iBusiness, 2010, 2, 268-273
doi:10.4236/ib.2010.23034 Published Online September 2010 (http://www.SciRP.org/journal/ib)
The Moderation Effects of Perceptions of
Organizational Politics on the Relationship
between Work Stress and Turnover Intention: An
Empirical Study about Civilian in Skeleton
Government of China*
Guangjin Zhang1, Gabriel Lee2
1Politics and Law School, China University of Geosciences (CUG), Wuhan, China; 2College of Business, City University of Hong
Kong (CityU), Hong Kong, China.
Received May 26th, 2010; revised July 5th, 2010; accepted August 19th, 2010.
ABSTRACT
It is disastrous for organization that valued employee leave their jobs. Many researches have found that work stress is
consistently and strongly related to intentions to turnover. Therefore, studying buffers of the stress-turnover intention
relationship is important. This paper examines the situational variables such as perceptions of organizational politics,
as moderators of the stress-turnover relationship. An investigation from 304 employees in Chinese skeleton government
found that: lower levels of perceived politics will minimize turnover intentions resulted from work stress. These findings
are helpful to understand the unique relationships between these variables and provide specific information for manag-
ers to retain valued employees.
Keywords: Perceptions of Organizational Politics, Work Stress, Turnover Intention, Moderation Effects
1. Introduction
Employees are recognized as a very important organ-
izational asset, Organizational cost incurred due to em-
ployees quitting their jobs and the subsequent hiring of
replacement personnel, new-hire training, and general
costs for administration can be tremendous in terms of
personal, work unit and organizational re-adjustments
[1]. Thus, voluntary employee turnover is an important
organizational issue that merits thorough exploration.
Although it is recognized that people leave their jobs
for a variety of reasons, there is consistent research evi-
dence indicating that voluntary turnover can be explained
by employees’ intention to leave the organization. In
investigations of turnover, actual voluntary turnover
measures are extremely difficult to obtain; therefore, in-
tentions to turnover are used in the high majority of
turnover studies, as these intentions are the only antece-
dent that has been consistently found to be a direct ante-
cedent to actual turnover. Like those prior works, this
study explores voluntary turnover by focusing on the
potential impact of employees’ intentions to leave the
organization.
The prediction and understanding of employee turn-
over has been studied from many different perspectives.
Most of the past studies have focused on single direct
antecedents and examined their roles in the context of
process models of turnover. Work stress has been identi-
fied as major contributors to voluntary turnover and has
ultimately been blamed for the loss of employees. On the
other hand, many models acknowledge that negative af-
fective reactions (such as work stress) do not necessarily
cause individuals to actually quit and have attempted to
improve their predictive ability by incorporating factors
outside the individual. Consequently, researchers have
attempted to better understand these variations by con-
*This research was supported by the Primary Research Foundation o
f
University belonging to Education Ministry of china under Grant
CUGW090211, the Research Project of S&T of Hubei province under
Grant B20080402, and the Humanity Social Science Foundation o
f
Hubei province under Grant 2008y054.

The Moderation Effects of Perceptions of Organizational Politics on the Relationship between
Work Stress and Turnover Intention: An Empirical Study about Civilian in Skeleton Government of China
269
sidering the effects of external variables such as job al-
ternatives, social support, and the quality of leader mem-
ber exchange (LMX) relationships.
Despite these variations, the stress-turnover relation-
ship is well established and can have negative bottom-
line consequences for an organization [2]. Thus, it re-
mains important to investigate how other variables in-
fluence this relationship. Specifically, other variables
often provide additional insight or clarification into the
exact nature of how the variables interact. The intention
of this study is to examine situational variables that
might affect the stress-turnover intention relationship. In
particular we are interested in understanding variables
that have the potential to buffer or intensify the impact of
stress on employees’ intentions to leave the organization.
The variable of interest in the study is perceptions of
politics (POPs). We expect that by studying buffers of
stress impact on turnover intentions, we can gain a better
understanding of internal organizational dynamics that
affect individuals and organizations as a whole. Thus, the
purpose of this study is to investigate how POPs moder-
ate the relationship between work stress and turnover
intentions.
