Liu et al.

BMC Psychiatry 2013, 13:89

http://www.biomedcentral.com/1471-244X/13/89

RESEARCH ARTICLE Open Access

Positive resources for combating depressive

symptoms among Chinese male correctional
officers: perceived organizational support and
psychological capital
Li Liu1, Shu Hu2, Lie Wang1*, Guoyuan Sui1 and Lei Ma1

Abstract
Background: Although correctional officers (COs) clearly suffer from depression, positive resources for combating
depression have been rarely studied in this population. The purpose of the study was to examine the associations
of perceived organizational support (POS) and psychological capital (PsyCap) with depressive symptoms among
Chinese COs.
Methods: A cross-sectional survey was conducted in a province of northeast China during March–April 2011. A
self-administered questionnaire was distributed to 1900 male COs from four male prisons. Depressive symptoms,
POS, and PsyCap (self efficacy, hope, resilience, and optimism) were measured anonymously. A total of 1428
effective respondents with 953 frontline COs (FL-COs) and 475 non-frontline COs (NFL-COs) became our final
sample. Hierarchical linear regression was performed to explore the factors associated with depressive symptoms.
Asymptotic and resampling strategies were used to examine the mediating roles of PsyCap and its four
components.
Results: The level of depressive symptoms of FL-COs was significantly higher than that of NFL-COs (t = 2.28, p = 0.023).
There were significant negative associations of POS, PsyCap, hope, resilience, and optimism with depressive symptoms
among FL-COs. In NFL-COs, POS, PsyCap, and optimism were negatively associated with depressive symptoms. POS
was positively associated with PsyCap and its four components among both FL-COs and NFL-COs. For FL-COs, PsyCap
(a*b = −0.143, BCa 95% CI: –0.186, –0.103, p < 0.05), resilience (a*b = −0.052, BCa 95% CI: –0.090, –0.017, p < 0.05),
and optimism (a*b = −0.053, BCa 95% CI: –0.090, –0.016, p < 0.05) significantly mediated the association between
POS and depressive symptoms. For NFL-COs, PsyCap (a*b = −0.126, BCa 95% CI: –0.186, –0.065, p < 0.05) and
optimism (a*b = −0.066, BCa 95% CI: –0.116, –0.008, p < 0.05) significantly mediated the association.
Conclusions: Perceived organizational support and psychological capital could be positive resources for
combating depressive symptoms in Chinese male COs. Psychological capital and its components (resilience and
optimism) partially mediate the association between perceived organizational support and depressive symptoms.
Therefore, organizational support and psychological capital investment (especially resilience and optimism)
should be included in depression preventions and treatments targeting Chinese male COs.
Keywords: Depressive symptoms, Positive psychological capital, Perceived organizational support, Mediating role,
Occupational psychology, Correctional officers

* Correspondence: liewang@mail.cmu.edu.cn
1
Department of Social Medicine, School of Public Health, China Medical
University, 92 North 2nd Road, Heping District, Shenyang, Liaoning 110001,
People's Republic of China
Full list of author information is available at the end of the article

© 2013 Liu et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Liu et al. BMC Psychiatry 2013, 13:89 Page 2 of 9
http://www.biomedcentral.com/1471-244X/13/89

