5.99 26.73 ± 11.18 44.90 ± 8.51
30–39 420(34.5) 10.21 ± 7.76 6.84 ± 6.40 28.56 ± 11.27 44.42 ± 8.76
> 40 362(29.7) 8.41 ± 6.89 5.41 ± 5.34 26.51 ± 12.55 44.61 ± 8.43
F 9.042** 8.841** 3.827* 0.338
Gender
male 923(75.7) 9.82 ± 7.21 6.68 ± 5.94 27.11 ± 11.52 44.36 ± 8.82
female 296(24.3) 9.58 ± 7.65 5.93 ± 6.10 27.87 ± 12.09 45.53 ± 7.68
T 0.486 1.894 −0.969 −2.181*
Marriage
Married/cohabitation 886(72.7) 9.57 ± 7.35 6.33 ± 5.97 27.39 ± 11.83 44.57 ± 8.61
Single/dicorced/widow 333(27.3) 10.28 ± 7.23 6.96 ± 6.02 27.04 ± 11.21 44.85 ± 8.48
T 1.510 1.644 −0.467 0.517
Educational level
Technical secondary school 232(19.0) 9.37 ± 7.94 6.36 ± 6.22 24.96 ± 12.65 44.06 ± 9.29
and the following
Junior college 492(40.4) 10.13 ± 7.62 7.08 ± 6.44 26.91 ± 12.00 44.10 ± 8.63
Bachelor degree or above 495(40.6) 9.58 ± 6.69 5.99 ± 5.34 28.78 ± 10.59 45.47 ± 8.09
F 1.109 4.134* 9.033** 3.808*
Income(RMB, yuan)
< 4000 919(75.4) 10.14 ± 7.44 6.82 ± 6.13 26.85 ± 11.68 44.12 ± 8.90
≧4000 300(24.6) 8.60 ± 6.81 5.53 ± 5.42 28.67 ± 11.51 46.26 ± 7.24
T 3.340** 3.450** −2.356* −4.179**
Weekly working hours
< 40 840(68.9) 8.84 ± 6.86 6.07 ± 5.67 27.30 ± 11.83 45.08 ± 8.47
≧40 379(31.1) 11.81 ± 7.89 7.45 ± 6.55 27.28 ± 11.29 43.70 ± 8.73
T −6.346** −3.557** 0.031 2.609*
Working years
< 12 674(47.1) 10.50 ± 7.21 6.97 ± 5.95 27.26 ± 11.32 44.94 ± 8.62
≧12 645(52.9) 9.11 ± 7.36 6.09 ± 6.00 27.33 ± 11.96 44.38 ± 8.52
T 3.323** 2.570** −0.113 1.137
EE emotional exhaustion, CY cynicism, PE professional efficacy, WB well-being
*p < 0.05, **p < 0.01 (two-tailed)

(P < 0.05). Age, educational level, and monthly income Psycap and self-esteem were negatively related to emotional
were related to professional efficacy (P < 0.05). Gender, exhaustion, cynicism and positively related to professional
education, income, and weekly working hours were efficacy and well-being.
related to well-being (P < 0.05).
Associations among ERI, PsyCap, self-esteem, and burnout
Correlations among continuous variables All independent variables were correlated with the three
The correlations among ERI, PsyCap, self-esteem, compo- components of burnout in univariate analyses. Thus, they
nents of burnout, and well-being were shown in Table 2. were entered in the multiple hierarchical linear regression
As revealed in the table, ERR, overcommitment were posi- models. Results were presented in Table 3. In step3, after
tively correlated with emotional exhaustion and cynicism, adjusting for age, gender, education, income, working
respectively, and had negative correlations with professional years, and weekly working hours, ERR, overcommitment,
efficacy, well-being and positive psychological variables. and PsyCap were associated with emotional exhaustion.
