J Management Studies - 2013 - Liu - Why Do I Feel Valued and Why Do I Contribute A Relational Approach To Employee S

bs_bs_banner
Journal of Management Studies 50:6 September 2013

doi: 10.1111/joms.12037
Why Do I Feel Valued and Why Do I Contribute?

A Relational Approach to Employee’s
Organization-Based Self-Esteem and Job Performance
Jun Liu, Chun Hui, Cynthia Lee and Zhen Xiong Chen
Renmin University of China; University of Hong Kong; Northeastern University and Hong Kong Polytechnic
University; The Australian National University
ABSTRACT Drawing on the relational perspective and self-consistency theory, we theorize how
relationships involving work-centric, off-work-centric, and/or personal components can affect
an employee’s organization-based self-esteem and job performance in Chinese organizational
contexts. Matched data were collected from a multi-source sample that included 219
employee–supervisor dyads from a Chinese bank. Results based on hierarchical regression
analyses reveal that a high-quality relationship with a supervisor through work and off-work
domains (leader–member exchange and guanxi) is positively related to organization-based
self-esteem. Organization-based self-esteem plays a mediating role in the relationship between
guanxi and job performance. Additionally, career mentoring from a supervisor (a work-centric
relationship involving personal components) moderates the relationship between
organization-based self-esteem and job performance.
Keywords: career mentoring, guanxi, job performance, leader-member exchange,
organization-based self-esteem
INTRODUCTION
Organization-based self-esteem (OBSE; Pierce et al., 1989) is ‘the self-perceived value
that individuals have of themselves as organization members acting within an organiza-
tional context’ (p. 625). Similar to the importance of self-concept and self-evaluation to
human beings, OBSE is important to employees and it has implications for linking
organizational initiatives to employee behaviours. A number of studies have examined
and found support for the mediating effect of OBSE on the relationship between work
environment characteristics (e.g. managerial respect and job complexity) and employees’
behaviours and attitudes (Pierce et al., 1989); pay level and job performance (Gardner
et al., 2004); demographic dissimilarity and organizational citizenship behaviour
Address for reprints: Jun Liu, Department of Organization and Human Resources, School of Business, Renmin
University of China, Beijing 100872, China ( [email protected])
© 2013 John Wiley & Sons Ltd and Society for the Advancement of Management Studies
14676486, 2013, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/joms.12037 by Universitat Rovira i Virgili, Wiley Online Library on [08/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Relational Approach to OBSE and Job Performance 1019
(Chattopadhyay, 1999); organizational support and affective commitment (Lee and
Peccei, 2007); and delegation and job performance (Chen and Aryee, 2007). Surpris-
ingly, there is no systematic effort in examining how OBSE would mediate the effects of
work relationships on employee behaviours (Aryee et al., 2005) despite Pierce and
Gardner’s (2004) suggestion that social messages sent from significant others constitute
an important source of OBSE. This indicates the importance of examining how rela-
tionships (especially those with his or her supervisor) may affect an employee’s OBSE and
job outcomes.
We propose a relational approach to studying how OBSE can be enhanced and how
OBSE may lead to performance. The relational approach to OBSE recognizes that
relationships with people are critical for an employee’s success in an organization. There
are different forms of relationships at work, however, and we know little about how social
messages concerning the self are ‘transmitted by role models, teachers, mentors, and
those who evaluate the individual’s work’ (Pierce and Gardner, 2004, p. 594). In the
present study, we examine how relationships that vary in work-relatedness and personal
components may affect OBSE and performance. Research in the West has identified
leader–member exchange (LMX), the relationship with one’s immediate supervisor, as
the major powerful connection an employee can build in an organization (e.g. Graen and
Scandura, 1987; Wayne et al., 1997). LMX pertains to an effective relationship that is
characterized by mutual trust, respect, and obligation between a supervisor and a
subordinate (Graen and Uhl-Bien, 1995). LMX is a work-centric construct in that it does
not have to involve personal components beyond work (cf. Graen and Uhl-Bien, 1995;
Law et al., 2000). Research on LMX relies heavily on the social exchange framework for
understanding how relationships at work affect employee outcomes, but it rarely
addresses how LMX may affect an employee’s sense of self-worth. Quality relationships
with significant others would not only prompt employees to reciprocate with better
efforts and output, but they would also convey to employees the signal that they are
valued, important, competent, and capable members of the organization. Our relational
model fills this gap in both the OBSE and the LMX literature by examining how
different forms of relationships may affect employee performance via OBSE.
Human beings are gregarious, and relationships they build at work may not be
restricted entirely within work contexts. In particular, relationships at work may span
into off-work or personal domains. This is especially the case for collectivistic people like
the Chinese (Hui et al., 2004). For example, it has been noted that off-work activities
such as visiting the family of a boss are important components of Chinese relationships
(e.g. Law et al., 2000). Thus, in addition to work-centric relationships, off-work-centric
relationships with significant others in organizations may also affect employee outcomes.
Guanxi is an indigenous form of relationship in China that is off-work-centric and refers
to a relationship that is formed based on particularistic ties such as kinship or social
networks (Tsui and Farh, 1997). ‘Guan’ in Chinese literally means ‘a door lock’ or a
‘gateway’; ‘xi’ means ‘linkage’ or a ‘system of links’. Combing the two characters, ‘guanxi’
refers to ‘the connection between two parties through a system of links when one party
can choose to “lock oneself away from” or “open the link” to the other party’ (Law et al.,
2000, p. 753). Chinese employees cultivate their guanxi with supervisors, primarily, via
those informal and off-work activities such as gift presenting, family visits, and resolving
1020 J. Liu et al.
the supervisors’ personal problems (Law et al., 2000). Although the norm of reciprocity
drives the employee–supervisor guanxi relationship, given the two parties’ unequal social
statuses within the high power distance cultural context of China (Hofstede and Bond,
1988), the superiors hold the initiatives in deciding the quality of guanxi. In this study, we
propose that off-work-centric relationships may also enhance employee performance via
OBSE, much like the effects of work-centric relationships, as guanxi conveys to employees
that they are important and valuable to their superiors.
There are also relationships at work that are dual-natured, in that both work and
off-work components are involved. We characterize this kind of relationship as career
mentoring that refers to providing advice and support for career-growth activities such as
helping the protégé learn the ropes, coaching, facilitating exposure, and promotion
nomination (Kram and Isabella, 1985; Scandura, 1992). Mentors are typically more
senior supervisors and serve to guide and develop more junior subordinates (Hunt and
Michael, 1983; Kram, 1985). Mentorship can be a work relationship, as mentors provide
advice on work-related matters. Mentorship may also involve personal components,
however. During the process of mentoring, especially on personal issues such as career
development, disclosure of personal interests, priorities and wishes by the protégé,
developing a personal relationship with a protégé would greatly enhance the mentor’s
ability to offer specific advice. Thus, an effective mentor would build a relationship with
the protégé that is grounded on specific work elements, as in providing career and
work-related advice, but extend to non-work elements, as in developing a personal
relationship with the protégé or being a confidant of the protégé (Kwan et al., 2011). In
the present study, we propose that career mentoring strengthens the effects of OBSE on
employee performance.
Figure 1 depicts our conceptual model. In this model, OBSE is the mechanism through
which LMX and guanxi lead to increased performance with career mentoring as a
moderator of the OBSE–performance relationship. We suggest that having a work-
centric relationship in the form of LMX and an off-work-centric relationship in the form
of guanxi would affect an employee’s OBSE and ultimately his/her task and contextual
performance. Task performance is a form of in-role performance and refers to the
particular duties that employees are hired to perform. Contextual performance is a form
of extra-role performance that is defined as activities that ‘support the broader organi-
zational, social, and psychological environment in which the technical core must
function’ (Borman and Motowidlo, 1993, p. 73). This kind of performance is not as
quantifiable as task performance but is important to the efficient and effective functioning
Supervisor-provided
career mentoring
Leader-member exchange
Employee contribution
Organization-based Task performance
self-esteem Contextual performance
Guanxi with the supervisor
Figure 1. Conceptual model of the study
of the organization. A dual-natured relationship in the form career mentoring would
enhance the effects of OBSE on performance.
Our study makes three contributions to the literature. First, research on OBSE
primarily has used self-consistency (Ferris et al., 2009; Korman, 1970), self-enhancement
(Korman, 2001), self-protection (Korman, 2001), or behavioural plasticity (Brockner,
1988) as the theoretical perspective from which to examine the origins and consequences
of OBSE (Pierce and Gardner, 2004). We extend such theorizing to include the rela-
tional perspective. Our model integrates self-consistency theory and relational theory for
understanding employee development and behaviours. Self-consistency theory suggests
that individuals’ self-esteem is formed around their social and organizational experi-
ences, which in turn play a significant role in determining their attitudes and behaviours
(Korman, 1970, 1976). Relational theory, on the other hand, proposes that personal
growth is related to the ability to connect oneself to others in ways that foster mutual
development and learning (Fletcher and Käufer, 2003). Second, Pierce and Gardner
(2004) have identified an important antecedent of OBSE in the form of social messages
that communicate to the employees their value and competence. However, there has not
been systematic examination of the source and contents of such messages. We charac-
terize the source and contents of such messages and distinguish relationships that vary in
how much they pertain to work versus off-work domains in terms of LMX and guanxi.
Third, we theorize and examine how a dual-natured relationship, career mentoring, may
strengthen the effects of OBSE on performance. We suggest that high OBSE employees
self-verify through higher performance and the presence of career mentoring further
facilitates the self-verification process. Our study contributes to theorization of how
relationships may enhance employee self-concept as well as performance in a cross-
cultural context.
THEORETICAL BACKGROUND AND HYPOTHESES