Background depression and anxiety [15,16]. As an important personal

Depression is common in the general population and is resource, the higher-order core construct of psychological
associated with extensive adverse effects in the work- capital (PsyCap) includes four state-like psychological
place. In the European Union, 20% of the workforce capacities, including self efficacy, hope, resilience, and op-
suffers from some type of mental health disorder at any timism, which can be measured, developed, and effectively
given time, particularly depression [1]. In China, with managed [17]. PsyCap has significantly positive effects on
the rapid development of the society and economy, employee performance, satisfaction, organizational com-
many workers face occupational mental health disorders. mitment, and well-being in the workplace [17-19]. It is
The average prevalence of depressive symptoms has negatively related to job stress, cynicism, and burnout
been reported as 46.2% among various Chinese occupa- [20-22]. However, to our knowledge, the roles of POS and
tional populations including teachers, foreign enterprise PsyCap in combating depression have not been studied
employees, managers, researchers, traffic police, and among Chinese COs. Employees who perceive they work
community health workers [2]. The World Health for more supportive organizations may be more likely to
Organization (WHO) estimates that depression will experience higher levels of PsyCap, which in turn has
become the second-leading cause of disability worldwide positive effects on their work behavior and job perform-
by 2020 [3]. Depression has substantial adverse effects ance [23,24]. In addition, whether PsyCap mediates the
on workers’ behaviors, job performance, and quality of association between POS and depression has not been
life [4-6]. confirmed. It is important to examine the positive effects
Correctional officers (COs), working in a closed, coer- of the four PsyCap components to develop specific preven-
cive, and harsh prison environment, bearing multiple tions and treatments for depression because each compo-
responsibilities for the correction, education and man- nent is conceptually and psychometrically distinct [25-27].
agement of prisoners, and the maintenance of security In light of the above concerns, the purpose of the
and order inside prisons, are prone to suffer from present study was to verify the following three assump-
various mental health disorders, especially depression tions among Chinese COs: 1) POS as well as PsyCap
[7-9]. In a French study, the overall proportion of COs and its components are negatively associated with
with depressive symptoms was 24% [7]. Approximately depressive symptoms, 2) POS is positively associated
one-third of the COs employed at two state prisons in with PsyCap and its components, and 3) PsyCap and its
the northeastern U.S. reported serious psychological components mediate the association between POS and
distress [9]. This issue is particularly serious among depressive symptoms.
Chinese COs because of high work demand and short-
age of COs due to heavy prisoner load resulting from Methods
China’s huge population [10]. The ratio of COs to Study design and sample
prisoners is 1:5 in China, which is lower than that in A cross-sectional survey was conducted in a province of
Western countries (1:3–1:4) [11]. As a result, COs are northeast China during March–April 2011. The province
often required to work overtime, especially frontline comprises three metropolitan cities with more than
COs (FL-COs) who have more contact with prisoners 1,000,000 population, seven medium-sized cities with
than other prison employees. Moreover, career ad- 500,000–1,000,000 population, and four small cities with
vancement of COs is stunted by the relative disadvan- 200,000–500,000 population. Based on geographic div-
tage associated with lack of higher education [12]. ision and economic development, one metropolitan, two
Overall, COs’ efforts are not met with adequate reward medium-sized cities, and one small city were selected. In
[10]. Hu et al. [13] reported that the prevalence of China, the Ministry of Justice and the Bureau of Prison
depressive symptoms was 60.5% among Chinese COs. Administration supervise all correctional facilities and
Depression can impact not only their career development provide funds to ensure their normal operation (The
and well-being, but also affect their functions related to Prison Law of the People’s Republic of China). After
correction and education that can help prisoners return obtaining agreement from the provincial department of
successfully to society. Therefore, more research on the justice and prison superintendents, one male prison was
prevention and treatment of depression in Chinese COs is selected from each city. The selected four prisons house
urgently needed. However, positive resources for combat- prisoners with a sentence of more than 10 years. Female
ing depression have been rarely studied in this population. COs engage mainly in administrative and technical work
According to organizational support theory [14], em- in Chinese male prisons. Thus, male COs were our
ployees develop higher levels of perceived organizational focus. We randomly sampled 50% of male COs from
support (POS) when their organization cares about their each department of selected prisons. In addition, these
well-being and values their contributions. Consequently, participants were divided into two occupational categories:
employees with high POS are less likely to experience FL-COs (working in direct contact with prisoners) and
Liu et al. BMC Psychiatry 2013, 13:89 Page 3 of 9
http://www.biomedcentral.com/1471-244X/13/89

non-frontline COs (NFL-COs, including administrative and adequately applied among Chinese occupational
staff and educational, technical, and healthcare workers). populations [22,33,34]. For the total scale, alpha was
After the participants were given a brief and complete de- 0.93 for both FL-COs and NFL-COs. Alpha for self effi-
scription of the study, written informed consent was cacy, hope, resilience, and optimism subscales were
obtained. A self-administered questionnaire was directly 0.84, 0.86, 0.83, and 0.75 for FL-COs and 0.86, 0.86,
distributed to 1900 male COs. The sampled COs com- 0.87, and 0.77 for NFL-COs, respectively.
pleted the questionnaire anonymously in a private place The age, marital status, education, occupational category
after their shift was over. Complete responses were and monthly income (RMB) of COs were obtained in this
obtained from 1428 individuals (response rate: 75.2%), of study. Marital status was categorized as single/widowed/
whom 953 (66.7%) were FL-COs and 475 (33.3%) were divorced or married/cohabiting. Education was catego-
NFL-COs. The study was approved by the Committee rized as junior college or lower and college or higher.
on Human Experimentation of China Medical Univer- Occupational category was categorized as FL-COs and
sity, and the study procedures were in accordance with NFL-COs. Monthly income was categorized as ≤ 3000
ethical standards. yuan, 3001–4000 yuan and > 4000 yuan.