Wang et al. BMC Psychiatry (2017) 17:364 Page 6 of 10

Table 2 Correlations among dimension of MBI-GS, ERI and positive psychological variables
variables Mean SD 1 2 3 4 5 6 7 8
1. Emotional exhaustion 9.76 7.32 1
2. Cynicism 6.50 5.99 .751** 1
3. Professional efficacy 27.29 11.65 −.140** −.271** 1
4. Well-being −.322** −.399** .306** 1
5. ERR 0.76 0.58 .612** .614** −.193** −.251** 1
6. Overcommitment 13.42 3.45 .502** .371** −.070* −.257** .481** 1
7. PsyCap 4.31 0.80 −.390** −.486** .405** .616** −.350** −.279** 1
8. Self-esteem 21.39 4.65 −.262** −.357** .419** .522** −.248** −.285** .630** 1
ERR effort-reward ratio, PsyCap psychological capital
*p < 0.05, **p < 0.01 (two-tailed)

All variables could explain 47.7% of the variance in emo- Mediating effects of PsyCap and self-esteem
tional exhaustion, and positive psychological variables The indirect pahways of ERI with psychological out-
were responsible for an additional 2.6% in Step3. ERR and comes through PsyCap and self-esteem are shown in
PsyCap and self-esteem were associated with cynicism and Table 4. The path coefficients of ERI with positive psy-
were responsible for 47.3% of the variance in cynicism. chological variables (a), the relations of positive psycho-
ERR, overcommitment, PsyCap, self-esteem were associ- logical variables with dependents (b), the direct relations
ated with professional efficacy, and could explain 22.8% of of ERI with dependents(c), the relations of ERI with de-
its variance. pendents after controlling for the effects of the positive
psychological variables (c’), a*b (the indirect effects of
ERI on the dependent variables through the positive psy-
Associations between ERI, PsyCap, self-esteem, and well-being chological variables) products, and BCa 95% CI for these
After adjusting for controlling variables which were re- products are shown in Table 4. BCa 95% CI excluding 0
lated to well-being in univariate analyses, as shown in indicated a significant mediating role. If the indirect path
Table 3, ERR and overcommitment were associated with coefficient (c’) was significant (P < 0.05), the positive psy-
well-being in step 2. In step 3, PsyCap and self-esteem chological variable would play a part of the intermediary
were positively and significantly associated with well- role between independent and dependent variables. If
being, respectively. In addition, the effect of ERR and not (P > 0.05), it would have wholly mediator effect be-
overcommitment on well-being reduced in step3. tween independent and dependent variables.

Table 3 Hierarchical linear regression analyses of the factors associated with MBI-GS scores
variables EE CY PE WE
Step1 Step2 Step3 Step1 Step2 Step3 Step1 Step2 Step3 Step1 Step2 Step3
Age 0.041 0.015 0.008 0.002 −0.014 −0.036 0.038 0.04 0.088 −0.115 −0.099 −0.047
Gender 0.003 0.018 0.016 −0.042 −0.03 −0.026 0.01 0.007 −0.011 0.048 0.04 0.026
Education_1 −0.012 −0.022 −0.023 0.013 −0.001 0.002 0.082* 0.088* 0.075* 0.003 0.005 −0.005
Education_2 −0.081 −0.047 −0.034 −0.094* −0.056 −0.023 0.152** 0.138** 0.075* 0.075 0.062 −0.013
Income −0.063 −0.061** −0.041 −0.066* −0.061** −0.024 0.06* 0.058* 0.003 0.101** 0.100** 0.024
working hours 0.264** 0.089** 0.096** 0.159** 0.001 0.012 −0.015 0.031 0.017 −0.081** 0.004 −0.017
Working years −0.189** −0.105 −0.106* −0.131 −0.052 −0.048 −0.043 −0.066 −0.082 0.114 0.075 0.063
ERR 0.456** 0.407** 0.553** 0.47** −0.211** −0.109** −0.159** 0.001
Overcommitment 0.258** 0.235** 0.103** 0.048 0.02 0.117** −0.18** −0.062*