Self-esteem is ‘an attitude of approval or disapproval and indicates the extent to which
the individual believes the self to be acceptable, significant, successful and worthy’
(Coppersmith, 1967, p. 45). As a hierarchical and multifaceted construct, self-esteem has
been conceptualized both as a global measure of overall evaluation of the self and as a
situation-specific measure of self-evaluation that results from a specific situation or role.
Korman (1970, p. 33) suggested that ‘organizations and their environments and their
actors may be viewed in terms of self-evaluations that such environments cultivate’.
OBSE is a context-specific kind of self-esteem (Pierce et al., 1989). Employees with high
OBSE have come to believe that ‘I count around here (in the organization)’. OBSE,
therefore, reflects ‘an employee’s evaluation of his or her personal adequacy and wor-
thiness as an organizational member’ (Gardner and Pierce, 1998, p. 50).
To have high OBSE, employees need to have positive self-evaluations grounded in
their work environment. Epstein (1973) develops the notion that one’s self-concept, as
well as self-evaluation, develops out of personal experience, especially those interactions
with significant others. According to Korman (2001), positive feelings result when the
employee’s work environment provides the opportunity to attain positive goals through
the employee’s own efforts and when the employee sees him/herself as deserving of and
1022 J. Liu et al.
competent in attaining goals. For instance, an employee’s most relevant ‘significant
other’ or the immediate supervisor can effectively facilitate the development of the
employee’s positive self-evaluation by the messages that he/she conveys. This reasoning
is consistent with theory of ‘relational (or interpersonal) self’, which recognizes that
knowledge about the self is linked with knowledge about significant others (Andersen and
Chen, 2002; Brewer and Gardner, 1996).
LMX and OBSE

Our relational model suggests that relationships with significant others will enhance
OBSE, which in turn affects an employee’s behaviours. As suggested by Pierce and
Gardner (2004), a major source of OBSE emerges from social messages that come from
meaningful and significant others such as teachers, role models, mentors, or those who
evaluate the individual’s work. Relationships, in general, can foster mutual development
and learning through connecting to others (Fletcher and Käufer, 2003). Workplace
relationships, especially those with immediate supervisors, have long been recognized as
important antecedents to employee outcomes.
LMX enhances employee OBSE because it signals to the employees that they or their
job contributions are valued and recognized. A fundamental proposition of LMX theory
is that supervisors do not treat all subordinates the same. Instead, LMX represents a
dyadic approach to leadership in that the leader–member relationship should be exam-
ined within each leader–member dyad (cf. Graen and Uhl-Bien, 1995). To cultivate a
high-quality LMX, the supervisor continuously serves as the role model of the employee,
develops the employee’s job skills, communicates the employee’s value to the work and
organization, and builds a dyadic partnership (Fairhurst, 1993; Graen and Uhl-Bien,
1991, 1995; McClane, 1991). This high-quality work relationship conveys to the employ-
ees that they are able and competent and enables the employees to recognize their value
to the organization. Furthermore, employees with high LMX are trusted and respected
by their supervisors. When employees receive such signals from supervisors, this positive
experience from important representatives of their organizations would enhance the
employees’ sense of worth in organizations.
Hypothesis 1: LMX positively relates to employee OBSE.
Guanxi and OBSE

Guanxi also contributes to employee OBSE. Guanxi is an extremely important determi-
nant of interpersonal interactions in China (Hwang, 1987; Jacobs, 1982). In transitional
economies where laws and formal rules are under-developed and less relied upon,
personal relationships such as guanxi can serve as effective lubricants to facilitate trans-
actions (Luo, 1997). Pearce et al. (1994) also commented that workers in transitional
societies have been observed to rely more heavily upon person-specific relations, such as
guanxi. Personal relationships with powerful people in organizations in such societies are
often the primary mechanism through which individuals experience the employment
relationship (Pearce, 2001). In addition, in the Chinese society that is characterized more
by the rule of the people (where relationships matter) than by the rule of law (Walder,
1991), decision making (i.e. career promotion) is strongly influenced by the supervisor’s
personal prerogatives (Chen and Tjosvold, 2006; Liu and Wang, 2012; Wei et al., 2010).
When a Chinese employee establishes and develops high-quality guanxi with the super-
visor, he/she demonstrates his or her commitment (and loyalty) to the supervisor, as well
as to the organization. Reciprocally, the supervisor will more likely categorize him or her
as an in-group member, which enhances that employee’s identification with the organi-
zation. In China, guanxi is a critical form of social network that defines one’s place and
social status in any group, personal network, or organization. The particularistic ties
embodied in the employee–supervisor guanxi signify the ‘unique’ value of an employee to
his/her supervisor and also to the organization. Additionally, consistent with Pierce and
Gardner (2004), guanxi is another form of social message that further conveys to the
employees about their social standing with the authority who evaluates their work. The
off-work behavioural exchanges between supervisor and the employees internalize into
the evaluation that the employee is worthy of the organization’s investment. Thus, we
predict that:
Hypothesis 2: Employee–supervisor guanxi positively relates to employee OBSE.
Joint Effect of LMX and Guanxi

As argued previously, while LMX enhances employees’ job competency and leadership
development, employee–supervisor guanxi provides the additional recognition of
employee loyalty and opportunities for career growth that relate to employee OBSE. We
further propose that the work and off-work relationships jointly influence employee
OBSE with a synergistic effect in between. According to the theory of self-concept,
evaluations of self-esteem have to be drawn from both personal and environmental cues.
While the self-concept may be a summary evaluative judgment of how important and
valuable one is, the formation of such a judgment, in many cases, depends on more than
one type of message. Researchers have recognized that the self-concept includes a diverse
collection of information about the self (Cantor et al., 1986). When information of
different foci provides converging evidence of one’s importance and competence, the
resulting self-concept would be based on stronger foundations. In work contexts, LMX
signifies to employees their value to their supervisors in work-related matters. Perceptions
of one’s own importance and competence would be greatly strengthened if there are also
messages from the supervisor that the employee is worthy of his/her supervisor’s time
and effort beyond work contexts. Supervisors have at their disposal numerous ways to
spend their off-work time. Thus, when a supervisor is willing to develop guanxi – a
personal relationship beyond work contexts – with a subordinate, this sends a powerful
message to the subordinate that he/she is not only valued by the supervisor as a
subordinate at work, but is also as an important individual beyond work. LMX, coupled
with guanxi, provides a more holistic message that spans across both the work and
off-work lives of the employees. Such converging evidence of the importance and
competence of the subordinate will greatly strengthen OBSE. LMX conveys signals of
the employee’s competency for organizational tasks. Guanxi, in addition, signifies the
1024 J. Liu et al.
employee’s prominent social status to the supervisor and, hence, in the organization.
Taken together, LMX and guanxi jointly promote employee OBSE. Thus, we predict
that:
Hypothesis 3: Employee–supervisor guanxi moderates the effect of LMX on employee