2. Literature Review and Hypothesis
2.1. Work Stress-Turnover Relationship
The stress-turnover relationship has received consider-
able attention and has been empirically supported in dif-
ferent contexts, worker types, and cultures [3]. Past stud-
ies suggest that one factor related to turnover is tension
associated with the employees’ present work, This con-
ceptualization is in keeping with the prior established
notion that increasing levels of tension in the current job
may lead to a decision to quit the stressful work envi-
ronment. Additionally, Cao Yu-ping used causal model-
ing to find support for the notion that stress is a predictor
of turnover [4]. This has led to a general acceptance that
the relationship exists and has likely led to a decrease in
the number of studies on this topic.
As suggested by the cybernetic theory of stress, it is
important to note that not all individuals who experience
workplace stressors have the resulting stress and not all
people who experience stress have the adverse resulting
outcomes (i.e., turnover intentions) [5]. Some studies
have investigated gender, age, tenure, and other person-
ality variables which may influence the stress-turnover
relationship, but a number of variables have been ne-
glected. In particular, we feel that it is necessary to ex-
amine not only variables related to the person, but also
aspects of the organization. Up to this point, there are no
known studies that have explicitly examined situational
factors unique to the social milieu of the organization
such as organizational climate, perceived organizational
support, perceived politics, and organizational justice
perceptions to name a few.
2.2. Perceptions of Organizational Politics
Organizational politics is defined as “social influence
attempts directed at those who can provide rewards that
will help promote or protect the self-interest of the actor”.
Although we recognize that politics can have some con-
structive outcomes in organizations, it is generally (and
in the large majority of research studies) recognized that
political behavior deals with attempts to benefit, protect
or enhance ones self-interest without regard for the wel-
fare of the organization or others. Thus, the theory in this
article is based on Mintzberg’s definition that describes
political behavior as individual or group behavior that is
informal, ostensibly parochial, typically divisive and
illegitimate [6]. In keeping with this definition, our
measure of perceptions of politics really does not address
positive aspects of this construct. Additionally, since it is
difficult to observe or measure politics or political be-
haviors, individuals can report their perceptions of such
activities (i.e., POPs).
In the original conception, Ferris et al. suggested that
POPs led directly to job anxiety/stress and to different
forms of organizational withdrawal, including turnover.
In a revised model of POPs, Ferris et al. expanded the
model on POPs, but still retained the linkages between
POPs and stress and POPs and turnover. These relation-
ships have been examined and empirically supported in
numerous investigations [7].
Whereas POPs have been examined as a direct ante-
cedent to these outcomes, it has received minimal atten-
tion as a potential moderator. The only study we are
aware of investigating POPs as a moderator examined
how it influenced the conscientiousness job performance
relationship [8]. This leads to a natural question of what
other relationships POPs might influence as a moderating
variable. This question is especially relevant to the POPs-
stress-related relationships since POPs has only been
considered as a direct antecedent in such relationships.
Moreover, the direct relationship between POPs and
stress has generally been substantiated [9]. This study
expands the literature by considering and examining an
alternative pattern of the POPs-stress relationship.
2.3. POPs as a Moderator of the Stress-Turnover
Intention Relationship
Although it has been established that POPs are inde-
pendently related to stress and lead to higher self-reports
of turnover intentions [10], the interactive effect of work

The Moderation Effects of Perceptions of Organizational Politics on the Relationship between
Work Stress and Turnover Intention: An Empirical Study about Civilian in Skeleton Government of China
270
stress and POPs has never been examined. This is sur-
prising due to the fact that their combined influence is
likely to provide greater insight into the specific predic-
tion of turnover intentions.
Previous studies have examined the link between an
individual perceiving the workplace as political and vari-
ous outcomes, including intentions to turnover. Results
indicate that there is a positive relationship between po-
litical perceptions and individuals’ intentions to depart
from an organization. The reason for this is that POPs is
considered as a disconcerting phenomenon that decrease
an individual’s desire to remain in a political workplace
[11].
In some cases organizational politics are perceived as
threatening. When individuals experience stress and the
resulting feelings of wanting to leave the organization (i.e.,
turnover intentions), political situations are likely to in-
tensify these feelings. Due to the nature of organizational
politics, people are also less likely to feel that they have
reliable external coping mechanisms. Once politics is per-
ceived, individuals tend to view the workplace as a self-
focused environment characterized by self-serving be-
havior [10]. It is argued that this kind of self-reliance and
forced individualism diminishes peoples’ ability to deal
with workplace stressors and the resulting stress, and in-
creases their desire to escape from undesirable situations
[12]. Thus, the following hypothesis is proposed:
Hypothesis: Perceptions of organizational politics will
moderate the relationship between stress and turnover
intentions, specifically the positive relationship between
stress and turnover intention will be strongest when per-
ceptions of politics are highest.
