Guanxi with Supervisor and Counterproductive Work Behavior: The Mediating Role of Job

Satisfaction
Author(s): Long Zhang and Yulin Deng
Source: Journal of Business Ethics, Vol. 134, No. 3 (March 2016), pp. 413-427
Published by: Springer
Stable URL: https://www.jstor.org/stable/24703780
Accessed: 29-02-2020 11:21 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Springer is collaborating with JSTOR to digitize, preserve and extend access to Journal of
Business Ethics

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
J Bus Ethics (2016) 134:413-427
DOT 10.1007/sl0551-014-2438-7 CrossMark

Guanxi with Supervisor and Counterpr

The Mediating Role of Job Satisfaction

Long Zhang ' Yulin Deng

Received: 3 March 2014/Accepted: 20 October 2014/Published online: 31 O

© Springer Science+Business Media Dordrecht 2014

Abstract This study aims to explore the rolework

pervasive counterproductive of informal
behavior (CWB) whose
leader-member interactions in managing
impacts are notably negativecounterproduc
in the workplace (Klotz and
tive work behavior (CWB) in a non-Western context.
Buckley 2013). CWB refers generally We
to "a set of distinct
propose that under the Chinesebehaviors
background, guanxi
that share the characteristics with
that they are voli
supervisor increases employees' job
tional (as satisfaction,
opposed which
to accidental or mandated) and harm or
further reduces their CWB. Partial
intend to least square
harm organizations structural
and/or organization stake
equation modeling with a sample
holders." (Spector of 272
et al. 2006, Chinese
p. 447). It has also been
studied under
employees confirms this mediating various specifically
effect of job defined labels such as
satisfac
tion. However, we also find that jobabuse
emotional satisfaction passes
(e.g., Keashly and Harvey 2005), theft the
and
effect of guanxi with supervisor
withdrawalon
(e.g., to CWB
Chen and Spector targeting
1992), production
deviance
people, but not to CWB targeting the(e.g., organization.
Hollinger 1986), and aggression
Impli (e.g.,
Hershcovis
cations for research on CWB and guanxiet al. 2007).
with supervisor
are discussed. Management and organization researchers have made
abundant efforts in explaining how CWB occurs in order to
Keywords China • CWB • Job satisfaction ■ Guanxi prescribe what organizations have to do to reduce it (Klotz
PLS-SEM • Social exchange and Buckley 2013). A recent meta-analysis (Berry et al.
2012) revealed that CWB was most frequently attributed to
individual attributes (e.g., demographics and Big Five) and
Introduction contextual factors (e.g., constraints and organizational
justice). This finding is in line with the extant literature that
Organizations traditionally improve their competitiveness emphasized personality and work stressors as the ante
through boosting effective employee behaviors (e.g., cedents of CWB (e.g., Bolton et al. 2010; Jensen and Patel
organizational citizenship behavior, Farh et al. 1997; pro 2011; Meier and Spector 2013; Spector 2011). Surpris
active behavior, Lam et al. 2014). Western scholars, ingly, interpersonal factors have been largely ignored,
however, have gradually steered their attention to the given the substantial chance that human interactions in the
workplace be proximal causes of CWB (Spector et al.
2006). This research gap has remained until two recent
studies related CWB to leadership behaviors (Kessler et al.
2013) and leadership consideration and structure (Holtz
L. Zhang (0) • Y. Deng and Harold 2013). The current study attempts to expand
Business School, Hohai University, Rm. 1111, Boxue Building,
this stream of research by focusing on guanxi with super
8# West Focheng Road, Nanjing 211100, China
e-mail: zhangl@hhu.edu.cn visor, an informal leader-member relationship in China.
This informal relationship perspective is of much
Y. Deng
School of Business, Nanjing University, Nanjing 210093, China importance. Based on a historical review of CWB, Klotz
e-mail: yulin_d@163.com and Buckley (2013) identified two trends in the workplace

<£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 414 L. Zhang, Y. Deng

that had made compan

informal sources and st
reduce CWB. One is the
as a cause of CWB; the ot
have made CWB more a
development of new for
relationships have not b
recent study (Lin and Ho
increased employees' org
(OCB), which usually ha
CWB (Klotz and Bolino 2013). Guanxi, or "a quality relationship that determines the
Specifically, this paper explores how guanxi with appropriate behaviors and treatment of each oth
supervisor impacts employees' CWB through their job and Tjosvold 2006, p. 1730), is usually consider
satisfaction. We attempt to make several contributions to indigenous Chinese concept (Law et al. 2000).
the literature. First, this study expands the list of interper- by personal interests as well as needs to belong;
sonal causes of CWB and enriches the role of supervisors be built on a relationship by birth or blood (e.g
in the CWB management (Holtz and Harold 2013; Kessler and neighbors) or through social intera
et al. 2013). Rather than highlighting more formal inter- acquaintances; it is maintained and reinforce
personal interactions (e.g., employee selection, MacLane long-term reciprocal exchanges; it can be tra
and Walmsley 2010; leadership behaviors, Kessler et al. through a third party as the referral (Chen and
2013), we focus on the guanxi with supervisor, which is 2007; 2006; Luo 2000).
non-work-centered. As a form of guanxi, guanxi with supervisor refers to a
Second, this paper proposes a perspective of job satis- particularistic relationship between in
faction to explain the effect of guanxi with supervisor on immediate supervisors (Wong et al. 2003).
CWB. Guanxi with supervisor and CWB may have dif- mutual interest and benefits (Wong e
ferent targets: the former targets the supervisor, but the lished and developed mainly through
latter usually targets the organization or persons other than after work (Chen et al. 2009; Luo
the supervisor. Therefore, it is reasonable not to expect an potential of facilitating favor exchanges b
immediate effect of guanxi with supervisor on CWB. We parties involved (Hwang 1987). Despit
use the causal reasoning perspective of CWB (Martinko cultural modernization in China, guan
et al. 2002) to predict that guanxi with supervisor invokes remains the most critical interpersonal re
employees' cognitive processing of job satisfaction, which ferent Chinese settings and has been
further brings CWB that functions as a means to regain the sponding to the ruler-subject relationsh
balance between the contributions the employees have 2010). It has proven to be a key antecede
made and the benefits they have received. Since CWBs attitudes and behaviors of both emplo
differ in their targets, we also relate guanxi with supervisor supervisors (Zhang et al. 2013).
to CWBs with different targets, specifically CWB targeting
the organization (CWB-O) and CWB targeting people Guanxi with Supervisor and Job Satisfa
(CWB-P). Thus, this paper has the potential of revealing
the intricate mechanism about how different types of Job satisfaction refers to the degree t
CWBs occur. are satisfied with their jobs (Kalleberg 1977). According to
Third, conducted in the Chinese context, this study the social exchange theory (Blau
illustrates how CWB occurs in non-Western settings. investments and perceived obligat
Considering many CWBs have roots in negative emotions and the organization they se
(Spector and Fox 2005; Spector et al. 2006), and emotions reciprocity (Gouldner 1960). W
are regulated by culture and social norms (Johns 1997; and effort to their jobs, they anticipa
Keashly and Harvey 2005), studying CWB in a global will recognize their contributions
context is beneficial and indispensable for the full under- The extent to which this anti
standing of these behaviors. The CWB literature, however, individuals' job satisfaction.
has almost exclusively taken a Western perspective, with Guanxi with supervisor prom
very few exceptions (Peng 2012; Rotundo and Xie 2008). between employees and their
Fourth and finally, as an improvement upon prior work expressive and instrumental
that measured CWB by averaging the ratings of its items, spond to the expressive and in

Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 Guanxi and CWB 415

this informal tie (

takes place becau
stable and long-ter
parties' feelings
(Hwang 1987). Em
supervisor increa
devotion to their
1998), all of whi
their supervisors
mechanism, guan
achieve their car
alternative to for
1996). After all, t
(Chen and Chen 2
work issues can b
policies and proced
There has been am
of guanxi with s
increases breadth
Tjosvold 2006), f
Tjosvold 2007), an
2011). It not only
from their super
disposes supervis
strafing their em
Tjosvold 2006; Ch
supervisor even h
(Kossek et al. 201
sonal influence o
that satisfy their
Taken all those a
shapes employees
promising. From
job satisfaction,
proximal causes o
2008). Therefore
between guanxi w
satisfaction. result, they are more likely to perceive disequilibria in the
workplace and dissatisfy with their jobs. To bet
Job Satisfaction and CWB situation, they have to analyze the causes (i.e., attribution
processes) and then take measures accordingly to reduc
According to social exchange theory, job dissatisfaction their job dissatisfaction. Although CWB is not d
implies a perceived imbalance that the employees' contri- the organization or the employees, it can act as a te
butions are not fully compensated by the benefits they solution for employees' negative emotions (
receive from the organization. Although CWB cannot be Fox 2002). Therefore, we propose the
viewed functional for the organization or beneficial for hypothesis:
employees, it provides a temporal solution to regain the
Hypothesis 1 Guanxi with supervisor increases employ
balance through decreasing the employees' contributions
ees' job satisfaction, which further reduces their CWB.
or reducing the benefits of the organization (Spector and
Fox 2002). Scholars make a distinction between CWB targeting
The link between job satisfaction and CWB may be the organization (CWB
strengthened by negative affectivity in social exchanges. A (CWB-P) (e.g., Robinso
recent meta-analysis found an association between job the norm of reciprocity

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 416 L. Zhang, Y. Deng

job dissatisfaction pr
because as aforementi
dissatisfaction arises p
their contributions are
efits they receive from
however, may also le
First, according to th
faction, employees' job
aspects of their jobs (M
related exchanges with
essarily those with the
dissatisfaction. Second,
always clear (Sakurai an
the organization, part
considered representa
1987). Therefore, indi
(e.g., purposely failed
their dissatisfaction

Hypothesis 2 Guanxi w
ees' job satisfaction, wh

Hypothesis 3 Guanxi w
ees' job satisfaction, w
(2006) classi
ing sabotage
Methodology duction deviance (three items), theft (five items), and abuse
(18 items). Sample items included, "Purposely waste
Sample and Data employer's materials/supplies" (sabotage), "Came to work
late without permission" (withdrawal), "Purpose
The sample for this study included 350 employees from six your work incorrectly" (produ
companies in Nanjing, China. As a major city in East something belonging to your employ
China, Nanjing has undergone a dramatic transformation in nasty or rude to a client or custome
the past over thirty years. Numerous organizations in this asked for a rating of how often part
city have adopted Western ideas and techniques, but guanxi in the past 12 months on a f
as a traditional element in Chinese culture is still embraced (1 = never, 2 = once or twice a y
in business management (Zhang et al. 2013). This mixed month, 4 = once or twice a week
condition facilitates examining the effect of guanxi with A prerequisite for testing Hypo
supervisor under transition context. These sampled firms placement of the CWB items int
were in information technology, international trade, and Five graduate students in manage
electronic industries, respectively. At least one of the matter experts (SMEs) to fulfill this
authors got well acquainted with the head of the human them of the definition of CWB and the
resource department or one of the top managers from each CWB-0 and CWB-P, we asked th
company. This connection guaranteed close cooperation which category each item fit. They
from these companies because Chinese people are guanxi- pendently first and then discussed
oriented. After successful contact with the heads of human resolved them. Among the 33 CWB i
resource departments of these companies, one of the placed into CWB-O; another 17 item
authors went to collect data on-site. At the beginning of CWB-P; the remaining item (i.e.,
data collection, this author described the research objec- damaging or harmful rumor at wo
tives and the procedures of data collection to the respon- either category because of its vaguenes
dents, who were assured that all information they provided As Spector et al.'s (2006) CWB C
would be kept confidential. The completed questionnaires scale whose indicators are not inte
were returned directly to the author at the spot. a single underlying construct, we anal

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
Guanxi and CWB

Sal
Tel

Te2
Sa2

Te3
Sa3

Val
Wil

Va2
Wi2

Va3
Wi3

Unl
Pdl

Un2
Pd2

Thl ► // \\\—■— —
Thl Un3

Th2 Th2 Theft Pal

Pal

~' / \\l 71,3 Pa2

Th3 Pa2

Pa3
Pa3
Abl

Ab2 Obi

Ob2

Fig. 1 The HCMs of CWB-0 and CWB-P Model la: The HCM of CWB-0 Model lb: The HCM of CWB-P

formative measurement models of CWB, CWB-O, and indicator led to an increase in the average variance
CWB-0 using PLS-SEM. The results revealed several extracted (AVE) above the recommended threshold value
measurement problems, including multicollinearity, non- (i.e., 0.50) (Hair et al. 2014). The other indicator belonged
significant indicators, and co-occurrence of negative and to the LOC of theft, and deleting this indicator decreased
positive indicator weights. To eliminate these problems, we the number of the theft indicators to 3, thus making it
followed Cenfetelli and Bassellier's (2009) suggestion by closer to the number of indicators with the other LOCs.
establishing hierarchical component models (HCMs) of According to Becker et al. (2012), the equality of the
CWB, CWB-O, and CWB-P. number of indicators per LOC helps to eliminate potential
We first dealt with CWB-O. According to Spector et al. bias when using a repeated indicators approac
(2006), its 15 items fell into five categories (i.e., sabotage we had the HCM of CWB-O consistin
[three items], withdrawal [four items], production deviance indicators (i.e., three for sabotage, three f
[two items], theft [four items], and abuse [two items]), two for production deviance, three for t
which served as the lower order components (LOCs) of abuse; see Fig. 1; Table 3 in Appendix fo
CWB-O. Then we specified a formative-reflective HCM of The PLS-SEM analysis showed that the HCM of
CWB-O (Ringle et al. 2012). Specifically, we specified CWB-O fit the data well. The manifest indicators'loadings
formative relationships between the LOCs and CWB-O ranged between 0.64 and 0.88, with only one below 0.70.
because these LOCs were categories of CWB and thus However, since the criteria for reliability and convergent
were not interchangeable (Coltman et al. 2008). However, validity were both met, we retained this indicator with
we specified reflective relationships between the manifest loadings below 0.70. All the LOCs' composite reliabilities
indicators and the LOCs because each of these LOCs (i.e., 0.78 [sabotage], 0.79 [withdrawal], 0.87 [production
existed independent of their indicators, and the indicators deviance], 0.82 [theft], and 0.80 [abuse]) were well above
belonging to a particular LOC shared a common theme the critical value of 0.70, thus supporting internal consis
(Coltman et al. 2008). tency reliability. The LOCs' AVEs (i.e., 0.54 [sabotage],
We established the HCM of CWB-O using the repeated 0.55 [withdrawal], 0.76 [produ
indicators approach (Lohmoller 1989), whereby we and 0.67 [abuse]) were above 0.50, thereby supporting
assigned all the indicators from the LOCs to CWB-O. We convergent validity. In addition, the correlations between
subsequently removed two indicators with loadings below any two LOCs ranged between 0.42 and 0.65, lower than
the recommended threshold value (i.e., 0.70). One of them the square root of any LOCs AVE. Therefore, the Fornell
belonged to the LOC of withdrawal, and deleting this Larcker criterion was satisfied (Fornell and Larcker 1981),