Measures Statistical analysis

The Chinese version of the Center for Epidemiologic We analyzed the data for FL-COs and NFL-COs separ-
Studies Depression Scale (CES-D) was used as the indi- ately to allow for possible occupational differences in
cator of depressive symptoms [28]. It consists of 20 variable associations. Pearson’s chi-square (χ2) tests were
items with 4 possible options that describe the frequency used to compare differences in demographic characteris-
of respondents’ feelings in the past week ranging from 0 tics between FL-COs and NFL-COs. Differences in con-
‘rarely or none of the time (less than 1 day)’ to 3 ‘most tinuous variables were examined by t-tests or one-way
or all of the time (5 to 7 days)’. The severity of depres- ANOVAs. Pearson’s correlation analyses were executed
sive symptoms increases with a higher summed score to examine the correlations among continuous variables.
and ranges from 0 to 60. The CES-D has been widely Univariate analyses were carried out to determine vari-
used among Chinese occupational populations with ables that were correlated with depressive symptoms
good reliability and validity [2,29]. In the present study, with adjustment for age, marital status, education, and
Cronbach’s alpha coefficient for the CES-D was 0.91 for monthly income. The significant independent variables
FL-COs and 0.90 for NFL-COs. were then entered in the regression models. Hierarchical
POS was assessed using the 9-item version of the linear regression analysis was performed to explore the
Survey of Perceived Organizational Support (SPOS) effects of groups of independent variables on depressive
[14,30]. The scale consists of statements mainly concerning symptoms. In Block 1, age, marital status, education, and
the organization’s valuation and well-being of employees. monthly income were added. POS was added in Block 2.
Each item is scored on a 7-point Likert scale ranging from There were two models (Model 1 and Model 2) in Block 3.
1 ‘strongly disagree’ to 7 ‘strongly agree’. In this study, the PsyCap was added in Model 1, and self efficacy, hope, re-
average score of the 9 items was used as the indicator of silience and optimism were added in Model 2. Asymptotic
POS. Higher values indicated higher levels of POS. The and resampling strategies were used to examine the medi-
9-item SPOS has been extensively applied and validated ating roles (a*b product) of PsyCap and its four compo-
among Chinese occupational groups [31,32]. Cronbach’s nents on the association between POS and depressive
alpha for the POS scale was 0.89 for FL-COs and 0.88 for symptoms, respectively [35]. In these equations, POS was
NFL-COs. modeled as the independent variable, CES-D score as the
The Chinese version of the 24-item Psychological dependent variable, PsyCap and its components as the
Capital Questionnaire (PCQ) was used to measure mediators, and age, marital status, education, and monthly
PsyCap [17]. The PCQ consists of four subscales: self income as covariates. The bootstrap estimate was based on
efficacy (6 items), hope (6 items), resilience (6 items), 5000 bootstrap samples. Then, the bias-corrected and
and optimism (6 items). Each item has six responses accelerated 95% confidence interval (BCa 95% CI) for each
with categories ranging from 1 ‘strongly disagree’ to 6 a*b product was investigated, and a BCa 95% CI not
‘strongly agree’. Average scores for the total scale and including 0 indicated a significant mediating role. All study
each subscale were calculated as indicators of PsyCap, variables were centralized before analysis to account for
self efficacy, hope, resilience, and optimism in this differences in scale scores. Moreover, tolerance and
study, with higher scores indicating greater psycho- variance inflation factors were used to check for mul-
logical capacities. Adequate reliability and construct ticollinearity. All analyses were conducted using SPSS for
validity of the PCQ has been demonstrated across mul- Windows, Ver. 13.0. Statistical significance was defined as
tiple samples [17]. The Chinese PCQ has been initially p < 0.05 (two-tailed).
Liu et al. BMC Psychiatry 2013, 13:89 Page 4 of 9
http://www.biomedcentral.com/1471-244X/13/89

Results Associations of POS and PsyCap with depressive symptoms

Characteristics of subjects All independent variables including POS, PsyCap, self
Demographic characteristics and differences in depres- efficacy, hope, resilience, and optimism were associated
sive symptoms are presented separately for FL-COs and with depressive symptoms in univariate analyses with
NFL-COs in Table 1. NFL-COs were significantly more adjustment for age, marital status, education, and
likely to be married (χ2 = 39.28, p < 0.001), had lower edu- monthly income among both FL-COs and NFL-COs.
cation (χ2 = 6.08, p = 0.014) and earned more (χ2 = 75.31, Thus, they were entered in the hierarchical multiple
p < 0.001) than FL-COs. For FL-COs, married/cohabiting regression models of FL-COs and NFL-COs.
had a higher mean CES-D score than single/widowed/ Hierarchical multiple regression analysis results are
divorced (t = 2.29, p = 0.022), and those with junior college presented in Table 3. For FL-COs, POS was significantly
or lower education had higher CES-D score than those and negatively associated with depressive symptoms in
with college or higher education (t = 2.57, p = 0.010). Block 2. In Block 3 Model 1, POS and PsyCap were
For NFL-COs, those married/cohabiting had a lower significantly and negatively associated with depressive
mean CES-D score than those single/widowed/divorced symptoms. In Block 3 Model 2, POS, hope, resilience,
(t = −2.59, p = 0.010). and optimism were significantly and negatively associ-
ated with depressive symptoms. For NFL-COs, POS was
significantly and negatively associated with depressive
Differences and correlations among continuous variables symptoms in Block 2. In Block 3 Model 1, POS and
Differences in continuous variables between FL-COs and PsyCap were significantly and negatively associated with
NFL-COs and correlations among continuous variables depressive symptoms. In Block 3 Model 2, POS and
are presented in Table 2. The average age of FL-COs optimism were significantly and negatively associated
(37.7 years) was significantly younger than that of NFL- with depressive symptoms. In addition, the effect of POS
COs (43.2 years) (t = 10.29, p < 0.001). The mean CES-D on depressive symptoms in Block 3 was reduced com-
score of FL-COs was significantly higher than that of pared with that in Block 2 among both FL-COs and
NFL-COs (t = 2.28, p = 0.023). The mean POS score of NFL-COs, as indicated by smaller β coefficients.
FL-COs was significantly lower than that of NFL-COs
(t = −2.31, p = 0.021). However, there were no significant Mediating effects of PsyCap and its components
differences in PsyCap, self efficacy, hope, resilience, or opti- Path coefficients a (between POS and mediators) and b
mism between FL-COs and NFL-COs. Age was positively (between mediators and depressive symptoms), a*b
correlated with CES-D score among FL-COs (r = 0.07, products, and BCa 95% CI for these products are
p = 0.037). CES-D score had negative correlations with presented in Table 4. For both FL-COs and NFL-COs,
POS, PsyCap, self efficacy, hope, resilience, and optimism POS was significantly and positively associated with
among both FL-COs and NFL-COs. PsyCap, self efficacy, hope, resilience, and optimism.