3. Conclusions
3.1. Samples
The respondents were 304 civilian (40% response rate)
from skeleton government in Hunan province of china.
The sample was primarily male (57%). The years of edu-
cation ranged from 15 to 22 with a mean of 17 years. Ten-
ure with the organization was measured categorically to
protect respondents’ anonymity, the departments divided
into two types, the one is outer oriented that provide ser-
vice for the public; the other is internal oriented managing
internal personnel and hardware construction. The demo-
graphics for the sample have some of representative.
3.2. Measures
All of the items on the survey were responded to on a five-
point Likert scale. The anchors for the scale were strongly
disagree (1) to strongly agree (5). The items in the scales
for each sample were averaged to create an overall mean
for each variable. The items were coded such that high
values represented high levels of the constructs.
Perceptions of organizational politics were measured
with fifteen items (a = .87) developed by Kacmar and
Carlson [13]. A sample item from this scale was “People
in this organization attempt to build them up by tearing
others down”.
Job stress was measured with the seven-item scale (α
= .89) from Rizzo, House and Lirtzman [14]. A sample
item included “I work under a great deal of tension”.
Turnover intention was measured with three items (α
= .84) from Seashore, Lawler, Mirvis and Cammann [15].
A sample item from this scale was “I will probably look
for a new job in the near future”.
3.3. Control Variables
Four demographic variables were controlled in this study.
The variables were tenure, gender, education background
and department. The first three control variables were
included because previous research has shown the strong
impact these variables can have on turnover intentions
[1]. Department was included to eliminate any differ-
ences that might exist between individuals in different
departments of the organization.
4. Data Analysis and Results
We used hierarchical moderated regression analysis
(HMRA) to test our hypotheses. There were four steps in
the HMRA. In the first step, control variables were en-
tered. Stress was entered in the second step, and the
moderating variable (POPs) was entered in the third step.
In the final step, the interaction term was entered.
Table 1 presents means, standard deviations, and in-
ter-correlations for the variables. As can be seen in Table
1, it provided further evidence for the study, the correla-
tion between stress and turnover intentions was signifi-
cant (r = .382, ρ < .01). Additionally, the modera-
tors-perceptions of politics significantly related to turn-
over intentions (r = .430, p < .01).
Since all of the data used in this study were collected
from the same individuals, at the same time, and using
the same method, common method variance (CMV)
posed a potential threat to the results of this study. Before
proceeding with further analyses, we tested for CMV
using the Harmon one-factor test [16], where all of the
scales in the study were submitted to an exploratory fac-
tor analysis (EFA). CMV exists when there is only one
factor or the first factor explains a majority of the vari-
ance. The results of this EFA using principal components
analysis produced multiple factors with Eigenvalues
equal to or greater than 1.00, and the first factor ex-
plained only 29% of the variance. Thus, the Harmon one-
factor test indicates that CMV was not a serious problem
in this study.

The Moderation Effects of Perceptions of Organizational Politics on the Relationship between
Work Stress and Turnover Intention: An Empirical Study about Civilian in Skeleton Government of China
271
Table 1. Means, standard deviations, and inter-correlations.
Mean S.D. 1 2 3 4 5 6 7
1. Tenure 2.84 1.29
2. Gender 1.44 0.5 −.050
3. Education 16.8 5.14 −.323** −.212**
4. Department 1.16 0.41 −.031 .135** .010
5. Work stress 2.91 0.91 .137** −.061* −.054* −.013
6. POPs 3.20 0.70 .106** .048 −.061* .133** .451**
7. Turnover intention 2.39 1.10 −.140** −.156** −.170** .036 .382** .430**
*P < 0.05; ** < 0.01
Table 2. Hierarchical moderated regression analysis predicting turnover intention.