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 418 L. Zhang, Y. Deng

which established the

LOCs. Finally, the pat
CWB-0 ranged betwee
significant {p < 0.001). ged between 0.26 and 0.65, all of which were lower than
We then dealt with CWB-P. To place the items of CWB- the square root of any LOC's AVE. Theref
P into its sub-categories, the two authors of this paper Larcker criterion was satisfied, and the
engaged in a directed inductive process of developing and validity of the LOCs was established. Fi
refining the coding scheme (Krippendorff 2004). It was coefficient estimates from the LOCs to
decided that these items be classified into five sub-cate- between 0.09 and 0.20, all being highly
gories, including verbal aggression, physical aggression, (p < 0.001).
teasing, undermining, and other behaviors. They then gave
the SMEs the definitions of the five sub-categories and Guanxi with Supervisor
asked them to place the items into these sub-categories.
The SMEs coded the CWB-P items independently first and A six-item scale of guanxi with superv
then discussed the disagreements. They successfully from Law et al. (2000). The items included,
reached agreement (i.e., physical aggression [four items], idays or after office hours, I would call m
verbal aggression [four items], teasing [three items], visit him/her," "I always actively share with
undermining [four items], and other behaviors [two items]). about my thoughts, problems, needs, and f
Similarly, we specified a formative-reflective HCM for supervisor invites me to his/her home for lun
CWB-P, with formative relationships between CWB-P and "On special occasions such as my supervi
its LOCs and reflective relationships between the LOCs would definitely visit my supervisor an
and the manifest indicators. Although all the AVEs of the gifts," "I care about and have a good understa
LOCs reached the threshold value of 0.50, we deleted three supervisor's family conditions," and "W
items with the LOCs of verbal aggression, physical conflicting opinions, I will definitely stand
aggression, and undermining, respectively. As a result, we visor's side." Each item asked for a rating on
came up with the HCM whose LOCs had similar number of Likert scale (1 "strongly disagree" to 5 "
manifest indicators (i.e., verbal aggression [three items], The scale score was the average of all th
physical aggression [three items], undermining [three Cronbach's a for this scale in the current s
items], teasing [three items], and other behaviors [two indicating good reliability,
items]; see Fig. 1; Table 3 Appendix for detail). The PLS
SEM analysis showed that the HCM of CWB-P fit the data Job Satisfaction
well. The manifest indicators' loadings ranged between
0.66 and 0.88, with only two below 0.70. The AVEs of the We used Minnesota Satisfaction Questionn
LOCs were 0.67 (teasing), 0.58 (verbal aggression), 0.67 to measure job satisfaction. The responde
(undermining), 0.65 (physical aggression), and 0.51 (other items on a five-point Likert scale (1 "very dissa
behaviors), all being above 0.50. The composite reliability "very satisfied"). Sample items included
estimates for the LOCs were 0.80 (teasing), 0.82 (verbal work alone on the job," "The chance to do d
aggression), 0.86 (undermining), 0.85 (physical aggres- from time to time," and "The way my boss
sion), and 0.76 (other behaviors), all being above 0.70. In workers." The Cronbach's a for this scale
addition, the correlations between any two LOCs ranged sample was 0.97, indicating excellent reli
between 0.34 and 0.68, lower than the square root of any
LOC's AVE. Therefore, the Fornell-Larcker criterion was Control Variables
satisfied, and the discriminant validity of the LOCs was
established. Finally, the path coefficients from the LOCs to In response to Martinko et al. (2002) emphas
CWB-P ranged between 0.21 and 0.29, all being highly of individual differences and situational
significant (p < 0.001). occurrence of CWB, we included the respondents' demo
Regarding CWB, we also specified a formative-reflec- graphics (i.e., gender, age, educational
five HCM, with the LOCs from CWB-0 and CWB-P as its tenure), compensation, and organization
LOCs. The PLS-SEM analysis showed that this HCM of rule out their possible impacts on C
CWB fit the data well. All the manifest indicators' loadings measured with a dummy variable (1 =
ranged between 0.65 and 0.89, with three being below 0.70. all except two respondents had a colle
Nevertheless, since all the AVEs of the LOCs were above tional level was measured with a d
0.50 (i.e., between 0.51 and 0.76), and all their composite (1 = "graduate"). Compensation wa