Table 1 Demographic characteristics and differences in depressive symptoms

Characteristics FL-COs (N = 953) NFL-COs (N = 475)
N % CES-D N % CES-D
Mean ± SD Mean ± SD
Marital status
Single/widowed/divorced 250 26.2 17.71 ± 10.94 56 11.8**a 21.06 ± 12.85*
Married/cohabitation 703 73.8 19.59 ± 11.20* 419 88.2 17.25 ± 9.98
Education
Junior college or lower 318 33.4*a 20.40 ± 10.60* 190 40.0 18.44 ± 10.22
College or higher 635 66.6 18.44 ± 11.38 285 60.0 17.20 ± 10.53
Monthly income (yuan)
≤ 3000 355 37.3 18.00 ± 11.27 79 16.6**a 18.68 ± 12.40
3001–4000 522 54.8 19.69 ± 11.09 315 66.3 17.36 ± 10.24
> 4000 76 8.00 20.14 ± 10.83 81 17.1 18.06 ± 9.84
FL-COs: frontline correctional officers; NFL-COs: non-frontline correctional officers; CES-D: the Center for Epidemiologic Studies Depression Scale; POS: perceived
organizational support; PsyCap: psychological capital.
a
Comparison between FL-COs and NFL-COs.
* p < 0.05, ** p < 0.01.
Liu et al. BMC Psychiatry 2013, 13:89 Page 5 of 9
http://www.biomedcentral.com/1471-244X/13/89

Table 2 Differences and correlations among continuous variables

Variables FL-COs (N = 953) NFL-COs (N = 475)
Mean ± SD Pearson r with CES-D Mean ± SD Pearson r with CES-D
CES-D 19.10 ± 11.16 – 17.70 ± 10.41* –
Age 37.73 ± 9.80** 0.07* 43.22 ± 8.84 −0.01
POS 4.49 ± 1.23* −0.41** 4.65 ± 1.15 −0.32**
PsyCap 4.31 ± 0.74 −0.42** 4.35 ± 0.72 −0.38**
Self efficacy 4.39 ± 0.85 −0.29** 4.40 ± 0.86 −0.27**
Hope 4.23 ± 0.90 −0.39** 4.27 ± 0.87 −0.34**
Resilience 4.36 ± 0.85 −0.36** 4.42 ± 0.84 −0.32**
Optimism 4.24 ± 0.90 −0.37** 4.28 ± 0.87 −0.36**
FL-COs: frontline correctional officers; NFL-COs: non-frontline correctional officers; CES-D: the Center for Epidemiologic Studies Depression Scale; POS: perceived
organizational support; PsyCap: psychological capital.
* p < 0.05, ** p < 0.01.

Consistent with the results from hierarchical multiple Chinese male general population [37]. The prevalence was
regression, PsyCap, hope, resilience, and optimism were also higher than that of male COs (24.9%) in France [7].
significantly and negatively associated with depressive In addition, the level of depressive symptoms in FL-
symptoms after controlling for age, marital status, edu- COs was higher compared with NFL-COs. Previous
cation, monthly income, and POS among FL-COs. Thus, studies have shown that FL-COs suffer more serious
significant mediating roles of PsyCap (a*b = −0.143, BCa mental health disorders than NFL-COs that can be
95% CI: –0.186, –0.103), resilience (a*b = −0.052, BCa attributed to occupational differences [38-40]. FL-COs
95% CI: –0.090, –0.017), and optimism (a*b = −0.053, and NFL-COs represent two significantly different work
BCa 95% CI: –0.090, –0.016) on the association between groups, despite working at the same place [39]. FL-COs
POS and depressive symptoms were revealed among work night shifts and spend almost all their working
FL-COs. For NFL-COs, PsyCap and optimism were sig- time in direct contact with the criminals. They bear
nificantly and negatively associated with depressive great responsibility for prisoner correction as well as
symptoms. Thus, PsyCap (a*b = −0.126, BCa 95% CI: – prison safety and control, and often face more stressors
0.186, –0.065) and optimism (a*b = −0.066, BCa 95% than NFL-COs in the workplace. NFL-COs have less
CI: –0.116, –0.008) significantly mediated the associ- direct contact with the criminals and fewer night shifts
ation between POS and depressive symptoms. as well as more flexibility in performing tasks. We also
The proportion of total effect of POS on depressive found significant differences in demographic characteris-
symptoms by mediator role was calculated with the for- tics and POS between FL-COs and NFL-COs. Therefore,
mula ‘(a*b)/total effect’. For FL-COs, the proportions of potential occupational differences should be taken into
mediating roles of PsyCap, resilience, and optimism were account to assess and effectively deal with depressive
35.3%, 12.8%, and 13.1%, respectively. For NFL-COs, the symptoms. We analyzed the data for FL-COs and NFL-
proportions of mediating roles of PsyCap and optimism COs separately.
were 38.5% and 20.2%, respectively. It has been suggested that POS decreases workplace
stressors and has the potential for combating job-related
burnout, anxiety, and depression [41]. Our finding that
Discussion POS was negatively associated with depressive symptoms
Findings show that in China, male COs seriously suffer among both FL-COs and NFL-COs supports this. POS
from depressive symptoms, and POS and PsyCap can be is a highly effective resource that can predict a wide
positive resources for combating these symptoms. This range of positive work attitudes and outcomes [41], and
is the first study to confirm the mediating roles of may help employees avoid depression and other mental
PsyCap and its components on the association between health disorders. The significant difference in POS might
POS and depressive symptoms. be one of the important explanations for the difference
The prevalence of at least mild to moderate depression in depressive symptoms between FL-COs and NFL-COs.
was quite high, as 59.7% of FL-COs and 56.2% of NFL- Furthermore, the positive effects of PsyCap on work
COs scored above 16 on the CES-D [2,29,36]. The level of attitudes and outcomes can help employees combat job
depressive symptoms in our sample was higher than that stress, cynicism, burnout, and depression. Components
of various occupational groups in Shanghai [2] and ap- of PsyCap have positive associations with desirable work
proximately 1.5 times higher than that found in the attitudes and work performance [25-27,42]. Some of
Liu et al. BMC Psychiatry 2013, 13:89 Page 6 of 9
http://www.biomedcentral.com/1471-244X/13/89