Dependent variable Turnover Intention
Independent variable Model 1Model 2 Model 3 Model 4 VIF
>= 3 and =< 10 −.25** −.12** −.11** −.11** 1.72
Tenure
(years) >10 −.27** −.14** −.14** −.12** 2.13
Gender female −.21** −.15** −.09 −.08 1.47
>= 25 and =< 35 .16** .13** .07 .05 1.46
Education
(years) > 35 .23** .15** .11** .09 1.15
Outer oriented .04 .01 .02 .01 1.31
The first step
Department Internal oriented .07 .01 .05 .03 1.18
The second step Work stress .18** .19** .19** 1.62
The third step POPs .13** .14** 2.05
The fourth step Work stress * POPs .12** 2.16
Adj.R2 0.09 .238 .332 .364
△R2 0.09 .148 .094 .032
F 9.40** 20.01** 23.52** 27.98**
D-W
1.96
**<0.01
Table 2 provides the results from the HMRA. Results
from the analyses showed that the control variable de-
partment was not significantly related to turnover inten-
tions, whereas tenure, gender, and education were all
related, specifically, tenure and female is negatively re-
lated to turnover intention, and education is positively
related to it. Which maybe resulted form that internal
governments lack for competition compared with outer
society, the civilian who has more tenure within govern-
ments would have less courage to enter into market sys-
tem accompanied with competition, most of the female
also like stable and middle-high reward offices position,
In general, the more education, the more capability, the
more chance to acquire a new job, therefore, the higher
turnover intention. The control variables explained 9%
variance. In the second step, work stress was positively
and significantly related to turnover intentions, and ex-
plained additional 14.8% of variance beyond the control
variables. In step three, the POPs was significantly re-
lated to turnover intentions, and POPs explained 9.4%
unexplained variance. Finally, in the fourth step, we en-
tered the interaction terms and found support for hy-
pothesis as it was positively and significantly (β = .12, p
< .01) related to turnover intentions, further, the interac-
tion term explained an additional 3.2% of the variance
beyond the first three steps. The final column of Table 2
showed the VIF and the Durbin-Watson in Model 4, the
biggest VIF is the interaction term (2.16), which indi-
cated that the predicators haven’t serious multiple col-
linearity, the D-W is 1.96(  2), it indicated that the error
items in model 4 are independent.
A graphical representation of this relationship is pro-
vided in Figure 1 and illustrates how individuals with
high stress and high POPs exhibit the strongest turnover
intention. Of particular interest are the low levels of
turnover intention when POPs are low, regardless of

The Moderation Effects of Perceptions of Organizational Politics on the Relationship between
Work Stress and Turnover Intention: An Empirical Study about Civilian in Skeleton Government of China
272
Work stress * POPs
4.2
4.4
4.6
4.8
5
5.2
5.4
5.6
Low High
Levels of work strtess
Levels of intention
To turnover
High
POPs
Low
POPs
Figure 1. Graphical representation of perceptions of politics moderating the work stress-turnover intention relationship.
level of work stress, but the same does not hold when
POPs are high. Specifically, when stress is low and POPs
are high, turnover intention is moderate, but when both
variables are high the resulting turnover intention is con-
siderably higher than for any other combination of the
two variables.
5. Conclusions and Discussions
5.1. Findings
The results from this study provide support for the
study's hypothesis. That is POPs moderates the relation-
ship between work stress and turnover intentions. Al-
though individuals may experience work stress, which
could lead to turnover intentions, this positive relation-
ship is strongest when political perceptions are the high-
est (see Figure 1). Thus, when POPs are at low levels,
turnover intentions are considerably lower. The finding
points that the importance of organizations promoting
desired behaviors and discouraging self interested be-
haviors that epitomized politics [9].
5.2. Implications for Management Practice
Organizations are inherently stressful places. There are a
number of different workplace stressors that ultimately
produce stress for employees. If we are to acknowledge
that workplaces do and will continue producing stress, it
is necessary for organizations and their managers to find
ways to reduce the negative consequences from this felt
stress. The findings from this study found that this can be
accomplished in this way. That is, managers can attempt
to reduce perceptions of politics. Politics have been iden-
tified as a source of stress themselves [9], but it may be
that if managers can decrease political perceptions, they
can decrease the negative feelings and behavioral inten-
tions from work stress. Some ways to decrease political
perceptions are to make rules and procedures, ensure that
they are understood, and then to enforce them [12].
Clearly establishing and enforcing rules for appropriate
behavior leaves little room for alternative interpretations.