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 Guanxi and CWB 419

average monthly sal

RMB," 2 = "2,001-5
RMB," 4 = "8,001-
RMB"). Organizationa
categories: private co
foreign-investment e
two dummy variables
erence type. Team te
worked for their cu
multicollinearity prob
age to logs before hypothesis testing. second stage (Hair et al. 2014).
We assessed the results of each PLS-SEM analysis by
evaluating the reflective measurement models, the formative
Analytic Methodology measurement model(s), and the structural model in sequence
(Hair et al. 2014). The parametric significance tests were
We used PLS-SEM (Lohmoller 1989; Wold 1974) to test performed using the bootstrapping procedure (i.e., 272
our hypotheses for two reasons. First, PLS-SEM facilitates observations for each subsample, 5,000 subsamples). We
the inclusion of formative measurement models (Hair et al. also assessed whether our test results were flawed by com
2014) so that we can simultaneously examine the mea- mon method bias using Harman's single-factor test (Pod
surement model of CWB and the proposed path model. sakoff and Organ 1986) and the unmeasured latent marker
Second, PLS-SEM is deemed advantageous over covari- construct (ULMC) technique (Liang et al. 2007).
ance-based SEM in terms of the robustness of estimations The descriptive and correlation statistics were computed
and statistical power when used with small sample sizes with SPSS 20.0. Harman's single-factor test was run with
(Reinartz et al. 2009). Our sample consisted of 272 SPSS 20.0 and AMOS 20.0. The PLS-SEM analyses were
observations with 60 manifest indicators, resulting in a performed with SmartPLS 2.0 M3 (Beta) (Ringle et al.
subject-to-variable ratio of 4.53 that is lower than the 2005). All reported p-values were two-tailed. Statistical
threshold value of 5 for covariance-based SEM (Hair et al. significance was set at 5 percent.
2010). The data, however, well exceeded the minimum
sample size as required by PLS-SEM (i.e., "10 times the
largest number of formative indicators used to measure one
construct or 10 times the largest number of structural paths Results
directed at a particular construct in the structural model";
Hair et al., 2014, p. 23). Preliminary Analysis
We tested Hypotheses 1-3 by first examining the indi
rect effect of guarvci with supervisor on CWB, CWB-O, Table 1 reports the means, sta
and CWB-P through job satisfaction. According to the tions, and reliability estimates o
recent development of mediation analysis, only one means of guanxi with supervisor
requirement is needed to establish mediation that the faction (i.e., 3.50) were above 3,
indirect effect of the predictor on the criterion through the CWB, CWB-O, and CWB-P
mediator being significant (MacKinnon 2008; Zhao et al. particular interest to this discu
2010). We also made a more comprehensive examination correlations among guarvci with
of the mediation effects using the Tippins and Sohi (2003) tion, and CWB. Guanxi with
criteria. Tippins and Sohi included Baron and Kenny's correlated with job satisfaction
(1986) conditions for mediation effects but were more negatively correlated with CW
applicable to SEM because they involved the comparison CWB-P (r = -0.15, p < 0.05
of two competing models: the direct model and the medi- p < 0.05). Job satisfaction was
ated model. Specifically, Tippins and Sohi specified four CWB-P (r = -0.16, p < 0.
conditions for mediation to occur: (a) the mediated model p < 0.05) and marginally cor
explains more variance in the criterion than does the direct -0.11, p < 0.10). In addition
model; (b) the predictor relates to the criterion in the direct between predicting variables was
model; (c) the predictor relates to the mediator, and the indicators of organizational ow
mediator relates significantly to the criterion in the medi- below the recommended threshold
ated model; and (d) the significant relationship between the concern of the collinearity proble

£) Spring*:er

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 420 L. Zhang, Y. Deng

Table
Table
1 Means, standard deviations, 1
correlations,
Means,
and reliability estimates of the study standard
variables devia
\/f™~ or* 1 1 1 A < ti 1
Variables Mean
Mean SD
SD 1 2 3 4 5 6 7 8 8 99 10 11 12

Gender
Gender 0.66 0.47

Age 31.10 3.16 0.04

Educational 0.20 0.40 0.04 0.12

level

Team tenure 2.60 1.77 -0.01 0.62" 0.01

Compensation 3.32 0.92 -0.09 -0.01 0.07 0.02

Organizational 0.19 0.39 -0.25" -0.15* -0.01 -0.06 0.00

ownership 1
* •

Organizational 0.59 0.49 0.22" i

© 2 -0.16**
0.12 0.16" -0.58**
ownership 2
Guanxi with 3.09 0.87 -0.10 0.05 -0.07 0.26** 0.52** -0.03 0.10 P bo
(0.87)b
supervisor3
Job 3.50 0.73 -0.07 0.01 0.10 0.01 0.62** -0.01 0.08 0.57**
0.57** (0.97)b
(0.97)b
satisfaction"
*
CWB-O" 1.52 0.39 0.04 -0.05 0.10 0.00 -0.11 0.03 -0.03 -0.12*
1

© <N
1
©
—0.11
CWB-P"
CWB-P"
1.24 0.32 0.09 0.00 0.08 0.07 -0.14* -0.06 0.09 -0.15* -0.16"
-0.15* —0.16*"* 0.73**
»

-0.13* -0.14*
*

CWB"
CWB" 1.37 0.33 0.07 -0.03 0.10 0.03 1

© -0.01 0.03 i
© -0.14*
cn 0.94" 0.92**

* pp< <0,05,
0.05, p <<0.01;
** p 0.01;n =n 272
= 272
"" The
Thescale
scale
score
score
was was
calculated
calculated
by taking
by the
taking
meanthe
of the
mean
itemofratings
the item
included
ratings
in theincluded
scale in the scale
bb The
Thenumbers
numbersin parentheses
in parentheses
are Cronbach's
are Cronbach's
a's. We dida's.
notWe
report
did the
notCronbach's
report thea's for
Cronbach's
CWB, CWB-O,
a's for
and CWB,
CWB-P CWB-O,
because these
andconstructs
CWB-P because these constructs
were
weremeasured
measuredwith
with
formative
formative
indicators
indicators
for which
for
internal
which consistency
internal reliability
consistencyare not
reliability
meaningful
are not meaningful

Hypothesis Testing As anticipated, the indirect effect of guanxi with supervisor

on CWB through job satisfaction was significant (—0.11,
Hypothesis 1 proposed that job satisfaction mediates t = -2.30, p < 0.05).
between guanxi with supervisor and CWB. To test this The four conditions for mediation specified by
hypothesis, we ran PLS-SEM analyses that included the and Sohi (2003) were also met. Figure 2 shows t
hypothesized relationships and the measurement models of peting models analysis. Model 2a examined th
guanxi with supervisor, job satisfaction, and CWB. All the relationship between guanxi with supervisor a
control variables (i.e., gender, age, educational level, team Model 2b examined the same relationship with
tenure, compensation, and organizational ownership) were faction acting as the mediator. Tippins and Soh
specified as predictors of CWB and job satisfaction. first requirement was satisfied since the mediated
The first-stage PLS-SEM analysis showed that the accounted for more variance in CWB than did the
measurement models of guanxi with supervisor and job model (R2 = 0.18 vs. 0.14; see Fig. 2). The secon
satisfaction fit the data well. Specifically, the AVE of gu- dition was also met because guanxi with superv
anxi with supervisor was 0.61, and the factor loadings directly and negatively related to CWB {fi
ranged between 0.66 and 0.84. The AVE of job satisfaction p < 0.01; see Fig. 2, Model 2a). As to the third
was 0.60, and the factor loadings ranged between 0.69 and ment, guanxi with supervisor was related dire
0.87. The HCM of CWB also fit the data well. As the positively to job satisfaction (/? = 0.30, p <
measurement statistics were highly similar to those repor- Fig. 2, Model 2b), and job satisfaction was related dir
ted in the Measures section, we did not repeat them for and negatively to CWB (/? = -0.36, p < 0.05; se
brevity. The collinearity analysis revealed that the maxi- Model 2b). The fourth criterion was satisfied because w
mum VIF was 3.40 (for physical aggression with CWB as job satisfaction was included as the mediator, th
the criterion), well below the conservative threshold of 4 effect of guanxi with supervisor on CWB becam
for a serious multicollinearity problem (O'Brien 2007). nificant (/? = -0.37, p < 0.01 in Model 2a vs. /? =
We obtained the latent variable scores for the LOCs of p > 0.10 in Model 2b; see Fig. 2). These results p
CWB in the first stage and used them as manifest indicators support for Hypothesis 1.
of CWB in the second stage (Hair et al. 2014). Table 2 Hypotheses 2 proposed that job satisfaction
reports the results of the second-stage PLS-SEM analysis. between guanxi with supervisor and CWB-O; Hyp