PsyCap −0.193** −0.268** 0.219** 0.466**
Self-esteem 0.033 −0.055* 0.286** 0.208**
F 19.012** 110.79** 99.953** 10.297** 88.214** 98.594** 3.305** 8.113** 32.336** 4.930** 15.806** 78.308**
R2 0.099 0.451 0.477 0.056 0.396 0.473 0.019 0.057 0.228 0.028 0.105 0.416
R2-change 0.099 0.352 0.026 0.056 0.340 0.077 0.019 0.038 0.171 0.028 0.078 0.311
Education_1 means “Junior college” vs. “Technical secondary school and the following”, Education_2 means “Bachelor degree or above” vs. “Technical secondary
school and the following”; *P < 0.05; **P < 0.01(two-tailed)
Wang et al. BMC Psychiatry (2017) 17:364 Page 7 of 10

Table 4 Intermediary function test

X M Y c a b c’ a*b(BCa95%)
Model1
ERR PsyCap EE 5.8505** −0.3879** −1.7822** 5.2241** 0.6914(0.4137,1.0648)
Self-esteem −1.1191** 0.0580 −0.0649(−0.2154,0.0337)
Overcommitment PsyCap EE 0.5641** −0.0372** −1.7822** 0.5148** 0.0662(0.0353, 0.1087)
Self-esteem −0.2912** 0.0580 −0.0169(−0.0477,0.0106)
Model2
ERR PsyCap CY 5.7204** −0.3879** −1.9967** 4.8659** 0.7746(0.4721,1.1169)
Self-esteem −1.1191** −0.0714* 0.0799(0.0023,0.1980)
Overcommitment PsyCap CY 0.1839** −0.0372** −1.9967** 0.0889* 0.0742(0.0406, 0.1166)
Self-esteem −0.2912** −0.0714* 0.0208(0.0003,0.0459)
Model3
ERR PsyCap PE −4.2256** −0.3879** 3.1443** −2.1995** −1.2198(−1.9028,-0.6473)
Self-esteem −1.1191** 0.7205** −0.8063(−1.3403,-0.3646)
Overcommitment PsyCap PE 0.0661 −0.0372** 3.1443** 0.3927** −0.1168(−0.1913,-0.0620)
Self-esteem −0.2912* 0.7205** −0.2098(−0.3089,-0.1346)
Model4
ERR PsyCap Well-being −2.3444** −0.3879** 4.9939** 0.0171 −1.9374(−2.6831,-1.2118)
Self-esteem −1.1191** 0.3789** −0.4241(−0.7464,-0.1908)
Overcommitment PsyCap Well-being −0.4467** −0.0372** 4.9934** −0.1508* −0.1855(−0.2783,-0.1058)
Self-esteem −0.2912** 0.3789** −0.1103(−0.1675,-0.0622)
*P < 0.05, **P < 0.01
a: associations of ERI with PsyCap and self-esteem
b: associations of PsyCap/ self-esteem with burnout/well-being
c: direct associations of ERI with burnout/well-being
c’: indirect associations of ERI with burnout/well-being
age, gender, income, working hours, working years and education level are covariates

Given that, for Model 1 (Table 4), PsyCap were medi- Discussion
ated the association of ERR with emotional exhaustion This study enriched the theoretical framework of posi-
(a*b = 0.6914, BCa 95% CI: 0.4137, 1.0648), and mediated tive psychology in terms of the associations between oc-
the association of overcommitment with emotional cupational stress, positive psychological resources and
exhaustion (a*b = 0.0662, BCa95%CI: 0.0353, 0.1087). mental health outcomes, as well as the potential mediat-
PsyCap played a partial mediating effect in the rela- ing roles of positive psychological resources in the
tionship between ERR/overcommitment and emo- relationships. According to the results of multiple hier-
tional exhaustion (c’ = 5.2241, P < 0.01). Similarly, for archical regressions, lower levels of education and
Model 2, PsyCap and self-esteem partially mediated income, longer weekly working hours, and shorter work-
the relationship between ERR and cynicism, and ing years were associated with burnout. Older workers
partially mediated the relationships between over- were less likely to have symptoms of emotional exhaus-
commiment and cynicism. For Model 3, PsyCap and tion, which was in line with the findings of previous
self-esteem were partially mediated the relationship studies [4]. They had more experience than younger