OBSE. Specifically, for employees with stronger guanxi with their supervisors, the
relationship between LMX and OBSE would be stronger than for employees with
weaker guanxi with their supervisors.
Mediating Role of OBSE

In our model, OBSE is an important employee attitude that links work relationships to
the employee’s job performance. We propose that when a supervisor is successful in
strengthening an employee’s OBSE from high-quality LMX and guanxi experiences, the
employee is likely to expend effort to achieve high performance goals, thereby making
contributions to the organization. Both LMX and guanxi provide positive and favourable
information that satisfies the employee’s self-views (i.e. OBSE). To the extent that the
employee’s esteem need is satisfied by others’ evaluations, the employee will respond
favourably by contributing to the welfare of the organization. The self-consistency system
is activated when an individual sees an opportunity to achieve high performance goals,
believes that he/she can achieve them, and also sees that the authorities or the supervisor
encourages them to do so (Korman, 1970). Thus, employees who are socialized by their
supervisors inside and outside of the work environment will be motivated to seek oppor-
tunities and encouragement to attain goals. Furthermore, subordinates who have low
OBSE may perform at a low level, as they would behave in ways that are consistent with
their low self-esteem. This suggests that the high-quality relationship from the work and
off-work interactions between the employee and supervisor should help enhance the
employee’s sense of OBSE. When established, this sense of OBSE would drive the
employee to achieve high performance goals and to contribute to the organization at a
level that is consistent with his/her level of self-esteem.
Research has demonstrated that OBSE motivates employees’ work attitudes and
behaviours (Chen and Aryee, 2007; Chen et al., 2005; Hui and Lee, 2000; Pierce and
Gardner, 2004; Pierce et al., 1989). Additionally, OBSE has been shown to relate
positively to job and extra-role performance (Alarcon, 2010; Pierce and Gardner, 2004).
Borman and Motowidlo (1997) demonstrated that supervisors roughly weight subordi-
nate tasks and contextual performance equally when making overall judgments of their
performance. Consistent with the self-consistency model and previous research, we
examined both task and contextual performance as the outcome of OBSE. Since work
relationships are only one source of OBSE (Aryee et al., 2005), there are some other
important mediating mechanisms such as personal learning (Lankau and Scandura,
2002), career motivation (Day and Allen, 2002), and role stressors (Lankau et al., 2006)
linking supervisor–employee relationships and employee outcomes. Therefore, in the
present study, we argue for a partial mediation role of OBSE in the relationship among
LMX, guanxi, and employee job performance.
Hypothesis 4: Employee OBSE partially mediates the effects of (a) LMX and (b)
employee–supervisor guanxi on an employee’s task and contextual performance.
The Moderating Role of Career Mentoring in the OBSE–Job

Performance Relationship
While LMX and guanxi communicate to the employee their value to the supervisor and
promote employee OBSE, career mentoring is a work relationship that includes both
work and off-work elements through which the supervisor can facilitate employee per-
formance. As discussed above, employees with high levels of OBSE would be more likely
to perform at higher levels to be consistent with their self-images. We suggest that career
mentoring helps high OBSE people self-verify by improving their performance. Con-
sistent with the employee’s self-image, those low in OBSE would perform at a lower level
(Pierce and Gardner, 2004; Pierce et al., 1989). In this case, even when supervisors
provide mentoring, low-OBSE employees would be less likely to take advantage in
exerting the additional effort to attain higher levels of performance. This is because
attaining higher performance is inconsistent with their self-image. However, the high-
OBSE individuals would be more likely to take advantage of supervisor mentoring to
enhance their performance. The attainment of higher performance would verify their
self-image as high OBSE people (Dipboye, 1977; Korman, 1970). Career mentoring
includes career-growth activities such as helping the protégé learn the ropes, coaching,
facilitating exposure, and nominating the employee for promotions. Career mentoring
by the supervisor constitutes an informal skill development mechanism that not only
enhances the employability of subordinates but also develops the employee’s skills and
competencies that are critical to effective implementation of the organization’s strategy.
In addition, mentors have personal interest in and care for the well-being of the protégé.
In this growth-fostering interaction, the mentor has the skills to enable the protégé, on
the one hand, while the protégé is willing to be influenced by the mentor on the other.
With such mentoring help, protégés would be better motivated and equipped to perform
better in both their task and contextual performance. High OBSE enables self-
verification through higher performance, and the presence of career mentoring makes it
more likely they can succeed at self-verification (i.e. have higher performance). We
propose that:
Hypothesis 5: Career mentoring by the supervisor moderates the effects of OBSE on