Table 3 Results from the hierarchical multiple regression Table 4 Mediating roles of PsyCap and its components on
analyses the POS-depressive symptoms association
Variables Block 1 Block 2 Block 3 (β) Mediators a b a*b (BCa 95% CI)
(β) (β)
Model 1 Model 2 FL-COs
FL-COs (N = 953) PsyCap 0.485** −0.294** −0.143* (−0.186, −0.103)
(N = 953) Age −0.015 −0.022 −0.034 −0.037 Self efficacy 0.358** 0.025 0.009 (−0.031, 0.042)
Marital status 0.045 0.009 0.014 0.013 Hope 0.500** −0.104* −0.052 (−0.106, 0.001)
Education −0.070* −0.045 −0.025 −0.028 Resilience 0.342** −0.152** −0.052* (−0.090, −0.017)
Monthly income 1 0.033 0.070 0.096* 0.099** Optimism 0.425** −0.124** −0.053* (−0.090, −0.016)
Monthly income 2 0.046 0.054 0.078 0.080* NFL-COs
POS −0.405** −0.263** −0.257** (N = 475) PsyCap 0.423** −0.299** −0.126* (−0.186, −0.065)
PsyCap −0.294** Self efficacy 0.230** −0.093 −0.021 (−0.060, 0.012)
Self efficacy 0.025 Hope 0.491** −0.040 −0.020 (−0.115, 0.066)
Hope −0.104* Resilience 0.358** −0.045 −0.016 (−0.068, 0.029)
Resilience −0.152** Optimism 0.339** −0.195** −0.066* (−0.116, −0.008)
Optimism −0.124** PsyCap: psychological capital; POS: perceived organizational support; FL-COs:
frontline correctional officers; NFL-COs: non-frontline correctional officers; BCa
F 2.238* 32.993** 42.088** 31.043** 95% CI: the bias-corrected and accelerated 95% confidence interval.
Adjusted R2 0.006 0.168 0.232 0.240 a: associations of POS with PsyCap and its components; b: associations of
PsyCap and its components with depressive symptoms after controlling for
ΔR2 0.012* 0.161** 0.065** 0.075** the predictor variables; a*b: the product of a and b.
Age, marital status, education, and monthly income were covariates.
NFL-COs * p < 0.05, ** p < 0.01.
(N = 475) Age 0.016 0.010 0.000 −0.002
Marital status −0.123* −0.125* −0.121* −0.119* when faced with adversity in a constantly changing work-
Education −0.055 −0.051 −0.059 −0.056 place environment [44]. Optimism is associated with a
Monthly income 1 0.006 −0.011 −0.002 −0.001
positive outlook or attribution of success, which includes
positive emotions and motivation [45]. Workers with
Monthly income 2 −0.005 −0.027 −0.025 −0.016
higher levels of optimism are less likely to experience
POS −0.327** −0.201** −0.204** symptoms of stress in the workplace [46]. However, only
PsyCap −0.299** optimism was negatively associated with depressive symp-
Self efficacy −0.093 toms for NFL-COs. The different psychological environ-
Hope −0.040 ment related to occupational categories is one possible
Resilience −0.045
reason for this finding. In prison, FL-COs face more
occupational stressors considered risk factors for de-
Optimism −0.195**
pression than NFL-COs, such as long working hours,
F 1.726 11.102** 16.465** 12.152** role conflict, personal safety, and effort-reward imbal-
Adjusted R2 0.008 0.113 0.186 0.190 ance [39]. Moreover, mean age of FL-COs was signifi-
ΔR 2
0.018 0.107** 0.073** 0.083** cantly lower than that of NFL-COs in our current data.
FL-COs: frontline correctional officers; NFL-COs: non-frontline correctional These younger employees may face more problems out-
officers; POS: perceived organizational support; PsyCap: psychological capital. side the workplace related to family and social life [2,47].
Marital status: married/cohabitation vs. single/widowed/divorced; Education:
college or higher vs. junior college or lower; Monthly income 1: 3001–4000 Compared with NFL-COs, FL-COs may need more com-
yuan vs. ≤ 3000 yuan; Monthly income 2: > 4000 yuan vs. ≤ 3000 yuan. prehensive psychological capacities to accomplish their
There were two models (Model 1 and Model 2) in Block 3. PsyCap was added
in Model 1, and self efficacy, hope, resilience and optimism were added in
multiple work functions and effectively deal with negative
Model 2. psychological consequences.
* p < 0.05, ** p < 0.01. This is the first study to confirm the mediating roles
of PsyCap and its components on the POS-depressive
them also have positive effects on individual’s emotional symptoms association. The COs who perceive more
functions [21,43-46]. For FL-COs, hope, resilience, and POS at work may be more likely to experience higher
optimism had negative associations with depressive levels of PsyCap, which in turn reduces their depressive
symptoms. Hope may provide individuals a positive symptoms. Among the PsyCap components, resilience
resource for anxiety while protecting against perceptions and optimism partially mediated the association between
of vulnerability, uncontrollability and unpredictability POS and depressive symptoms in FL-COs, and only op-
[21,43]. Resilient individuals show more emotional stability timism partially mediated the POS-depressive symptoms
Liu et al. BMC Psychiatry 2013, 13:89 Page 7 of 9
http://www.biomedcentral.com/1471-244X/13/89