between ERR and professional efficacy, and partially workers and were more likely to better deal with differ-
mediated the relationship between overcommitment ent situations. Workers with less working years often
and professional efficacy. begin with great expectations and may go through a dif-
In the Model 4, there were significant and full mediat- ficult adjustment period, due to the lack of experience
ing roles of PsyCap and self-esteem between ERR and and practical skills. Above all, the three variables of ERI,
well-being (both of the mediators’ Bca95%CI excluded 0, PsyCap and self-esteem explained a large proportion of
c’ = 0.0171, P > 0.05). PsyCap and self-esteem partially the variance in burnout. For the well-being model, all
mediated the associations between overcommitment and the demographics indicators had no significant associ-
well-being (both of the mediators’ Bca95%CI excluded 0, ation with well-being when they were controlled. The
c’ = −0.1508, P < 0.05). results suggested that occupational stress and positive
Wang et al. BMC Psychiatry (2017) 17:364 Page 8 of 10

psychological resources could have relatively major the understanding that PsyCap and self-esteem were posi-
impacts on employees’ well-being. tive resources for combating burnout and encouraged us
The percentage of our sample workers who had occu- to explore the mediating roles of positive psychological re-
pational stress was 22.4%, which was similar to that re- sources in the relationship between occupational stress
ported by university teachers [44], and lower than some and burnout in this population.
other groups of employees [45]. With regard to the rela- Results revealed that PsyCap played a mediating role
tionship between occupational stress and mental health, in the relationships between ERI and emotional exhaus-
workers who experienced high ERR and overcommit- tion and cynicism, and also partially mediated the effects
ment had higher level of emotional exhaustion and cyni- of ERR on professional efficacy. Self-esteem partially me-
cism, and lower professional efficacy and well-being diated the effects of ERI on cynicism and partially medi-
among the manufacturing workers. These results were ated the effects of ERR on professional efficacy. Both
consistent with the findings of previous studies [3, 46]. PsyCap and self-esteem mediated the relations between
Workers with high effort, such as time and energy, ERI and well-being. Workers who perceived more effort-
might tend to experience more negative emotions. When reward imbalance and overcommitment would be more
the level of extrinsic effort exceeded intrinsic reward, the likely to experience lower levels of positive psychological
impact on both physical and mental health could easily resources which in turn increased the possibilities of de-
lead to burnout and decrease the level of psychological veloping burnout symptoms and decreasing their well-
well-being. Based on the Allostatic Load model, Girardi being. Compared with decreasing workers’ occupational
et al. [47] demonstrated that work-related stress could stress, it is a more positive and strategic way for com-
associate with biomarkers of inflammation, through panies to develop programs to increase the inner posi-
negative affectivity which affected the exposure to psy- tive resources of their workers, thus to enhance the
chological stressors. In addition, most of the manufac- mental health and improve the organizational perform-
turing workers are often exposed to adverse working ance in the long run.