Managers need to consistently reward desired behaviors
that follow the guidelines and ignore and/or punish be-
havior that does not. If good things only come to those
individuals who follow the rules, employees will quickly
learn and enact behaviors that will reap rewards, and in
this process, lower individuals' levels of perceptions of
politics.
REFERENCES
[1] P. W. Hom and R. W. Griffeth, “Employee Turnover,”
South/Western Ohio, Cincinnati, 1995.
[2] S. Parasuraman, “Predicting Turnover Intentions and
Turnover Behavior - A Multivariate Analysis,” Journal of
Vocational Behavior, Vol. 21, No. 1, 1982, pp. 111-121.
[3] X. P. Zhao, L. Liu and C. Z. Zhang, “A Multi-Variables
Analysis on Factors Influencing Employee’s Turnover
Intention,” China Soft Science, Vol. 18, No. 3, 2003, pp.
71-73.
[4] Y. P. Cao, “A Review of Occupational Stress, Burnout
and Departure Intention among University Faculty,”
Journal of Shanxi Finance and Economics University,
Vol. 27, No. 3, 2005, pp. 41-44.
[5] J. R. Edwards, “A Cybernetic Theory of Stress, Coping,
and Well-Being in Organizations,” Academy of Manage-
ment Review, Vol. 17, No. 2, 1992, pp. 238-274.
[6] H. Mintzberg, “Structure in Fives: Designing Effective
Organizations,” Prentice-Hall, Englewood Cliffs, New
York, 1983.
[7] Z. D. Huang, “A Literature Review about Perception of
Organizational Politics in Foreign Country,” Foreign
Economics and Management, Vol. 26, No. 3, 2004, pp.
15-18.
[8] W. A. Hochwarter, L. A. Witt and K. M. Kacmar, “Per-
ceptions of Organizational Politics as a Moderator of the
Relationship between Conscientiousness and Job Per-
formance,” Journal of Applied Psychology, Vol. 85, No. 3,
2000, pp. 472-478.

The Moderation Effects of Perceptions of Organizational Politics on the Relationship between
Work Stress and Turnover Intention: An Empirical Study about Civilian in Skeleton Government of China
273
[9] K. M. Kacmar, D. P. Bozeman, D. S. Carlson and W. P.
Anthony, “A Partial Test of the Perceptions of Organiza-
tional Politics Model,” Human Relations, Vol. 52, No. 3,
1999, pp. 383-416.
[10] G. R. Ferris, G. Harrell-Cook and J. H. Dulebohn, “Or-
ganizational Politics: The Nature of the Relationship be-
tween Politics Perceptions and Political Behavior,” In: S.
B. Bacharach and E. J. Lawler, Eds., Stamford Research
in the Sociology of Organizations, Vol. 17, JAI Press,
New York, 2000.
[11] R. S. Cropanzano, J. C. Howes, A. A. Grandey and P.
Toth, “The Relationship of Organizational Politics and
Support to Work Behaviors, Attitudes, and Stress,” Jour-
nal of Organizational Behavior, Vol. 18, No. 2, 1997, pp.
159-180.
[12] G. R. Ferris, G. Adams, R. W. Kolodinsky, W. A.
Hochwarter and A. P. Ammeter, “Perceptions of Organ-
izational Politics: Theory and Research Directions,” In: F.
Daiisereau and F. J. Yammarino, Eds., Research in Multi-
Level Issues, Elsevier Science/JAI Press, Oxford, 2002.
[13] K. M. Kacmar and D. S. Carlson, “Further Validation of
the Perceptions of Politics Scale (Pops): A Multiple Sam-
ple Investigation,” Journal of Management, Vol. 23, No.
5, 1997, pp. 627-658.
[14] J. R. Rizzo, R. J. House and S. I. Lirtzman, “Role Con-
flict and Ambiguity in Complex Organizations,” Admin-
istrative Science Quarterly, Vol. 15, No. 2, 1970, pp.
150-163.
[15] S. E. Seashore, E. E. Lawler, P. Mirvis and C. Cammann,
“Observing and Measuring Organizational Change: A
Guide to Field Practice,” Wiley, New York, 1982.
[16] P. M. Podsakoff and D. W. Organ, “Self-Reports in Or-
ganizational Research: Problems and Prospects,” Journal
of Management, Vol. 12, No. 4, 1986, pp. 531-544.