association in NFL-COs. POS can create positive condi- used a cross-sectional design, making it impossible to
tions for developing PsyCap [48]. Individuals who per- draw causal relations among study variables. However,
ceive a high level of organizational support feel study hypotheses were built on a solid theoretical and
confident and hopeful about their desired job goals and research foundation. The results of our cross-sectional
are more likely to engage in voluntary behaviors. In study need to be confirmed in prospective settings. Second,
addition, organizational support can help employees deal the study population comprised only Chinese male COs
with problems, and promote the development of opti- from prisons that house individuals with a sentence of
mism and self-attribution regarding personal accom- more than 10 years, which may limit the generalization of
plishment [48]. the results to other Chinese prison populations. Finally,
Findings from our study have practical implication for gender differences in the associations between variables
prevention and treatment of depressive symptoms remain unaddressed and need further examination.
among Chinese male COs because PsyCap can be devel-
oped in a variety of ways [49,50]. Continued investment Conclusions
in financial, human, and social capital may no longer be Chinese male COs suffer from severe depressive symptoms.
sufficient in the workplace. Investment in PsyCap may POS and PsyCap could be positive resources for combating
yield substantial positive returns beyond the traditional depressive symptoms, with POS also being a positive re-
forms of capital investment [20,51]. source for developing PsyCap and its four components
As a composite of psychological capacities, the direct among both FL-COs and NFL-COs. In addition, PsyCap,
effect of PsyCap on depressive symptoms and its mediat- resilience, and optimism partially mediate the association
ing role on the POS-depressive symptoms association between POS and depressive symptoms in FL-COs. For
were greater than the specific effects of PsyCap compo- NFL-COs, PsyCap and optimism partially mediate the
nents among both FL-COs and NFL-COs. This result POS-depressive symptoms association. Therefore, POS and
was consistent with a previous study that indicated the PsyCap investment (especially resilience and optimism)
synergistic role of PsyCap components [52]. Therefore, should be included in depression preventions and treat-
investment in each component is necessary to achieve ments targeting Chinese male COs.
maximum positive outcomes.
Effective strategies should be implemented with Chinese Abbreviations
male COs to improve POS and PsyCap, and to relieve CES-D: The center for epidemiologic studies depression scale;
FL-COs: Frontline correctional officers; NFL-COs: Non-frontline correctional
depressive symptoms. Especially, resilience and optimism
officers; POS: Perceived organizational support; PsyCap: Psychological capital.
should be given more attention in PsyCap investment.
Policy makers and managers should improve the level of Competing interests
POS in COs by establishing fair and impartial procedures, The authors declare that they have no competing interests.
affirming the contribution of COs and giving adequate
praise and awards, maintaining effective communication Authors’ contributions
LL designed the research, carried out data analysis and wrote the paper. SH
with COs, helping them with career planning, providing organized the investigation. LW provided guidance in study design and was
good working conditions, improving job autonomy, and the corresponding author of the paper. GYS and LM provided help in the
caring about their well-being [41,53,54]. For self-efficacy data collection, data analysis, results interpreting and paper writing. All
authors read and approved the final manuscript.
improvement, managers should provide opportunities for
COs to develop career interests and skills, experience
Acknowledgments
success, and boost their confidence. To facilitate a sense The authors would like to thank the administrators from each selected
of hope, managers should help COs set appropriate, spe- prison who helped us get the written informed consents and distribute the
questionnaires. We also thank all the correctional officers voluntary
cific, and challenging job goals, create multiple pathways
participating in this study.
to achieve them, and activate a reward system. To im-
prove optimism, managers should encourage COs to Author details
1
Department of Social Medicine, School of Public Health, China Medical
regard past failures and setbacks as valuable experience,
University, 92 North 2nd Road, Heping District, Shenyang, Liaoning 110001,
develop a positive attribution style, and enhance their abil- People's Republic of China. 2Department of Basic Law, School of Forensic
ity to discover and pursue various opportunities. For resili- Medicine, China Medical University, 92 North 2nd Road, Heping District,
Shenyang, Liaoning 110001, People's Republic of China.
ence improvement, managers should encourage COs to
fully use their existing personal resources, develop solu- Received: 11 June 2012 Accepted: 12 March 2013
tions to avoid or overcome obstacles, facilitate critical Published: 19 March 2013
reflection, and get adequate rest when burnout occurs
[48-50,55,56]. References
1. McDaid D, Curran C, Knapp M: Promoting mental well-being in the
However, before conclusions can be drawn, limitations workplace: A European policy perspective. Int Rev Psychiatry 2005,
of this study must be acknowledged. First, our study 17:365–373.
Liu et al. BMC Psychiatry 2013, 13:89 Page 8 of 9
http://www.biomedcentral.com/1471-244X/13/89