environment such as noise, high temperature, dust, The findings have theoretical and practical implications
organic solvents, and prolonged working hours on the for occupational health and performance management. In
assembly line with monotonous and repetitive tasks. theory, different from previous studies which focused on
These adverse factors may impact the mental health, the effects of single intervening organizational behavior
and contribute to their burnout [48]. There is no doubt variable, our study explored the concurrent effects of posi-
that the objective working conditions are unlikely to tive psychological resources on both negative and positive
change within a foreseeable future. Therefore, measures mental health constructs, and revealed that PsyCap and
and strategies improving the inner psychological re- self-esteem might be positive resources for combating
sources of the manufacturing workers should be intro- stress symptoms. In the competitive society, mental health
duced in psychological health interventions. issues of employees are becoming more common, and
An increasing number of scholars have become inter- psychological health is a basic requirement for the staff to
ested in the associations between positive psychological bring their intelligence into play. Organizational managers
resources and mental health outcomes. PsyCap can help usually care more about advanced technologies and skills
employees fight against stressors and work more effect- than positive inner resources of their staff to improve
ively. It has been reported that individuals with higher organizational performance. According to the theory of
levels of PsyCap are able to have more confidence and the positive psychology, positive psychological capital is
make greater efforts to pursue success, preserve the will open to be developed and managed, including optimism,
to accomplish tasks or goals, bounce back from adversity self-efficacy, hope, resilience, self-esteem, and some other
or personal setbacks, and perceive positive expectations positive psychological resources. Interventions designed to
and attributes regarding consequences [27, 49]. In the enhance positive psychological resources have been
present study, PsyCap was found to be negatively associ- introduced in previous studies and attracting more
ated with emotional exhaustion and cynicism, respect- and more attention. Therefore, practical strategies of
ively, and positively associated with professional efficacy enhancing workers’ positive psychological resources
and well-being among the manufacturing workers. As should be developed to help workers better adjust to
another important inner resource for psychological and their working situations.
social functioning, self-esteem was found to be nega- Several limitations have to be mentioned in this study.
tively correlated with cynicism, and positively correlated The primary limitation of our study is the cross-
with professional efficacy and well-being, which were sectional nature of the data, which do not permit causal
consistent with the findings in previous studies [50, 51]. conclusions to be drawn. Although positive psycho-
Low self-esteem has been found to be a risk or vulnerabil- logical resources can account for a large part of the
ity factor for mental health. These results contributed to predictive associations between occupational stress and
Wang et al. BMC Psychiatry (2017) 17:364 Page 9 of 10

mental health, our study cannot confirm a meditational Ethics approval and consent to participate
model through cross-sectional design. Longitudinal This study received ethics approval from the Committee on Human
Experimentation of China Medical University (Shenyang, Liaoning, China).
studies are required to test these hypotheses, along with Written informed consent concerning conduct of the survey was obtained
more complex statistical models. Besides, as a study from the participants.
based on self-report measure, the results cannot rule out
other possible factors. Third, we couldn’t neglect that Consent for publication
Not applicable.
some associations such as the correlations between over-
commitment and professional efficacy are statistically Competing interests
significant, but very small. Finally, generalizations from The authors declare that they have no competing interests.
our data should be taken with caution since the sample
comprised only a small proportion of all workers in Publisher’s Note
manufacturing industries. Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

Author details
Conclusions 1
Department of Social Medicine, School of Public Health, China Medical
To summarize, our findings revealed that both extrinsic University, No.77 Puhe Road, Shenyang North NewArea, Shenyang, Liaoning,
stress (ERR) and intrinsic stress (overcommitment) were People’s Republic of China. 2School of Public Health, China Medical
University, No.77 Puhe Road, Shenyang North New District, Shenyang,
positively associated with emotional exhaustion and
Liaoning, People’s Republic of China.
cynicism, respectively, among manufacturing workers.
Only ERR was negatively associated with professional ef- Received: 16 March 2017 Accepted: 7 November 2017
ficacy. PsyCap mediated the effects of ERI on emotional
exhaustion and cynicism, and also mediated the effects References
of ERR on professional efficacy. Self-esteem mediated 1. Wang Y, Ramos A, Wu H, Liu L, Yang X, Wang J, et al. Relationship between
the effects of ERI on cynicism, and the effects of ERR on occupational stress and burnout among Chinese teachers: a cross-sectional
survey in Liaoning, China. International Archives of Occupational & Environmental
professional efficacy. Both PsyCap and self-esteem Health. 2015;88:589–97. doi:10.1007/s00420-014-0987-9.
mediated the relations between ERI and well-being. 2. Landsbergis PA, Diezroux AV, Fujishiro K, Baron S, Kaufman JD, Meyer JD, et al.