<£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 Guanxi and CWB 421

Table
Table 2 Direct,
2 Direct, indirect, indirect, and
and total effects total effects p > 0.10). Table 2 also shows evidence consistent with
Iffa^
EffectDirect effect
on Direct indirect
effect Hypothesis
Indirect Total 3 that the indirect effect of guanxi with
mediator/
mediator/effect effect
effect supervisor
effect on CWB-P through job satisfaction was signif
criterion icant (-0.09, t = 2.04, p < 0.05).
criterion

The mediated
mediatedmodel
modelexplaining CWB
explaining CWB
We also examined Hypotheses 2
The
Job
Job satisfaction
satisfaction Tippins and Sohi's (2003) criteria.
CR2 = 0.62)
(J?2 competing models analysis. It is cle
Guanxi with 0.30" (7.47)
(7.47) 0.30
0.30 was rejected because job satisfacti
supervisor CWB-0 (-0.21, p > 0.10; see Fig. 3, Model 3b).
% =
CWB (R2 ii
0.18)
© oo
i—»
Hypothesis 3, however, received supported because all the
Guanxi
Guanxiwith
with -0.09 (0.56)-0.11*
(0.56) -0.11*(-2.30)
(-2.30) -0.20 Tippins and Sohi (2003) criteria were met. Specifically, the
-0.20
supervisor first requirement was satisfied since the mediated model
Job satisfaction -0.36" (2.58)
(2.58) -0.36
-0.36 accounted for more variance in CWB-P than the direct
The
Themediated CWB-0
model explaining
mediated and CWB-P
CWB-O
model and CWB-P model (R2
explaining = 0.12 vs. 0.09; see Fig. 3). The second con
Job satisfaction
Job satisfaction(J?2
(R20.60)
== dition was met because guanxi with supervisor was directly
Guanxi
Guanxi with
with 0.29" (6.98)
(6.98) 0.29
0.29 and negatively related to CWB-P (jS = -0.30, p < 0.05;
supervisor
see Fig. 3, Model 3a). As to the third requirement, guanxi
CWB-0
CWB-O (R2 = 0.10) with supervisor was related directly and positively to job
Guanxi with -0.14 (1.50)
(1.50) -0.06
-0.06 (1.42)
(1.42) -0.20
-0.20 satisfaction (/? = 0.29, p < 0.01; see Fig. 3, Model 3b),
supervisor
and job satisfaction was related directly and negatively to
Job satisfaction -0.21 (1.09)
(1.09) -0.21
-0.21
CWB-P (/? = -0.30, p < 0.05; see Fig. 3, Model 3b).
CWB-P
CWB-P(R2 = 0.12) (R2 = 0.12)
Support for the fourth criterion was obtained because when
Guanxi
Guanxiwith with
-0.07 (0.47) -0.09*
(0.47) -0.09*(2.04)
(2.04)-0.16
-0.16 job satisfaction was included as the mediator, the
supervisor
direct effect of guanxi with supervisor on CWB-P became
Job satisfaction -0.30* (2.20)
(2.20) -0.30
-0.30
nonsignificant (/? = -0.30, p < 0.05 in Model 3a vs.
*p
p < <0.05, "P< 0.01
0.05, p < p =0.01
-0.07, p > 0.10 in Model 3b; see Fig. 3).
The
Thenumbers
numbers
in parentheses are r-values
in parentheses
for the direct/indirect effects
are f-values for the direct/indir

Significance
Significancetests were based on
tests
Student's two-tailed
were t test.
basedThe on Student's two-tailed t
indirect
indirecteffects and f-values were determined
effects using the bootstrapping
and f-values wereCommon Method Bias
determined using the boo
procedure
procedure(272 observations
(272
for eachobservations
subsample, 5,000 subsamples) for each subsample, 5,000 su

Because all the data were self-reported and collected

through the same questionnaire during the same period
proposed that job satisfaction mediates between guanxi with with cross-sectional design, we checked for com
supervisor and CWB-P. We examined these two hypotheses method variance. We first conducted a Harman's s
by following the same procedure as used in the test of factor test (Podsakoff & Organ, 1986). All the man
Hypothesis 1. In the first stage, we ran a PLS-SEM analysis indicators of guanxi with supervisor, job satisfaction, an
that included the hypothesized relationships and the mea- CWB were entered into an exploratory factor ana
surement models of guanxi with supervisor, job satisfaction, using the unrotated principal components method. The
CWB-O, and CWB-P. All the control variables were spec- results revealed that no single factor accounted for
ified as predictors of job satisfaction, CWB-O, and CWB-P. majority of the variance, and the largest factor
The results showed that all the measurement models fit the accounted for 25.07 % of the variance. Confirmatory fa
data well. We did not repeat the measurement statistics analysis also showed that the single-factor model did not fi
because they were highly similar to those reported before. the data well (x2 = 8328.56, d/"= 1,377, x2/d/ = 6
The collinearity analysis revealed that the maximum VIF RMSEA = 0.14, CFI = 0.39, TLI = 0.36, GFI = 0.33).
was 3.02 (for verbal aggression with CWB-P as the crite- Second, following Liang et al. (2007), we included
rion), well below the conservative threshold of 4 for serious the PLS-SEM analysis explaining CWB (see Fig. 2, Model
concern of the collinearity problem. 2b) a ULMC with all the indicators from the prin
The latent variable scores obtained in the first stage for constructs (i.e., guanxi with supervisor, job satisfaction,
the LOCs of CWB-O/CWB-P served as the manifest and CWB) as its indicators. We found that the ave
indicators of CWB-O/CWB-P in the second stage. Table 2 variance of the indicators explained by the principal
shows evidence against Hypothesis 2 that the indirect struct was 0.60, while the average variance of the indic
effect of guanxi with supervisor on CWB-O through job explained by the ULMC was just 0.016. In addition,
satisfaction was not significant (-0.06, t = 1.42, method factor loadings were not significant.

^ Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 422 L. Zhang, Y. Deng

Model
Model
2a: The direct model 2a: The direct mod

Control
Control
variables variables
Gender
Gender
Age
Educational level
Team tenure
Compensation
Organizational ownership

-.37**

Model
Model
2b: The mediated model 2b: The mediated m
Control variables
Gender
Age
Job Educational level
satisfaction Team tenure
J?2 = .62 Compensation
Organizational ownership

^-.36*

CWB
-.09
R1 = .18

Fig. 2 The competing models explaining CWB. * p < 0.05, ** p < 0.01. Significance tests were based on Student's two-tailed t test, and
/-values were determined using the bootstrapping procedure (272 observations for each subsample, 5,000 subsamples)