2. Dai J, Yu H, Wu J, Wu C, Fu H: Analysis on the association between job 26. Larson M, Luthans F: Potential added value of psychological capital in
stress factors and depression symptoms. Wei Sheng Yan Jiu 2010, predicting work attitudes. J LOS 2006, 13:75–92.
39:342–346 (in Chinese). 27. Youssef CM, Luthans F: Positive organizational behavior in the workplace:
3. Murray CJ, Lopez AD: Alternative projections of mortality and disability the impact of hope, optimism, and resiliency. J Manag 2007, 33:774–800.
by cause 1990–2020: global burden of disease study. Lancet 1997, 28. Radloff LS: The CES-D scale: A self-report depression scale for research in
349:1498–1504. the general population. Appl Psychol Meas 1977, 1:385–401.
4. Haslam C, Atkinson S, Brown SS, Haslam RA: Anxiety and depression in the 29. Qiu P, Caine E, Yang Y, Chen Q, Li J, Ma X: Depression and associated factors
workplace: effects on the individual and organisation (a focus group in internal migrant workers in China. J Affect Disord 2011, 134:198–207.
investigation). J Affect Disord 2005, 88:209–215. 30. Eisenberger R, Fasolo P, Davis-LaMastro V: Perceived organizational
5. Adler DA, McLaughlin TJ, Rogers WH, Chang H, Lapitsky L, Lerner D: Job support and employee diligence, commitment, and innovation.
performance deficits due to depression. Am J Psychiatry 2006, 163:1569–1576. J Appl Psychol 1990, 75:51–59.
6. Rusli BN, Edimansyah BA, Naing L: Working conditions, self-perceived 31. Chen Z, Aryee A, Lee C: Test of a mediation model of perceived
stress, anxiety, depression and quality of life: a structural equation organizational support. J Vocat Behav 2005, 66:457–470.
modelling approach. BMC Publ Health 2008, 8:48. 32. Wang LH, Xu CJ: Impact of primary and secondary school teachers’
7. Goldberg P, David S, Landre MF, Goldberg M, Dassa S, Fuhrer R: Work perceived organizational support on their happiness and job burnout.
conditions and mental health among prison staff in France. Scand J Work Chin J Clin Psychol 2008, 16:574–578 (in Chinese).
Environ Health 1996, 22:45–54. 33. Wen L, Qi SS, Zhang YZ: A preliminary revision of Luthans psychological
8. Bourbonnais R, Jauvin N, Dussault J, Vézina M: Psychosocial work capital questionnaire. Chin J Clin Psychol 2009, 17:148–150 (in Chinese).
environment, interpersonal violence at work and mental health among 34. Sun T, Zhao XW, Yang LB, Fan LH: The impact of psychological capital on
correctional officers. Int J Law Psychiatry 2007, 30:355–368. job embeddedness and job performance among nurses: a structural
9. Obidoa C, Reeves D, Warren N, Reisine S, Cherniack M: Depression and equation approach. J Adv Nurs 2012, 68:69–79.
work family conflict among corrections officers. J Occup Environ Med 35. Preacher KJ, Hayes AF: Asymptotic and resampling strategies for
2011, 53:1294–1301. assessing and comparing indirect effects in multiple mediator models.
10. Cui FJ, Zheng XB, Xiang YM, Guan YS: Comparison in work stress, self- Behav Res Meth 2008, 40:879–891.
esteem and subjective well-being between prison policemen and street 36. Nakasujja N, Skolasky RL, Musisi S, Allebeck P, Robertson K, Ronald A,
office workers. China J Health Psychol 2011, 19:802–804 (in Chinese). Katabira E, Clifford DB, Sacktor N: Depression symptoms and cognitive
11. Dong SZ: Comparative studies of Chinese and foreign systems for prison function among individuals with advanced HIV infection initiating
officers. J Beijing Manage College Politics and Law 2002, 4:51–54 (in Chinese). HAART in Uganda. BMC Psychiatry 2010, 10:44.
12. Lu JG: Research on the mental health and occupational stress status of 37. Zhang J, Wu ZY, Fang G, Li J, Han BX, Chen ZY: Development of the
prison polices. Sci Soc Psychol 2010, 25:208–213 (in Chinese). Chinese age norms of CES-D in urban area. Chin Ment Health J 2010,
13. Hu S, Meng LY, Ma L, Wang L: The interaction of occupational stress, 24:139–143. in Chinese.
depressive symptoms and perceived organizational support among 38. Zhang L, Deng XP, Chen CY: Survey on social support, general well-being
correctional officers. Chinese J Health Stat 2012, 29:155 (in Chinese). and mental health in prison guards. China J Health Psychol 2009,
14. Eisenberger R, Huntington R, Hutchison S, Sowa D: Perceived 17:936–939 (in Chinese).
organizational support. J Appl Psychol 1986, 71:500–507. 39. Qin ZG, Liu DF, Zong Y, Wang XG, Huang TY, Zhang QS: Study on the
15. Grant-Vallone EJ, Ensher EA: An examination of work and personal life psychological health status of the western provinces under the age of
conflict, organizational support and employee health among 40 prison police using symptom checklist 90. Chinese J Social Med 2008,
international expatriates. Int J Intercult Relat 2001, 25:261–278. 25:348–350 (in Chinese).
16. Haar J, Brougham D: Cultural and organisational perceptions of support 40. Song SZ, Du YH, Li L: A survey report on the mental health and
towards mental health outcomes: A study of Maori employees. Labour, psychological quality of correctional officers. China Prison J 2010,
employment and work in New Zealand. Proceedings of the Fourteenth 5:124–129 (in Chinese).
Conference. 2010. https://www.victoria.ac.nz/som/industrial-relations-centre/ 41. Rhoades L, Eisenberger R: Perceived organizational support: a review of
irc-events/lew-conference-welcome/lew-papers/ the literature. J Appl Psychol 2002, 87:698–714.
J9_Haar_Brougham_LEW_final.pdf. 42. Bandura A, Locke EA: Negative self-efficacy and goal effects revisited.
17. Luthans F, Avolio BJ, Avey JB, Norman SM: Positive psychological capital: J Appl Psychol 2003, 88:87–99.
measurement and relationship with performance and satisfaction. 43. Snyder CR: Hope, goal-blocking thoughts, and test-related anxieties.
Pers Psychol 2007, 60:541–572. Psychol Rep 1999, 84:206–208.
18. Zhong LF: Effects of psychological capital on employees’ job 44. Tugade MM, Fredrickson BL: Resilient individuals use positive emotions to
performance, organizational commitment, and organizational citizenship bounce back from negative emotional experiences. J Pers Soc Psychol
behavior. Acta Psychologica Sinica 2007, 39:328–334 (in Chinese). 2004, 86:320–333.
19. Avey JB, Luthans F, Smith RM, Palmer NF: Impact of positive psychological 45. Luthans F: The need for and meaning of positive organizational
capital on employee well-being over time. J Occup Health Psychol 2010, behavior. J Organ Behav 2002, 23:695–706.
15:17–28. 46. Totterdell P, Wood S, Wall T: An intra-individual test of the demands-
20. Avey JB, Wernsing TS, Luthans F: Can positive employees help positive control model: a weekly diary study of psychological strain in portfolio
organizational change? Impact of psychological capital and emotions on workers. J Occup Organ Psychol 2006, 79:63–84.
relevant attitudes and behaviors. J Appl Behav Sci 2008, 44:48–70. 47. Guan NH, Zhang JP, Wang XD, Lin HS, Pan J, Gan ZY, Zhang ZL, Wu XL, Wei
21. Avey JB, Luthans F, Jensen SM: Psychological capital: a positive resource QL: A study of psychological state of prison policemen of guangdong
for combating employee stress and turnover. Hum Resour Manage 2009, province. Chinese J Behav Med Sci 2005, 14:260–262 (in Chinese).
48:677–693. 48. Zhong LF: Review of psychological capital research. Adv Psychol Sci 2007,
22. Luo H, Hao ZH: The relationships among psychological capital, job 15:482–487 (in Chinese).
burnout and turnover intention in 466 nurses. Chin J Nurs 2010, 49. Luthans F, Avey JB, Avolio BJ, Norman SM, Combs GM: Psychological
45:933–935 (in Chinese). capital development: toward a micro-intervention. J Organ Behav 2006,
23. Luthans F, Norman SM, Avolio BJ, Avey JB: The mediating role of 27:387–393.
psychological capital in the supportive organizational climate-employee 50. Luthans F, Avey JB, Patera JL: Experimental analysis of a web-based
performance relationship. J Organ Behav 2008, 29:219–238. training intervention to develop positive psychological capital. AMLE
24. Tian XZ, Xie JY: The influence of POS on working behaviors of 2008, 7:209–221.
employees: Empirical research on mediating role of psychological 51. Luthans F, Avey JB, Clapp-Smith R, Li WX: More evidence on the value of
capital. Nankai Business Review 2010, 13:23–29 (in Chinese). Chinese workers’ psychological capital: a potentially unlimited
25. Luthans F, Avolio BJ, Walumbwa FO, Li WX: The psychological capital of competitive resource? Int J Hum Resource Manag 2008, 19:818–827.
Chinese workers: exploring the relationship with performance. Manag 52. Avey JB, Patera JL, West BJ: Positive psychological capital: a new lens to
Organ Rev 2005, 1:247–269. view absenteeism. JLOS 2006, 13:42–60.
Liu et al. BMC Psychiatry 2013, 13:89 Page 9 of 9
http://www.biomedcentral.com/1471-244X/13/89