Hence, interventions to decrease Chinese manufactur- Job strain, occupational category, systolic blood pressure, and hypertension
prevalence: the multi-ethnic study of atherosclerosis. Journal of Occupational &
ing workers’ occupational stress and to enhance their Environmental Medicine. 2015;57:1178–84.
positive psychological resources should be developed 3. Gosselin E, Bourgault P, Lavoie S. Association between job strain, mental
and put into practice. health and empathy among intensive care nurses. Nurs Crit Care. 2016;21:
137–45. doi:10.1111/nicc.12064.
4. Wu H, Liu L, Sun W, Zhao X, Wang J, Wang L. Factors related to burnout
Additional file among Chinese female hospital nurses: cross-sectional survey in Liaoning
Province of China. J Nurs Manag. 2014;22:621–9. doi:10.1111/jonm.12015.
5. Rasmussen V, Turnell A, Butow P, Juraskova I, Kirsten L, Wiener L, et al.
Additional file 1: Survey questionnaires. (DOCX 11 kb) Burnout among psychosocial oncologists: an application and extension of
the effort-reward imbalance model. Psychoonchology. 2015;25:194–202.
doi:10.1002/pon.3902.
Abbreviations 6. Marano D, Boice Jr J, Fryzek J, Morrison J, Sadler C, McLaughlin J. Exposure
CY: Cynicism; EE: Emotional exhaution; ERI: Effort-Reward Imbalance; assessment for a large epidemiological study of aircraft manufacturing
ERR: Effort-Reward Imbalance Rate; FS: Flourishing scale; MBI-GS: Maslach workers. Appl Occup Environ Hyg. 2000;15:644–56. doi:10.1080/1047322
burnout inventory-general survey; OC: Over-Commitment; PE: Professional 0050075653.
efficacy; PsyCap: Psychological capital; WB: Well-being 7. Siegrist J. Adverse health effects of high-effort/low-reward conditions at
work. J Occup Health Psychol. 1996;1:27–43.
8. Siegrist J, Starke D, Chandola T, Godin I, Marmot M, Niedhammer I, et al.
Acknowledgements
The measurement of effort–reward imbalance at work: European comparisons.
The authors would like to thank all the participants who helped to get the
Socl. Sci Med. 2004;58:1483–99.
written informed consent about the survey and to distribute the questionnaires
9. Bakker A, Killmer C, Siegrist J, Schaufeli W. Effort-rewarding imbalance and
to the subjects.
burnout among nurses. J Adv Nurs. 2000;31:884–91.
10. Honkonen T, Ahola K, Pertovaara M, Isometsä E, Kalimo R, Nykyri E, et al. The
Funding association between burnout and physical illness in the general population-
None to declare. results from the Finnish health 2000 study. J Psychosom Res. 2006;61:59–66.
11. Maslach C, Schaufeli WB, Leiter MP. Job burnout. Ann Rev Psychol. 2001;52:
397–422. doi:10.1146/ [Link].52.1.397.
Availability of data and materials
12. Shin H, Noh H, Jang Y, Park Y, Lee S. A longitudinal examination of the
The datasets during and/or analysed during the current study available from
relationship between teacher burnout and depression. J Employ Couns.
the corresponding author on reasonable request.
2013;50:124–37. doi:10.1002/j.2161-1920.2013.00031.x.
13. Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev Psychol. 2001;
Authors’ contributions 52(1):397–422.
LW and ZYW conceived and designed the experiments; ZYW, HJY and SC 14. Schaufeli W, Enzmann D. The burnout companion to study and practice: a
performed the survey, collected data and input data; ZYW and YWW input critical analysis. London: Taylor & Francis; 1998.
data and analyzed the data; ZYW wrote the paper; HBL participated to review 15. Maslach C, Goldberg J. Prevention of burnout: new perspectives. Appl Prev
the paper. All authors have read and approved the final manuscript. Psychol. 1998;7:63–74.