employee task and contextual performance. Specifically, for employees with stronger
supervisor-provided career mentoring, the relationship between OBSE and employee
performance would be stronger than for employees with weaker supervisor-provided
career mentoring.
METHOD
Sample and Procedure
Respondents were employees and their immediate supervisors in a locally owned bank in
a major city of Southern China. In the organization, all supervisors were required to play
1026 J. Liu et al.
the role of career mentor to their subordinate employees. Subordinates were randomly
selected from a list of non-managerial employees obtained from the personnel depart-
ment. A structured questionnaire was used to obtain data on the employee’s experiences
of LMX, guanxi with his/her supervisor, career mentoring, and OBSE. Each employee
was also requested to provide the name of his/her immediate supervisor. After complet-
ing the questionnaire, the employee sent it directly to the research team. We ceased
employee questionnaire collection after three months. From the employees’ responses,
we obtained the names of the immediate supervisor and sent questionnaires to the
supervisors to obtain employee performance data. Identity numbers were assigned to
each employee–supervisor dyad to facilitate the matching of responses. Attached to each
questionnaire was a cover letter that explained the objectives of the survey, provided
assurances of confidentiality, informed respondents that participation in the survey was
voluntary, and explained procedures for completing and returning the questionnaires.
Enclosed in each questionnaire package was a self-addressed envelope for returning the
completed instrument to the research team.
Of the 289 employee questionnaires administered, 246 completed questionnaires were
returned, resulting in a response rate of 85 per cent. Twenty-seven supervisors did not
complete the supervisor questionnaires, so the responses of their immediate subordinates
were excluded from the analysis. The final sample therefore consisted of 219 employee–
supervisor dyads representing an overall response rate of 76 per cent, far exceeding the
55.6 per cent average response rate norm suggested by Baruch (1999).
Of the 219 employee respondents, 41 per cent were male. They reported an average
age of 31.45 years (SD = 7.87), an average of 13.56 years of education (SD = 2.09), and
an average organizational tenure of 9.61 years (SD = 6.73). In terms of occupational
composition, 27 per cent were tellers, 32.7 per cent were in accounting/finance, 12.5 per
cent were in sales/marketing, 14.1 per cent were in customer service, 5.2 per cent were
in administration/human resources, 2 per cent were in computer/information technol-
ogy, and 5.6 per cent did not indicate their job functions.
Measures
The survey instrument was initially constructed in English and was translated into
Chinese and then back-translated into English. Translation and back-translation is a
standard procedure in cross-cultural research designed to ensure the equivalence of
measures used in both versions of the survey instrument (Brislin, 1980). We subsequently
performed a pilot test of the Chinese survey instrument using employees in the partici-
pating organization who were not included in the final sample. Based on feedback
obtained from the pilot test, we reworded a few items to enhance clarity. With the
exception of LMX, guanxi, and OBSE, which used a seven-point rating format, response
options for all other measures ranged from (1) ‘strongly disagree’ to (5) ‘strongly agree’.
Leader-member exchange (LMX). Employees rated the perceived quality of their exchange
with their immediate supervisor using a 7-item LMX scale developed by Graen et al.
(1982). This scale is the most frequently used measure of LMX (Liden and Maslyn,
1998). The Cronbach’s alpha of this scale was 0.75.
Guanxi with the supervisor. An employee’s guanxi with his/her supervisor was measured
using the six-item scale developed and validated by Law et al. (2000). We asked respond-
ents to indicate the likelihood of exhibiting each of these behaviours/activities: ‘During
holidays or after office hours, I call my supervisor or visit him/her’, ‘My supervisor
invites me to his/her home for lunch or dinner’, ‘On special occasions such as my
supervisor’s birthday, I would definitely visit my supervisor and send him/her gifts’, ‘I
always actively share my personal thoughts, problems, needs and feelings with my
supervisor’, ‘I care about and have a good understanding of my supervisor’s family
conditions’, and ‘When there are conflicting opinions, I will definitely stand on my
supervisor’s side’. The Cronbach’s alpha of this scale was 0.87.
Career mentoring. We adapted four items from a vocational mentoring subscale developed
by Scandura and Ragins (1993). The original mentoring measure also includes a psy-
chosocial mentoring dimension that deals with friendship and counselling of the protégé.
We omitted this dimension because we wanted to focus on the career-enhancing function
of mentoring that has a specific work component and also a personal component. The
four items were: ‘My immediate supervisor has given me important assignments’, ‘My
immediate supervisor advised me about promotional opportunities’, ‘My immediate
supervisor takes a personal interest in my career’, and ‘My immediate supervisor has
devoted special time and consideration to my career’. We dropped one item from the
original scale that reads ‘My mentor helps me coordinate professional goals’ since in our
sample (a bank), part of the respondents (e.g. tellers) were not professionals, and it would
be difficult to set a clear professional goal for this type of employee. The Cronbach’s
alpha of the scale was 0.88.
Organization-based self-esteem. This was measured by ten items developed by Pierce et al.
(1989). Sample items included: ‘I count around here’, ‘I am taken seriously around here’,
and ‘I am an important part of this place’. Pierce et al. (1989, 1993) reported alpha
coefficients of 0.86 to 0.96 for this scale. In our study, the alpha coefficient was 0.90. The
strength of the internal consistency estimates provided evidence of the homogeneity of
the scale items.
Job performance. In the study, we measured both task and contextual performance. Task
performance consists of job-specific behaviours including core job responsibilities. Con-
textual performance consists of non-job-specific behaviours, such as cooperating with
co-workers and showing dedication (Borman and Motowidlo, 1993). Research has
shown that contextual performance contributes uniquely (over and above task perform-
ance) to an individual’s contribution to the organization (Conway, 1999). Task performance
was measured via a four-item scale developed by Farh and Cheng (1999). Sample items
were: ‘This employee makes an important contribution to the overall performance of our
work unit’ and ‘The performance of this employee always meets requirements and
expectations’. The Cronbach’s alpha of the scale was 0.83. Motowidlo and van Scotter’s
(1994) 15-item scale was used to measure the two dimensions of contextual performance,
interpersonal facilitation and job dedication. Following the stem ‘While performing his
or her job, how likely is it that this employee would . . .’, supervisors rated the likelihood
1028 J. Liu et al.
that an employee would, for example, ‘praise co-workers when they are successful’
(interpersonal facilitation) and ‘persist in overcoming obstacles to complete a task’
( job dedication). Interpersonal facilitation and job dedication were measured by
seven and eight items, respectively. Their Cronbach’s alpha values were 0.90 and 0.94,
respectively.
Control variables. In the present study, we controlled for the demographic variables of age,
gender, education, and organizational tenure, as these variables were found to influence OBSE
(Bowling et al., 2010; Campbell, 1990) and job performance (Beaudry et al., 2005;
Giniger et al., 1983). Gender was coded female = ‘0’ and male = ‘1’. Education was
measured with an item that ascertained the number of years of formal education the
respondent had completed.
RESULTS
Discriminant Validity Testing
We conducted confirmatory factor analyses (CFAs) using LISREL 8.50 (Jöreskog and
Sörbom, 2001) to examine the distinctiveness of the multi-item variables in the study.
Seven latent constructs were involved in the analysis: LMX, guanxi, career mentoring,
OBSE, task performance, and the two dimensions of contextual performance, interper-
sonal facilitation and job dedication. Given the small sample size relative to the meas-
urement items, we created item-parcels prior to conducting the analyses (Landis et al.,
2000). Specifically, we first conducted an exploratory factor analysis (EFA) to obtain each
item’s factor loading coefficient. We then created item-parcels until we yielded three
indicators for each construct. For example, items with the highest and the lowest loadings
were averaged to form the first new indicator, and the items with the next highest and the
next lowest factor loadings were averaged to form the second new indicator, and so on.
The method is a common and acceptable practice in management research (e.g. Aryee
et al., 2002; Hui et al., 1999; Mathieu and Farr, 1991).
Using this method, we obtained 21 parcels that were specified to seven prescribed
latent factors. We also compared the seven-factor model to an alternative six-factor
model (combining interpersonal facilitation and job dedication), a second six-factor
model (combining LMX and guanxi), a five-factor model (combining the three perform-
ance measures), a second five-factor model (combining LMX, guanxi, and career men-
toring), and a one-factor model (combining all items into one latent factor). Results of
comparison are presented in Table I. As shown, the hypothesized seven-factor model
(c2 = 290.50, d.f. = 168; TLI = 0.95, GFI = 0.89, RMSEA = 0.06) fit the data better than
did the two six-factor models (c2 = 422.43/456.21, d.f. = 174; TLI = 0.89/0.89,
GFI = 0.84/0.83, RMSEA = 0.08/0.09), the two five-factor models (c2 = 587.14/
738.97, d.f. = 179; TLI = 0.84/0.78, GFI = 0.80/0.74, RMSEA = 0.10/0.13), and the
one-factor model (c2 = 1889.48, d.f. = 190; TLI = 0.37, GFI = 0.53, RMSEA = 0.21).
The results indicated support for the hypothesized seven-factor model and, therefore, the
distinctiveness of the variables in the study.
Table I. Results of confirmatory factor analyses of the measures
c2 d.f. TLI a GFI b RMSEAc
Distinctiveness test models

Seven-factor model 290.50 168 0.95 0.89 0.06
Six-factor model 1 (interpersonal facilitation and job 422.43 174 0.89 0.84 0.08
dedication were combined into one factor)
Six-factor model 2 (guanxi and leader–member exchange 456.21 174 0.89 0.83 0.09
were combined into one factor)
Five-factor model 1 (three performance measures were 587.14 179 0.84 0.80 0.10
combined into one factor)
Five-factor model 2 (guanxi, leader–member exchange, 738.97 179 0.78 0.74 0.13
and mentoring were combined into one factor)
One-factor model (all 21 items were combined into one 1889.48 190 0.37 0.53 0.21
overall factor)
Common method variance test models
Harman one-factor model for employee data 809.24 54 0.35 0.58 0.27
Harman one-factor model for supervisor data 312.12 27 0.69 0.73 0.24
Notes: a TLI = Tucker–Lewis Index.

b
GFI = Goodness of Fit Index.
c
RMSEA = Root Mean Square Error of Approximation.
Common Method Variance Issues Testing