53. Liu K, You LM, Chen SX, Liu HP, Li XM, Li XH, He GP, Shang SM, Hu Y, Jiang
XL, Zheng J, Zhu XW, Zhang LF, Bu XQ: Organizational support and its
related factors among Chinese hospital nurses. Chin J Nurs 2011,
46:1009–1011 (in Chinese).
54. Wang Y, Ma L, Meng LY, Wang L: Perceived organizational support and
job performance among police officers in prison: the mediating role of
organizational commitment. J China Medl Univ 2012, 41:172–174
(in Chinese).
55. Luthans F, Avey JB, Avolio BJ, Peterson SJ: The development and resulting
performance impact of positive psychological capital. Hum Resource Dev
Q 2010, 21:41–67.
56. Luthans F, Vogelgesang GR, Lester PB: Developing the psychological
capital of resiliency. Hum Resource Dev Rev 2006, 5:25–44.

doi:10.1186/1471-244X-13-89
Cite this article as: Liu et al.: Positive resources for combating
depressive symptoms among Chinese male correctional officers:
perceived organizational support and psychological capital. BMC
Psychiatry 2013 13:89.

Submit your next manuscript to BioMed Central

and take full advantage of:

• Convenient online submission

• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution

Submit your manuscript at

www.biomedcentral.com/submit