Similarly, we controlled for the common method vari- Theoretical Implications and Suggestions for Future
ance in the model predicting CWB-0 and CWB-P (see Research
Fig. 3, Model 3b) using the ULMC technique. We found
that the average variance of the indicators explained by This study contributed to theories on CWB and guanxi with
their principal constructs (i.e., guanxi with supervisor, job supervisor in three ways. First, by confirming guanxi with
satisfaction, CWB-O, or CWB-P) was 0.60, and the aver- supervisor as an antecedent to CWB, this study expanded
age method-based variance was just 0.016. In addition, the research that related leadership to CWB from two
most method factor loadings were not significant. aspects. On one hand, this study highlighted the importance
Although the results of the above analyses did not pre- of informal leader-member connections, rather than more
elude common method bias, they did suggest that common formal leadership structure and behaviors (Holtz and Har
method variance was not of high concern and thus unlikely old 2013; Kessler et al. 2013) in CWB management. On the
to confound the interpretations of our results. other hand, this study was among the first to explore the
leadership-CWB relationship in a non-Western culture. As
little has been known about the dynamics between informal
Discussion interpersonal ties and CWB, particularly in a global con
text, we encourage further study along this line
The objective of this study was to investigate how guanxi Second, this paper reve
with supervisor impacted employees' CWB. The data satisfaction between guanxi
showed that employees' job satisfaction passed the effect particular, job satisfactio
of guanxi with supervisor on to overall CWB and CWB-P anxi with supervisor on
but not CWB-O. this nonsignificant finding is that organizations usually

Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 Guanxi and CWB 423

Model 3a: The dire

i i

-.29

Control variables
Gender
Age
Educational level
Team tenure
Compensation
Organizational ownership

-.30
1 '

Model 3b: The mediated model

CWB-0
^ = .10
i i

-.14
-.21

Control variables
Gender
Age
Educational level
Team tenure
Compensation
Organizational ownership
-.30*

CWB-P
^ = .12

Fig. 3 The competing models explaining CWB-0 and CWB-P. * p < 0.05, ** p < 0.01. Significance tests were based on Student's two-tailed t
test, and t-values were determined using the bootstrapping procedure (272 observations for each subsample, 5,000 subsamples)

have high vigilance against CWB-0 (Spector et al. 2006). employees whose job dissatisfaction resulted from this
As a result, organizational factors dominate employees' interpersonal tie are more likely to attribute their job dis
decisions whether to conduct CWB-O, potentially mini- satisfaction to people-related factors than to organizational
mizing the impact of other factors. This account, however, factors. Correspondingly, employees are more likely to
can only partially apply to the current situation because regain balance through conducting CWB-P than through
guanxi with supervisor as a people-centered factor impac- conducting CWB-O.
ted CWB-O even after controlling for employees' demo- The contingent role of job satisfaction between guanxi
graphics (i.e., gender, age, educational level, and team with supervisor and CWB has important implications for
tenure), compensation, and organizational ownership future research. Since guanxi with supervisor had a "main"
(-0.29, t = 2.31; see Fig. 3, Model 3a). A more plausible effect on CWB-O (see Fig. 3, Model 3a), but job satis
explanation relates to employees' attribution processes of faction did not mediate between them (see Fig. 3, Model
job dissatisfaction. This study addressed how employees 3b), further work is needed to explore how guanxi with
handled job dissatisfaction resulting from guanxi with supervisor affects CWB-O and to identify viable mediators
supervisor by conducting CWB. Since guanxi with super- between them. In addition, although job satisfaction was
visor centers on people, rather than the organization, hypothesized to relate more strongly to the organizational

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 424 L. Zhang, Y. Deng

factors than to people

of reciprocity, our dat
finding provided empir
that the intentions be
readily recognized (Sak
future research shoul
reciprocity (Gouldne
A third and final contr
debate on the ethics of
2013). Some scholars ex
linking it to positive a
others criticized it by
(Warren et al. 2004). T
from a new perspecti
supervisor helped to r
behaviors. In addition
studies that put undue
bright side of guanxi w
more balanced view b
supervisor was not equa
types of CWBs. In a se
study (Chen et al. 2011)
as negative effects of g
simultaneously. Given
with supervisor, it will
explore the intricate me
impacts work outcomes. there were alternative measurement models of CWB. To
check the robustness of the test results for Hypothesis 1, we
ran PLS-SEM with different measurement models of CWB,
all other things being equal. Specifically, we examined three
Limitations alternative measurement models of CWB: (1) a first-order
formative model; (2) a formative-formative H
A limitation of this study related to the data that were self- LOCs from CWB-0 and
reported and collected through the same questionnaire ifest indicators as the fo
during the same period with cross-sectional research and (3) a reflective-form
design. Above all, this data collection method could have P as its LOCs and the lat
caused common method bias. Fortunately, Harman's sin- CWB-P in their respect
gle-factor test (Podsakoff and Organ 1986) and the ULMC cators. The results sho
technique (Liang et al. 2007) both suggested that common these measurement mo
method variance was not a real concern for the interpre- Hypotheses 2 and 3 w
tation of the results. Furthermore, the cross-sectional data CWB-0 and CWB-P measu
were not sufficient to establish the proposed causal rela- being equal. Specific
tionships among guanxi with supervisor, job satisfaction, measured with two ty
and CWB. Future research needs to address this issue using order formative mo
a longitudinal design. HCMs with their respective dimensions as the LOCs and
In addition, the self-reported data could have been the manifest indicators as the formative i
flawed due to social expectancy bias. In other words, the LOCs. The results rejected Hypothesi
respondents could have overstated the quality of guanxi Hypothesis 3, consistent with what we h
with supervisor while underreported the frequency of
CWB. This bias may partly explain the relatively low Practical Implications
frequency of the reported CWB in the data. Previous
research, however, revealed that CWB still tended to be In a practical sense, the present findi
low even when it was other-reported (e.g., the boss; Mount building and developing guanxi with

<£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 Guanxi and CWB 425

feasible approach t
panies. However, al
-j. . , . . , __ _ . . the Mmistry of Education of China (11YJC630285) and the Out
effective in reducing both CWB-0 and CWB-P, it impacts standing Innovative Talent ^ of Hohai
CWB-0 and CWB-P through different mechanisms. As descried in this paper by Yulin Deng was
shown in this study, job satisfaction successfully passes the Sciences Foundation of Jiangsu Prov
effect of guanxi with supervisor on to CWB-P but not to National Natural Science Foundation of C
CWB-O. Therefore, managers should take caution to
identify effective "translator(s)" before they apply guanxi
strategy in CWB management. Appendix
Acknowledgements We thank Mr. Can Wang for his help in data See Table 3.
collection and Dr. Deborah Poff and the anonymous reviewers for

Table
Table 3 Counterproductive
3 Counterproductive
work behaviors
work
by subscale
behaviors by subscale

CWB item Sub-category

CWB-0

Purposely damaged a piece of equipment or property Sabotage

Purposely wasted your employer's materials/supplies Sabotage
Purposely dirtied or littered your place of work Sabotage
Stayed home from work and said you were sick when you were not Withdrawal
Taken a longer break than you were allowed to take Withdrawal
Left work earlier than you were allowed to Withdrawal
Purposely did your work incorrectly Production deviance
Purposely worked slowly when things needed to get done Production deviance
Took supplies or tools home without permission Theft
Stolen something belonging to your employer Theft
Put into be paid for more hours than you worked Theft
Told people outside the job what a lousy place you work for Abuse
Been nasty or rude to a client or customer of the company Abuse
CWB-P