Of the seven multi-item variables we studied, four (LMX, career mentoring, guanxi, and
OBSE) were drawn from a common source (employees) and the other three (task
performance, interpersonal facilitation, and job dedication) were drawn from another
source (supervisor). We followed the suggestions of Podsakoff et al. (2003) to test if there
was a strong common method effect that underpinned the data from the same source.
We conducted analyses of the two sources separately. As shown in Table I, results of the
Harman one-factor test yielded unacceptable fit (c2 = 809.24/312.12, d.f. = 54/27;
TLI = 0.35/0.69, GFI = 0.58/0.73, RMSEA = 0.27/0.24). In addition, for each source,
we also specified a structural equation model consisting of the theoretical constructs
together with an unmeasured latent method factor. The model’s fitness was not superior
to the proposed measurement model. Taken together, these results indicated that the
constructs could be discriminated from each other and common method bias was not
likely to have detrimental effects on the validity of our results and conclusions.
Hypothesis Testing
Table II presents the descriptive statistics and zero-order correlations among the vari-
ables. To test our model that involved both moderating and mediating effects, we
adopted the moderated causal steps approach in the regression analysis (Muller et al.,
2005). We examined these effects hierarchically: direct effects (including control vari-
ables, LMX, and guanxi) first, then the LMX–guanxi interaction, mediating effects
1030
Table II. Means, standard deviations, and correlations of the study’s variables
Variable Mean SD 1 2 3 4 5 6 7 8 9 10 11
1. Age 31.45 7.87 –

2. Gendera 0.41 0.49 0.03 –
3. Education (yr) 13.56 2.09 -0.12 -0.02 –
4. Organizational tenure (yr) 9.61 6.73 0.79** -0.14 -0.18* –
5. Leader–member exchange 3.46 0.59 0.08 0.12 -0.02 0.07 (0.75)c
6. Guanxi with supervisor 3.57 1.09 -0.27** 0.19** -0.02 -0.26** 0.29** (0.87)
7. Career mentoring 3.13 0.84 -0.10 0.08 -0.05 -0.10 0.39** 0.31** (0.88)
8. OBSEb 4.60 0.91 0.00 -0.03 0.15* 0.01 0.40** 0.27** 0.32** (0.90)
J. Liu et al.
9. Task performance 3.38 0.63 -0.01 0.14 0.13 0.04 0.12 0.06 0.15* 0.31** (0.83)
10. Interpersonal facilitation 3.78 0.59 0.04 0.09 0.12 0.03 0.21** 0.15* 0.21** 0.31** 0.51** (0.90)
11. Job dedication 3.89 0.66 0.07 0.05 0.21** 0.03 0.18* 0.18* 0.25** 0.36** 0.57** 0.73** (0.94)
Notes: a Gender coding: 0 = female; 1 = male.

b
OBSE = organization-based self-esteem.
c
Numbers in the brackets are Cronbach’s alpha values.
* Significant at the 0.05 level.
** Significant at the 0.01 level.
Table III. Hierarchical regression results of the mediated moderation
tests
OBSE c
Model 1 Model 2 Model 3 Model 4
Step 1: Control variables

Age -0.13a -0.10 -0.10 -0.09
Gender 0.01 -0.08 -0.13* -0.15*
Education 0.11 0.11 0.13* 0.13*
Company tenure 0.13 0.07 0.13 0.11
Step 2: LMXb 0.46** 0.37** 0.36**
Step 3: Guanxi 0.23** 0.22**
Step 4: LMX ¥ Guanxi 0.10
R2 0.64 0.21** 0.26** 0.27**
F 0.02 7.84 8.04 7.30
DR2 – 0.19** 0.05** 0.01
DF – 37.48 6.75 1.79
Notes: a Standardized regression coefficients are reported.

b
LMX = leader–member exchange.
c
(OBSE), and finally the OBSE–career mentoring interaction. Multi-collinearity prob-

lems were investigated in these tests, and no serious threat was identified. The hierar-
chical regression results are presented in Tables III and IV.
Hypotheses 1 and 2 stated that LMX and guanxi have positive effects on OBSE. As
shown in Table III, LMX was positively related to OBSE (b = 0.46, p < 0.01; Model 2).
In addition, guanxi was positively related to OBSE (b = 0.23, p < 0.01; Model 3). The
results supported Hypotheses 1 and 2. Our Hypothesis 3 predicted that guanxi would
moderate the effect of LMX on employee OBSE. However, the regression coefficient of
the LMX–guanxi interaction on OBSE was not significant (b = 0.10, ns; Model 4).
Hypothesis 3 was not supported by the data.
Hypothesis 4 predicted that OBSE would partially mediate the effects of LMX and
guanxi on an employee’s job performance. According to Baron and Kenny’s (1986)
criteria (later elaborated by Muller et al. (2005) in the moderated-mediation situation),
mediation is supported if four conditions are met: (1) the predictor (i.e. LMX, guanxi, or
LMX¥guanxi) is significantly related to the outcome variable (i.e. job and contextual
performance); (2) the predictor is significantly related to the mediator (i.e. OBSE); (3) the
mediator is significantly related to the outcome variable; and (4) when both the predictor
and the mediator are entered into the model, the relationship between the predictor and
the outcome variable becomes non-significant (full mediation) or diminishes (partial
mediation). Following these criteria, we found four significant predictor–outcome rela-
tionships turned to non-significant or less significant in the presence of OBSE (see
1032
Table IV. Hierarchical regression results of the mediated moderation tests
Task performance Interpersonal facilitation Job dedication
Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14 Model 15 Model 16
Step 1: Control variables

Age -0.16a -0.11 -0.09 -0.09 -0.25* -0.22 -0.20 -0.20 -0.14 -0.09 -0.07 -0.07
Gender 0.22** 0.13 0.18* 0.17* 0.20* 0.11 0.14* 0.12 0.14* 0.03 0.07 0.06
Education 0.17* 0.19** 0.15* 0.12 0.14 0.16* 0.13 0.10 0.26** 0.28** 0.24** 0.22**
Company tenure 0.22 0.22 0.19 0.15 0.30* 0.30* 0.28* 0.24* 0.27* 0.27* 0.25* 0.22*
Step 2: LMXb and guanxi effects
LMX -0.03 -0.16 -0.14 0.12 0.04 0.04 0.04 -0.07 -0.07
Guanxi 0.20* 0.13 0.10 0.16* 0.10 0.06 0.27** 0.21* 0.17*
LMX ¥ Guanxi 0.21** 0.17* 0.13* 0.09 0.07 0.03 0.12 0.09 0.06
Step 3: Mediator
OBSEc 0.36** 0.37** 0.24** 0.24** 0.30** 0.28**
J. Liu et al.
Step 4: Moderator and interaction

Career mentoring -0.01 0.04 0.06
OBSE ¥ Career mentoring 0.18* 0.19* 0.17*
R2 0.07* 0.15* 0.25** 0.28** 0.07* 0.13** 0.18** 0.21** 0.09** 0.19** 0.25** 0.28**
F 3.00 3.77 5.98 5.43 2.69 3.10 3.78 3.71 3.72 4.62 5.86 5.37
DR2 – 0.08** 0.10** 0.03* – 0.06** 0.05** 0.03* – 0.10* 0.06** 0.03*
DF – 4.51 18.23 2.67 – 3.47 7.52 3.00 – 5.36 12.03 2.78
Notes: a Standardized regression coefficients are reported.

b
LMX = leader–member exchange.
c
High-level of career mentoring
Task performance
Average-level of career mentoring
Low-level of career mentoring
1SD below mean 1SD above mean