Wang et al. BMC Psychiatry (2017) 17:364 Page 10 of 10

16. Wang Y, Ramos A, Wang L, et al. Relationship between occupational stress 43. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing
and burnout among Chinese teachers: a cross-sectional survey in Liaoning, and comparing indirect effects in multiple mediator models. Behav Res Meth.
China. Int Arch Occup Environ Health. 2015;88(5):589–97. 2008;40:879–91.
17. Bakker AB, Schaufeli WB, Demerouti E, Janssen PPM, Van der Hulst R, Brouwer 44. Shen X, Yang YL, Wang Y, Liu L, Wang S, Wang L. The association between
J. Using equity theory to examine the difference between burnout and occupational stress and depressive symptoms and the mediating role of
depression. Anxiety Stress Coping. 2000;13:247–68. doi:10.1080/106158000085 psychological capital among Chinese university teachers: a cross-sectional
49265. study. BMC Psychiatry. 2014;14:329. doi:10.1186/s12888-014-0329-1.
18. WHO. Promoting mental health: Concepts, emerging evidence, practice. 45. Jerg-Bretzke L, Fenkl M, Traue HC, Limbrecht-Ecklundt K. Effort-reward
Geneva: WHO.2005. imbalance among medical students and physicians. BMC Health Serv Res.
19. de Jonge J, Bosma H. Job strain, effort-reward imbalance and employee 2014;14:143. doi:10.1186/1472-6963-14-S2-P143.
well-being: a large-scale cross-sectional study. Soc Sci Med. 2000;50:1317–27. 46. Chou LP, Li CY, Hu SC. Job stress and burnout in hospital employees:
20. Ryan RM, Deci EL. On happiness and human potentials: a review of research comparisons of different medical professions in a regional hospital in
on hedonic and eudaimonic wellbeing. Annu Rev Psychol. 2001;52:141–66. Taiwan. BMJ Open. 2014;25 doi:10.1136/bmjopen-2013-004185.
21. Diener E, Wirtz D, Biswas-Diener R, Tov W, Kim-Prieto C, Choi D, et al. New 47. Girardi D, Falco A, De CA, et al. The mediating role of interpersonal conflict
measures of well-being. In: Diener E, editor. Assessing well-being: the collected at work in the relationship between negative affectivity and biomarkers of
works of Ed Diener. New York: Russell Sage Foundation; 2009. p. 247–66. stress. J Behav Med. 2015;38(6):922.
22. Diener E, Wirtz D, Tov W, Kim-Prieto C, Choi D, Oishi S, et al. New well-being 48. Hämmig O, Bauer GF. Work, work-life conflict and health in an industrial
measures: short scales to assess flourishing and positive and negative feelings. work environment. Occup Med (Lond). 2014;64:34–8. doi:10.1093/occmed/
Soc Indic Res. 2010;97(2):143–56. kqt127.
23. Folkman S, Lazarus RS, Gruen RJ. Appraisal, coping, health status, and 49. Luthans F, Norman SM, Avolio BJ, Avey JB. The mediating role of psychological
psychological [Link] of Personnality and. Soc Psychol. 1986; capital in the supportive organizational climate-employee performance
50:571–9. relationship. J Organ Behav. 2008;29:219–38.
24. Lazarus RS, Folkman S. Transactional theory and research on emotions and 50. Langballe EM, Innstrand ST, Aasland OG, Falkum E. The predictive value of
coping. Eur J Personal. 1987;1(3):141–69. doi:10.1002/per.2410010304. individual factors, work-related factors, and work-home interaction on burnout in
25. Lazarus RS, Folkman S. Stress, appraisal and coping. New York: Springer; 1984. female and male physicians: a longitudinal study. Stress Health. 2011;27:73–87.