Threatened someone at work, but not physically Verbal aggression

Verbally abused someone at work Verbal aggression
Started an argument with someone at work Verbal aggression
Made an obscene gesture (the finger) to someone at work Physical aggression
Hit or pushed someone at work Physical aggression
Threatened someone at work with violence Physical aggression
Insulted or made fun of someone at work Teasing
Made fun of someone's personal life Teasing
Played a mean prank to embarrass someone at work Teasing
Insulted someone about their job performance Undermining
Did something to make someone at work look bad Undermining
Blamed someone at work for error you made Undermining
Purposely failed to follow instructions Other behavior
Stole something belonging to someone at work Other behavior

Definitions of the sub-categories: Sabotage defacing or destroying physical property belo

amount of time working to less than is required by the organization; Production deviance pu
way they are supposed to be performed; Theft take money or physical property belonging to
abuse behaviors against others that harm through making threats or nasty comments; Physic
harm others physically; Teasing the act of harassing others playfully or maliciously; Underm
effectively; Other behavior behaviors against others that cannot be included in other sub-ca

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 426 L. Zhang, Y. Deng

References Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation

models with unobservable variables and measurement error.
Journal of Marketing Research, 18(1), 39-50.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary
variable distinction in social psychological research: Conceptual,
statement. American Sociological Review, 25, 161-178.
strategic, and statistical considerations. Journal of Persoanlity
Green, D. E. (1978). Analyzing multivariate data. Hinsdale, IL:
and Social Psychology, 51(6), 1173-1182.
Becker, J.-M., Klein, K., & Wetzels, M. (2012). Hierarchical latent Dryden Press.
Hair, J. F, Jr, Black, W. C., Babin, B. J., & Anderson, R. E. (2010).
variable models in PLS-SEM: Guidelines for using reflective
Multivariate data analysis (7/E). Upper Saddle River: Prentice
formative type models. Long Range Planning, 45(5-6),
Hall.
359-394.
Hair, J. F„ Hult, G. T. M., Ringle, C. M„ & Sarstedt, M. (2014). A
Berry, C. M„ Carpenter, N. C., & Barratt, C. L. (2012). Do other
primer on partial least squares structural equation modeling
report of counterproductive work behavior provide an incre
(PLS-SEM). Thousand Oaks: Sage.
mental contribution over self-reports? A meta-analytic copari
Hershcovis, M. S., Turner, N., Barling, J., Arnold, K. A., Dupre, K.
son. Journal of Applied Psychology, 97(3), 613-636.
E., Inness, M., et al. (2007). Predicting workplace aggression: A
Blau, P. (1964). Exchange and power in social life. New York: Wiley.
meta-analysis. Journal of Applied Psychology, 92, 228-238.
Bolton, L. R., Becker, L. K„ & Barber, L. K. (2010). Big five trait
Hollinger, R. C. (1986). Acts against the workplace: Social bonding
predictors of differential counterproductive work behavior
and employee deviance. Deviant Behavior, 7, 53-75.
dimensions. Personality and Individual Differences, 49,537-541.
Holtz, B. C., & Harold, C. M. (2013). Effects of leadership
Bruk-Lee, V., Khoury, H. A., Nixon, A. E., Goh, A., & Spector, P. E.
consideration and structure on employee perceptions of justice
(2009). Replicating and extending past personality/job satisfac
and counterproductive work behavior. Journal of Organisational
tion meta-analyses. Human Performance, 22, 156-189.
Behaviour, 34, 492-519.
Cenfetelli, R. T., & Bassellier, G. (2009). Interpretation of formative
measurement in information systems research. MIS Quarterly,
Hwang, K. K. (1987). Face and favor: Chinese power game.
American Journal of Sociology, 92, 944-974.
33(4), 689-707.
Jensen, J. M., & Patel, P. C. (2011). Predicting counterproductive
Chen, X. P., & Chen, C. C. (2004). On the intricacies of the Chinese
work behavior from the interaction of personality traits.
guanxi: A process model of guanxi development. Asia Pacific
Personality and Individual Differences, 51, 466-471.
Journal of Management, 21, 305-324.
Johns, G. (1997). Contemporary research on absence from work:
Chen, Y„ Friedman, R„ Yu, E., Fang, W„ & Lu, X. (2009).
Correlates, causes and consequences. In C. L. Cooper & I.
Supervisor-subordinate guanxi: Developing a three-dimensional
T. Robertson (Eds.), International review of industrial and
model and scale. Management and Organization Review, 5(3),
375-399. organizational psychology (pp. 115-173). Chichester: John
Wiley.
Chen, Y., Friedman, R., Yu, E., & Sun, F. (2011a). Examining the
Judge, T. A., & Klinger, R. (2008). Job satisfaction: Subjective well
positive and negative effects of guanxi practice: A multi-level
being at work. In M. Eid & R. Larsen (Eds.), The science of
analysis of guanxi practices and procedural justice perceptions.
subjective well-being (pp. 393-413). New York: Guilford
Asia Pacific Journal of Management, 28(4), 715-735.
Publications.
Chen, P. Y., & Spector, P. E. (1992). Relationships of work stressors
Kalleberg, A. L. (1977). Work values and job rewards: Theory of job
with aggression, withdrawal, theft and substance use: An
satisfaction. American Sociological Review, 42, 124-143.
exploratory study. Journal of Occupational and Organizational
Keashly, L„ & Harvey, S. (2005). Emotional abuse in the workplace.
Psychology, 65, 177-184.
In S. Fox & P. E. Spector (Eds.), Counterproductive work
Chen, Y. F., & Tjosvold, D. (2006). Participative leadership by
behavior: Investigations of actors and targets (pp. 201-235).
American and Chinese managers in China: The role of relation
Washington, DC: APA.
ship. Journal of Management Studies, 43(8), 1727-1752.
Kessler, S. R., Bruursema, K., Rodopman, B., & Spector, P. E.
Chen, N„ & Tjosvold, D. (2007). Guanxi and leader member
(2013). Leadership, interpersonal conflict, and counterproductive
relationships between American managers and Chinese employ
work behavior: An examination of the stressor-strain process.
ees: Open-minded dialogue as mediator. Asia Pacific Journal of
Negotiation and Conflict Management Research, 6(3), 180-190.
Management, 24(2), 171-196.
Klotz, A. C., & Bolino, M. C. (2013). Citizenship and counterpro
Chen, N. Y„ Tjosvold, D„ Huang, X., & Xu, D. (2011b). New
ductive work behavior: A moral licensing view. Academy of
manager socialization and conflict management in China: Effects
Management Review, 38(2), 292-306.
of relationship and open conflict values. Journal of Applied
Klotz, A. C., & Buckley, M. R. (2013). A historical perspective of
Social Psychology, 41(2), 332-356.
counterproductive work behavior targeting the organization.
Cheung, M. F. Y„ Wu, W.-P., Chan, A. K. K„ & Wong, M. M. L.
Journal of Management History, 19(1), 114-132.
(2009). Supervisor-subordinate guanxi and employee work
Kossek, E. E., Pichler, S., Bodner, T., & Hammer, L. B. (2011).
outcomes: The mediating role of job satisfaction. Journal of
Business Ethics, 85(1), 77-89. Workplace social support and work-family conflict: A meta
analysis clarifying the influence of general and work-family
Coltman, T., Devinney, T. M„ Midgley, D. F., & Venaik, S. (2008).
specific supervisor and organizational support. Personnel Psy
Formative versus reflective measurement models: Two applica
chology, 64, 289-313.
tions of formative measurement. Journal of Business Research,
61, 1250-1262. Krippendorff, K. (2004). Content analysis: An introduction to its
Farh, J. L„ Earley, P. C., & Lin, S. C. (1997). Impetus for action: A
methodology. Thousand Oaks, London, New Delhi: Sage
Publications, Inc.
cultural analysis of justice and organizational citizenship
Lam, C. F., Spreitzer, G., & Fritz, C. (2014). Too much of a good
behavior in Chinese society. Administrative Science Quarterly,
thing: Curvilinear effect of positive affect on proactive behav
42(3), 421-444.
iors. Journal of Organisational Behavior, 35(4), 530-546.
Farh, J.-L., Tsui, A. S„ Xin, C. R„ & Cheng, B.-S. (1998). The
Law, K. S., Wong, C. S„ Wang, D. X., & Wang, L. H. (2000). Effect
influence of relational demography and Guanxi: The Chinese
of supervisor-subordinate guanxi on supervisory decisions in
case. Organization Science, 9(4), 471-488.

Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms
 Guanxi and CWB 427

China: Am empirica
Sakurai, K., & Jex, S. M. (2012). Coworker invivility and incivility
Human Resource targets' work effort and counterproductive work behaviors: The Man
Liang, H., moderatingSaraf,
role of supervisor social support. Journal of Occu N.
enterprisepational Health Psychology, 17(2), 150-161.
systems: Th
mediating Schneider, role
B. (1987). The people make the place. of Personnel top
Lin, L.-H., Psychology,& 40(3), 437-453.Ho, Y.-L
Journal of Spector, P. E. (2011). Business
The relationship of personality to counterpro E
Lohmoller, ductive workJ.-B.
behavior (CWB): An integration of perspectives.
(1989
least squares. Human Resource Management Review, 21, 342-352.
Heidelb
Luo, Y. (2000). Spector, P. E., Bauer, J. A., & Fox, S. (2010).Guanx
Measurement artifacts
MacKinnon, D.
in the assessment of counterproductive P.
work behavior and (2
analysis. New organizational citizenship behavior: DoYork:
we know what we think T
MacLane, C. N.,
we know? Journal of Applied &
Psychology, 95(4), 781-790. W
ductive Spector, P. E., & Fox, S. (2002). An emotion-centered
work behav model of
Resource Managemen
voluntary work behavior: Some parallels between counterpro
Martinko, ductive
M. work behavior and organizational
J., citizenship behavior.
Gund
integrative Human Resourcetheory
Management Review, 12, 269-292. of
causal reasoning Spector, P. E., & Fox, S. (2005). The stressor-emotion model persof
and Assessment, counterproductive work behavior. In S. Fox & P. E. Spector 10,
Meier, L. (Eds.), Counterproductive &
L., work behavior: Investigations
Spect of
stressors actors
and and targets (pp. 151-174). Washington,
counte DC: American
longitudinal Psychological Association. study.
529-539. Spector, P. E., Fox, S., Penney, L. M., Bruursema, K., Goh, A., &
Moorman, R. H. (1993). The influence of cognitive and affective Kessler, S. (2006). The dimensionality of counterproductivity:
based job satisfaction measures on the relationship between Are all counterproductive behaviors created equal? Journal of
satisfaction and organizational citizenship behavior. Human Vocational Behavior, 68(3), 446-460.
Relations, 6, 759-776. Tippins, M. J., & Sohi, R. S. (2003). IT competency and firm
Mount, M., Ilies, R., & Johnson, E. (2006). Relationship of performance: Is organizational learning a missing link? Strategic
personality traits and counterproductive work behaviors: The Management Journal, 24(8), 745-761.
mediating effects of job satisfaction. Personnel Psychology, 59,Warren, D. E., Dunfee, T. W., & Li, N. (2004). Social exchange in
591-622. China: The double-edged sword of guanxi. Journal of Business
O'Brien, R. M. (2007). A caution regarding rules of thumb for Ethics, 55, 355-373.
variance inflation factors. Quality & Quantity, 41, 673-690. Watson, D., & Pennabaker, J. W. (1989). Health complaints, stress,
Peng, H. (2012). Counterproductive work behavior among Chinese and distress: Exploring the central role of negative affectivity.
knowledge workers. International Journal of Selection and Psychological Review, 96, 234-254.
Assessment, 20(2), 119-138. Wei, L.-Q., Liu, J„ Chen, Y.-Y., & Wu, L.-Z. (2010). Political skills,
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organiza supervisor-subordinate guanxi and career prospects in Chinese
tional research: Problems and prospects. Journal of Manage firms. Journal of Management Studies, 47(3), 437-454.
ment, 12(A), 531-544. Wold, H. (1974). Causal flows with latent variables: Partings of ways
Reinartz, W„ Haenlein, M., & Henseler, J. (2009). An empirical in the light of NIPALS modelling. European Economics Review,
comparison of the efficacy of covariance-based and variance 5(1), 67-86.
based SEM. International Journal of Research in Marketing, Wong, Y. T„ Ngo, H. Y„ & Wong, C. S. (2003). Antecedents and
26(A), 332-344. outcomes of employees' trust in Chinese joint ventures. Asia
Ringle, C. M., Sarstedt, M., & Straub, D. W. (2012). A critical look at Pacific Journal of Management, 20(4), 481-499.
the use of PLS-SEM in MIS Quarterly. MIS Quarterly, 36(1), iii Xin, C. R., & Pearce, J. L. (1996). Guanxi: Connections as substitutes
xiv. for formal institutional support. Academy of Management
Ringle, C. M., Wende, S„ & Will, S. (2005). SmartPLS 2.0 (M3) Journal, 39(6), 1641-1658.
Beta, from http://www.SmartPLS.de. Zhang, L., Deng, Y., & Wang, Q. (2013). An exploratory study of
Robinson, S. L., & Bennett, R. J. (1995). A topology of deviant Chinese motives for building supervisor-subordinate guanxi.
behaviors: A multidimensional scaling study. Academy of Journal of Business Ethics,. doi:10.1007/sl0551-10013-l 1899
Management Journal, 38(2), 555-572. 10554.
Rotundo, M., & Xie, J. L. (2008). Understanding the domain of Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and
counterproductive work behavior in China. International Journal Kenny: Myths and truths about mediation analysis. Journal of
of Human Resource Management, 19, 856-877. Consumer Research, 37, 197-206.

£) Springer

This content downloaded from 115.178.237.10 on Sat, 29 Feb 2020 11:21:13 UTC
All use subject to https://about.jstor.org/terms