OBSE
Figure 2. Moderating effect of career mentoring on the relationship between OBSE and task performance
Table IV for details). The four predictor–outcome relationships were those between
guanxi and task performance (from 0.20 in Model 6 to 0.13 in Model 7), guanxi and
interpersonal facilitation (from 0.16 in Model 10 to 0.10 in Model 11), guanxi and job
dedication (from 0.27 in Model 14 to 0.21 in Model 15), and finally, the LMX–guanxi
interaction and task performance (from 0.21 in Model 6 to 0.17 in Model 7). We then
conducted a Sobel test (Preacher and Hayes, 2004) to test the significance of the
reduction of regression coefficient for the four predictor–outcome relationships. Results
showed that three of the four were significant: guanxi–OBSE–task performance (Sobel
t = 2.22, p < 0.05), guanxi–OBSE–interpersonal facilitation (Sobel t = 1.99, p < 0.05), and
guanxi–OBSE–job dedication (Sobel t = 2.08, p < 0.05). Taken together, these results
indicate that our data supported Hypothesis 4b but not Hypothesis 4a.
Our Hypothesis 5 predicted that career mentoring would moderate the effects of
OBSE on an employee’s task and contextual performance. Regression results in
Table IV showed that the OBSE–mentoring interaction was significantly related to task
performance (b = 0.18, p < 0.05; Model 8), interpersonal facilitation (b = 0.19, p < 0.05;
Model 12), and job dedication (b = 0.17, p < 0.05; Model 16). To interpret the general
pattern of the interactive influence, we followed the procedure suggested by Aiken and
West (1991). For instance, in Figure 2, we plotted OBSE–task performance relationships
according to three levels of career mentoring: the mean, one standard deviation above
the mean, and one standard deviation below the mean. As shown in the figure, OBSE
had a greater positive impact on task performance when the employee received higher-
rather than lower-level career mentoring from his/her supervisor. In addition, Figures 3
and 4 visually illustrate how the OBSE–interpersonal facilitation and OBSE–job dedi-
cation relationships varied with the level of career mentoring.
DISCUSSION
This study examined how work-centric (LMX) and off-work-centric ( guanxi ) employee–
supervisor relationships relate to employees’ OBSE, which in turn, relates to task and
contextual performance. We further examined how a dual-natured relationship (career
mentoring) moderates the relationship between OBSE and employee contributions.
Results from 219 employee–supervisor dyads supported our model in a general sense.
1034 J. Liu et al.
Interpersonal facilitation

OBSE
Figure 3. Moderating effect of career mentoring on the relationship between OBSE and interpersonal
facilitation