26. Wang Y, Chang Y, Fu J, Wang L. Work-family conflict and burnout among doi:10.1002/smi.1321.
Chinese female nurses: the mediating effect of psychological capital. BMC 51. Li B, Ma HY, Guo YY, FM X, Yu F, Zhou ZK. Positive psychological capital: a
Public Health. 2012;12:915. doi:10.1186/1471-2458-12-915. new approach to social support and subjective well-being. Social Behavior
27. Luthans F, Avolio BJ, Avey JB, Norman SM. Positive psychological capital: & Personality: An International Journal. 2014;42:135–44.
measurement and relationship with performance and satisfaction. Pers
Psychol. 2007;60:541–72.
28. Rahimnia F, Mazidi AK, Mohammadzadeh Z. Emotional mediators of
psychological capital on well-being: the role of stress, anxiety, and
depression. Manage. Sci Lett. 2013;3:913–26.
29. Avey JB, Luthans F, Jensen SM. Psychological capital: a positive resource for
combating employee stress and turnover. Hum Resour Manag. 2009;48:677–93.
30. Park JG, Yoon SW, Kim JS, et al. The effects of empowering leadership on
psychological well-being and job engagement: the mediating role of
psychological capital. Leadership & Organization Development Journal.
2017;38(3):350–67.
31. Wang Y, Liu L, Wang J, et al. Work-family conflict and burnout among
Chinese doctors: the mediating role of psychological capital. J Occup
Health. 2012;54(3):232.
32. Rosenberg M. Society and the adolescent self-image. Princeton, NJ: Princeton
University Press; 1965.
33. Rosse JG, Boss RW, Johnson AE, Crown DF. Conceptualizing the role of self-
esteem in the burnout process. Group and Organization Management.
1991;16:428–51.
34. Richter A, Schraml K, Leineweber C. Work-family conflict, emotional exhaustion
and performance based self-esteem: reciprocal relationships. Int Arch Occup
Environ Health. 2015;88:103–12. doi:10.1007/s00420-014-0941-x.
35. Li J, Yang WJ, Cheng YW, Siegrist J, Cho S. Effort-reward imbalance at work
and job dissatisfaction in Chinese healthcare workers: a validation study. Int
arch occupy. Environ Health. 2005;78:198–204.
36. Zhang K, Zhang S, Dong YH. Positive psychological capital: measurement
and its association with mental health. Stu Psychol Behav. 2010;8:58–64.
37. Cheng ST, Hamid PN. An error in the use of translated scales: the
Rosenberg self-esteem scale for Chinese. Percept Mot Skills. 1995;81:431–4. Submit your next manuscript to BioMed Central
38. Maslach C, Jackson SE. The measurement of experienced burnout. J Occup and we will help you at every step:
Behav. 1981;2:99–103.
39. Schaufeli WB, Leiter MP, Maslach C, Jackson SE. Maslach burnout inventory- • We accept pre-submission inquiries
general survey (MBI-GS). In: Maslach C, Jackson SE, Leiter MP, editors. MBI • Our selector tool helps you to find the most relevant journal
manual. 3rd ed. Palo alto, CA: Consulting psychologists press; 1996. p. 19–26.
• We provide round the clock customer support
40. Li CP. Work-family conflict and job burnout of doctors and nurses. Chin
Ment Health. 2003;17:807–9. (in Chinese) • Convenient online submission
41. Diener E, Wirtz D, Tov W, Kim-Prieto C, Choi D, Oishi S, et al. New measures • Thorough peer review
of well-being: flourishing and positive and negative feelings. Soc Indic Res.
• Inclusion in PubMed and all major indexing services
2009;39:247–66.
42. Sumi K. Reliability and validity of Japanese versions of the flourishing scale • Maximum visibility for your research
and the scale of positive and negative experience. Soc Indic Res. 2014;
118(2):601–15. Submit your manuscript at
[Link]/submit