Job dedication


OBSE
Figure 4. Moderating effect of career mentoring on the relationship between OBSE and job dedication
Specifically, we found that both LMX and guanxi related to OBSE. However, OBSE
mediated only the relationship of guanxi on both task and contextual performance. We
also found that career mentoring moderated the relationship between OBSE and both
task and contextual performance.
As suggested by Hui et al. (2004), a Chinese employee’s relationship with his/her
supervisor takes on paramount importance and is an essential component of the Chinese
social structure. The relationship with one’s supervisor, including guanxi, may anchor the
employee’s relationship with the organization and his/her willingness to contribute to it.
We argue that employees who form special relationships with supervisors would likely
feel that they are special to the supervisor and the organization. This argument is
supported by the effects of guanxi but not LMX on the employee’s performance. It is
possible that, especially in a relationship-oriented society such as China’s, developing
quality work-based relationships with subordinates is taken for granted as one of super-
visor obligations. Besides, Chinese employees had been found to define their job scopes
more broadly than American or Australian employees (Lam et al., 1999). Chinese
employees, especially those who espouse traditional Chinese values, are likely to engage
in extra-role performance regardless of LMX quality (Hui et al., 2004). Thus, having
high LMX at work is less of a discriminating signal for how special someone is to the
supervisor and the organization. Further, its effects on employee performance may be
due to mechanisms other than OBSE. Developing off-work relationships such as guanxi,
however, is not an obligation but is at the discretion of the supervisors. Thus, having
guanxi with a supervisor is more likely to signal a special relationship that strengthens
OBSE and, subsequently, performance.
We did not find support for the interactive effects of LMX and guanxi on OBSE. This
is slightly surprising given the importance of relationships in China. We found that
although both LMX and guanxi bolster OBSE, LMX shows a stronger relationship to
OBSE than guanxi. This may come from both LMX and OBSE referencing the work
domain. The correlation table showed that career mentoring, which contains relation-
ships from both work and off-work domains, had a moderate correlation with OBSE.
Future studies should pay attention to this pattern of relationships.
Further, our results support the moderating effects of a dual-natured, growth-fostering
relationship of career mentoring on the relationship between OBSE and employee
performance. This result suggests that for relationships to be effective in Chinese organi-
zations, they should contain off-work components. LMX is a work-centric relationship
and on its own, as compared to guanxi, provides more limited information on how special
someone is to the supervisor and the organization.
The findings of this study have theoretical and practical implications. In view of the
motivational implications of the relational ties between an organization and its employ-
ees, the first theoretical implication is an insight into the development of OBSE and
ultimately employee contributions in terms of task and contextual performance. Specifi-
cally, the findings revealed that work-related and off-work-related employee–supervisor
relationships constitute sources of OBSE. Developing social relationships and offering
these work and off-work activities to employees are signals to employees that their
organization cares for their well-being. McAllister and Bigley (2002) noted that by
centring its values and principles on fulfilling employees’ needs and acting in their best
interests, an organization will cultivate higher OBSE. However, in terms of affecting
employee performance, relationships that include off-work personal components are
important in China.
Our results show that the relational and consistency perspectives reinforced one
another in suggesting that career mentoring strengthens the OBSE effect on employee
performance. According to self-consistency theory, once self-esteem is crystallized,
people behave in ways and develop attitudes that are consistent with their level of
self-esteem (Ferris et al., 2009; Korman, 1970). However, in organizations, when
employees perceive an opportunity to improve their OBSE, such as when their super-
visors provide personal interest and support, they too will try to enhance their perform-
ance that reflects the growth of self-esteem (Korman, 2001). As we previously indicated,
the Chinese ‘self’ is more interdependent and largely built around a network of roles and
relationships than is the ‘self’ in Western settings. In traditional Confucian thinking, the
‘self’ does not exist outside of others. The ‘self’ is a relational one, emphasizing a sense of
connectedness and interdependence. Therefore, the growth-in-connection perspective
building from supervisor-provided career mentoring supports and extends the current
conceptualization of OBSE.
1036 J. Liu et al.
Conceptually, our study advances understanding of how the social messages embed-
ded in different relationships at work affect OBSE. Pierce and Gardner (2004) suggested
two other types of antecedents to OBSE: environmental structures and personal expe-
riences. Specifically, they suggested that organic organizational structures that offer
employees control and autonomy, and positive personal experiences such as success in
the organization, would be more conducive to cultivating high OBSE. It is likely that
these different antecedents may operate synergistically to enhance OBSE. For example,
as suggested by Pierce and Gardner (2004), employees operating in an organic organi-
zational structure would be more likely to develop high OBSE. The effects of organiza-
tional structure on OBSE would be strengthened to the extent that employees have
already attained previous success in the organization. OBSE would be maximized
when these employees develop important relationships with significant others in the
organization. The resulting high OBSE may have an impact on task performance, as
well as on career progression. Future studies may examine how these different categories
of OBSE antecedents interact together to affect OBSE and individual and organizational
outcomes.
From a practical or managerial perspective, the finding that components of the
self-concept have motivational implications for an employee’s work-related attitudes and
behaviours suggest a role for human resource practices in fostering positive employee
self-concepts. For these human resource practices to be effective, organizations could
consider communicating respect and addressing the relational and esteem needs of
employees. By satisfying the ‘belongingness’ needs of employees, these human resource
practices promote stronger personal relationships between employees and their employ-
ers, which, in turn, enhance feelings of self-worth in the context of work or OBSE. One
implication of the importance of off-work activities is that managers would have to spend
more time in off-work contexts with their subordinates for them to be effective. Future
studies should examine how other implicit (such as supportive management culture or
job security) and explicit cues (such as positive feedback, experience of success from
promotion, and salary increase) from the work environment interact with social messages
and contribute to the development and maintenance of OBSE.
As with most studies, this one has a number of limitations. First, the cross-sectional
design implies that causal relations cannot be inferred from our findings. Future research
may utilize longitudinal designs to ascertain the causalities that we examined in this study.
Second, with the exception of task and contextual performance (supervisor-rated), data
on the study variables were based on self-reports. Consequently, our findings may be
attributed to method bias, particularly with regard to the antecedents of OBSE. However,
statistical examination of this issue revealed that there is no serious common method bias.
To mitigate concerns about method bias and to build multi-level research, future studies
may obtain data from sources other than respondents as antecedents. Third, data were
obtained from a single organization in a mono-cultural context, which limits the gener-
alizability of our findings. Fourth, in theory, we speculate that our findings, in which
off-work relationships are important to employee OBSE and performance, could be
applicable in a Western setting. However, the variable we used to reflect this off-work
relationship in our study (i.e. guanxi) may be somewhat unique to Chinese culture. We
suggest that future research should test our findings, using other operationalization of
off-work relationships in the particular culture studied or examine the relevance of guanxi
in a cross-cultural context.
Despite these limitations, our study provides insight into how employers or superiors
can influence the self-concepts and performance of employees via work and off-work
relationships. In general, our findings revealed that it may be important to consider
personal elements in relationships at work and how such relationships affect the self
concepts of employees.
ACKNOWLEDGMENTS
The study was supported by a grant from the Research Grants Council of HKSAR, China (Project
HKU745410H). The work described in this paper by Cynthia Lee was supported by the Provost Enhance-
ment Award, College of Business Administration, Northeastern University.
REFERENCES
Aiken, L. S. and West, S. G. (1991). Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA:
Sage.
Alarcon, G. (2010). ‘A meta-analysis of the predictors and consequences of organization-based self-esteem’.
Journal of Occupational and Organizational Psychology, 83, 601–26.
Andersen, S. M. and Chen, S. (2002). ‘The relational self: an interpersonal social-cognitive theory’. Psycho-
logical Review, 109, 619–45.
Aryee, S., Budhwar, P. and Chen, Z. X. (2002). ‘Trust as a mediator of the relationship between organi-
zational justice and work outcomes: test of a social exchange model’. Journal of Organizational Behavior, 23,
267–85.
Aryee, S., Budhwar, P. and Tan, H. H. (2005). ‘Leader-member exchange and contextual performance: the
mediating influence of organization-based self-esteem’. In Turnipseed, D. L. (Ed.), Handbook of Organi-
zational Citizenship Behavior. New York: Nova Science, 67–77.
Baron, R. M. and Kenny, D. A. (1986). ‘The moderator-mediator variable distinction in social psychological
research: conceptual, strategic, and statistical considerations’. Journal of Personality and Social Psychology,
51, 1173–82.
Baruch, Y. (1999). ‘Response rate in academic studies: a comparative analysis’. Human Relations, 52, 421–38.
Beaudry, P., Collard, F. and Green, D. A. (2005). ‘Demographics and recent productivity performance:
insights from cross-country comparisons’. Canadian Journal of Economics, 38, 309–44.
Borman, W. C. and Motowidlo, S. J. (1993). ‘Expanding the criterion domain to include elements of
contextual performance’. In Schmitt, N. and Borman, W. C. (Eds), Personnel Selection in Organizations. San
Francisco, CA: Jossey Bass, 71–98.
Borman, W. C. and Motowidlo, S. J. (1997). ‘Task performance and contextual performance: the meaning
for personal selection research’. Human Performance, 10, 99–109.
Bowling, N. A., Eschleman, K. J., Wang, Q., Kirkendall, C. and Alarcon, G. (2010). ‘A meta-analysis of the
predictors and consequences of organization-based self-esteem’. Journal of Occupational and Organizational
Psychology, 83, 601–26.
Brewer, M. B. and Gardner, W. (1996). ‘Who is this “we”? Levels of collective identity and self represen-
tation’. Journal of Personality and Social Psychology, 71, 83–93.
Brislin, R. W. (1980). ‘Translation and content analysis of oral and written materials’. In Triandis, H. C. and
Berry, J. W. (Eds), Handbook of Cross-Cultural Psychology, Vol. 2. Boston, MA: Allyn & Bacon, 389–444.
Brockner, J. (1988). Self-Esteem at Work. Lexington, MA: Lexington Books.
Campbell, J. D. (1990). ‘Self-esteem and clarity of the self-concept’. Journal of Personality and Social Psychology,
59, 538–49.
Cantor, N., Markus, H., Niedenthal, P. and Nurius, P. (1986). ‘On motivation and the self-concept’. In
Sorrentino, R. M. and Higgins, E. T. (Eds), Handbook of Motivation and Cognition. New York: The Guilford
Press, 96–121.
Chattopadhyay, P. (1999). ‘Beyond direct and symmetrical effects: the influence of demographic dissimilarity
on organizational citizenship behavior’. Academy of Management Journal, 42, 273–87.
1038 J. Liu et al.
Chen, Y. F. and Tjosvold, D. (2006). ‘Participative leadership by American and Chinese managers in China:
the role of relationships’. Journal of Management Studies, 43, 1727–52.
Chen, Z. X. and Aryee, S. (2007). ‘Delegation and employee work outcomes: the cultural context of
mediating processes’. Academy of Management Journal, 50, 226–38.
Chen, Z. X., Aryee, S. and Lee, C. (2005). ‘Test of a mediated model of perceived organizational support’.
Journal of Vocational Behavior, 66, 457–70.
Conway, J. M. (1999). ‘Distinguishing contextual performance from task performance for managerial jobs’.
Journal of Applied Psychology, 84, 3–13.
Coppersmith, S. V. (1967). Antecedents of Self-Esteem. San Francisco, CA: Freeman.
Day, R. and Allen, T. (2002). Why Are Protégés More Successful: Explaining the Mentoring–Career Success Relationship.
Paper presented at the 17th Annual the Society for Industrial and Organizational Psychology, April
2002, Toronto, Canada.
Dipboye, R. L. (1977). ‘A critical review of Korman’s self-consistency theory of work motivation and
occupational choice’. Organizational Behavior and Human Performance, 18, 108–26.
Epstein, S. (1973). ‘The self-concept revisited, or a theory of a theory’. American Psychologist, 28, 404–16.
Fairhurst, G. T. (1993). ‘The leader-member exchange patterns of women leaders in industry: a discourse
analysis’. Communication Monograph, 60, 321–51.
Farh, J. L. and Cheng, B. S. (1999). ‘An investigation of modesty bias in self-ratings of work performance
among Taiwan workers’. Chinese Journal of Psychology, 39, 103–18 (in Chinese).
Ferris, D. L., Brown, D. J., Lian, H. and Keeping, L. M. (2009). ‘When does self-esteem relate to deviant
behavior? The role of contingencies of self-worth’. Journal of Applied Psychology, 94, 1345–53.
Fletcher, J. K. and Käufer, K. (2003). ‘Shared leadership: paradox and possibility’. In Pearce, C. L. and
Conger, J. C. (Eds), Shared Leadership: Reframing the Hows and Whys of Leadership. Thousand Oaks, CA:
Sage, 21–47.
Gardner, D. G. and Pierce, J. L. (1998). ‘Self-esteem and self-efficacy within the organizational context’.
Group and Organization Management, 23, 48–70.
Gardner, D. G., Van Dyne, L. and Pierce, J. L. (2004). ‘The effects of pay level on organization-based
self-esteem and performance: a field investigation’. Journal of Occupational and Organizational Psychology, 77,
307–22.
Giniger, S., Dispenzieri, A. and Eisenberg, J. (1983). ‘Age, experience, and performance on speed and skill
jobs in an applied setting’. Journal of Applied Psychology, 68, 469–75.
Graen, G. B. and Scandura, T. A. (1987). ‘Toward a psychology of dyadic organizing’. In Cummings, L. L.
and Staw, B. W. (Eds), Research in Organizational Behavior, Vol. 9. Greenwich, CT: JAI Press, 175–208.
Graen, G. B. and Uhl-Bien, M. (1991). ‘Partnership-making applies equally well to teammate-sponsor
teammate-competence network, and teammate-teammate relationships’. Journal of Management Systems,
3, 49–54.
Graen, G. B. and Uhl-Bien, M. (1995). ‘Relationship-based approach to leadership: development of leader-
member exchange (LMX) theory of leadership over 25 years: applying a multi-level multi-domain
perspective’. Leadership Quarterly, 6, 219–47.
Graen, G. B., Novak, M. A. and Sommerkamp, P. (1982). ‘The effects of leader-member exchange and job
design on productivity and satisfaction: testing a dual attachment model’. Organizational Behavior and
Human Performance, 30, 109–31.
Hofstede, G. and Bond, M. H. (1988). ‘The Confucius connection: from cultural roots to economic growth’.
Organizational Dynamics, 16, 4–21.
Hui, C. and Lee, C. (2000). ‘Moderating effects of organization-based self-esteem on organizational
uncertainty-employee response relationships’. Journal of Management, 26, 215–32.
Hui, C., Law, K. S. and Chen, Z. X. (1999). ‘A structural equation model of the effects of negative affectivity,
leader-member exchange and perceived job mobility on in-role and extra-role performance: a Chinese
case’. Organizational Behavior and Human Decision Processes, 77, 3–21.
Hui, C., Lee, C. and Rousseau, D. M. (2004). ‘Employment relationships in China: do workers relate to the
organization or the people?’. Organization Science, 15, 232–40.
Hunt, D. M. and Michael, C. (1983). ‘Mentorship: a career training and development tool’. Academy of
Management Review, 8, 475–85.
Hwang, K. K. (1987). ‘Face and favor: the Chinese power game’. American Journal of Sociology, 92, 944–
74.
Jacobs, J. B. (1982). ‘The concept of guanxi and local politics in a rural Chinese cultural setting’. In
Greenblatt, S. L., Wilson, R. W. and Wilson, A. A. (Eds), Social Interaction in Chinese Society. New York:
Praeger Publishers, 209–36.
Jöreskog, K. G. and Sörbom, D. (2001). LISREL 8.5: Structural Equation Modeling with the SIMPLIS Command
Language. Chicago, IL: Scientific Software International.
Korman, A. K. (1970). ‘Toward a hypothesis of work behavior’. Journal of Applied Psychology, 54, 31–41.
Korman, A. K. (1976). ‘Hypothesis of work behavior revisited and an extension’. Academy of Management
Review, 1, 50–63.
Korman, A. K. (2001). ‘Self enhancement and self protection: towards a theory of motivation’. In Erez, M.,
Kleinbeck, U. and Thierry, H. (Eds), Work Motivation in the Context of a Globalizing Economy. Mahwah, NJ:
Lawrence Erlbaum Associates, 121–30.
Kram, K. E. (1985). Mentoring at Work: Developmental Relationships in Organizational Life. Glenview, IL: Scott
Foreman.
Kram, K. E. and Isabella, L. A. (1985). ‘Mentoring alternatives: the role of peer relationships in career
development’. Academy of Management Journal, 28, 110–32.
Kwan, H. K., Liu, J. and Yim, F. H. (2011). ‘Effects of mentoring functions on receivers’ organizational
citizenship behavior in a Chinese context: a two-study investigation’. Journal of Business Research, 64,
363–70.
Lam, S. K., Hui, C. and Law, K. (1999). ‘Organizational citizenship behavior: comparing perspectives of
supervisors and subordinates across four international samples’. Journal of Applied Psychology, 84, 594–601.
Landis, R. S., Beal, D. J. and Tesluk, P. E. (2000). ‘A comparison of approaches to forming composite
measures in structural equation models’. Organizational Research Methods, 3, 186–207.
Lankau, M. J. and Scandura, T. A. (2002). ‘An investigation of personal learning in mentoring relationships:
content, antecedents, and consequences’. Academy of Management Journal, 45, 779–91.
Lankau, M. J., Carlson, D. S. and Nielson, T. R. (2006). ‘The mediating influence of role stressors in the
relationship between mentoring and job attitudes’. Journal of Vocational Behavior, 68, 308–22.
Law, K. S., Wong, C. S., Wang, D. and Wang, L. (2000). ‘Effects of supervisor-subordinate guanxi on
supervisory decisions in China: an empirical investigation’. International Journal of Human Resource Man-
agement, 11, 751–65.
Lee, H. J. and Peccei, R. (2007). ‘Organizational-level gender dissimilarity and employee commitment’.
British Journal of Industrial Relations, 45, 687–712.
Liden, R. C. and Maslyn, J. M. (1998). ‘Multidimensionality of leader-member exchange: an empirical
assessment through scale development’. Journal of Management, 24, 43–72.
Liu, X. and Wang, J. (2012). ‘Abusive supervision and organizational citizenship behaviour: is supervisor–
subordinate guanxi a mediator?’. International Journal of Human Resource Management, 23, 3983–4008.
Luo, Y. (1997). ‘Guanxi and performance of foreign-invested enterprises in China: an empirical inquiry’.
Management International Review, 37, 51–70.
Mathieu, J. E. and Farr, J. L. (1991). ‘Further evidence for the discriminant validity of measures of
organizational commitment, job involvement, and job satisfaction’. Journal of Applied Psychology, 76,
127–33.
McAllister, D. J. and Bigley, G. A. (2002). ‘Work context and the (re)definition of self: how organizational
care influences organization-based self esteem’. Academy of Management Journal, 45, 894–904.
McClane, W. E. (1991). ‘Implications of member role differentiation: analysis of a key concept in the LMX
model of leadership’. Group and Organization Studies, 16, 102–13.
Motowidlo, S. J. and van Scotter, J. R. (1994). ‘Evidence that task performance should be distinguished from
contextual performance’. Journal of Applied Psychology, 79, 475–80.
Muller, D., Judd, C. M. and Yzerbyt, V. Y. (2005). ‘When moderation is mediated and mediation is
moderated’. Journal of Personality and Social Psychology, 89, 852–63.
Pearce, J. L. (2001). Organization and Management in the Embrace of Government. Mahwah, NJ: Erlbaum.
Pearce, J. L., Branyiczki, I. and Bakacsi, G. A. (1994). ‘Person-based reward systems’. Journal of Organizational
Behavior, 15, 261–82.
Pierce, J. L. and Gardner, D. G. (2004). ‘Self-esteem within the work and organizational context: a review
of the organization-based self-esteem literature’. Journal of Management, 30, 591–622.
Pierce, J. L., Gardner, D. G., Cummings, L. L. and Dunham, R. B. (1989). ‘Organization-based self-esteem:
construct definition, measurement and validation’. Academy of Management Journal, 32, 622–48.
Pierce, J. L., Gardner, D. G., Dunham, R. B. and Cummings, L. L. (1993). ‘Moderating effects of
organization-based self-esteem on role condition-employee response relationships’. Academy of Manage-
ment Journal, 36, 271–88.
Podsakoff, P. M., MacKenzie, S. M., Lee, J. and Podsakoff, N. P. (2003). ‘Common method variance in
behavioral research: a critical review of the literature and recommended remedies’. Journal of Applied
Psychology, 88, 879–903.
1040 J. Liu et al.
Preacher, K. J. and Hayes, A. F. (2004). ‘SPSS and SAS procedures for estimating indirect effects in simple
mediation models’. Behavior Research Methods, Instruments, and Computers, 36, 717–31.
Scandura, T. A. (1992). ‘Mentorship and career mobility: an empirical investigation’. Journal of Organizational
Behavior, 13, 169–74.
Scandura, T. A. and Ragins, B. R. (1993). ‘The effects of sex and gender role orientation on mentorship in
male-dominated occupations’. Journal of Vocational Behavior, 43, 251–65.
Tsui, A. S. and Farh, J. L. (1997). ‘Where guanxi matters: relational demography and guanxi in the Chinese
context’. Work and Occupations, 24, 56–79.
Walder, A. G. (1991). ‘Workers, managers, and the state: the reform era and the political crisis of 1989’.
China Quarterly, 127, 467–92.
Wayne, S. J., Shore, L. M. and Liden, R. C. (1997). ‘Perceived organizational support and leader-member
exchange: a social exchange perspective’. Academy of Management Journal, 40, 82–111.
Wei, L. Q., Liu, J., Chen, Y. Y. and Wu, L. Z. (2010). ‘Political skill, supervisor-subordinate guanxi and career
prospects in Chinese firms’. Journal of Management Studies, 47, 437–54.

J Management Studies - 2013 - Liu - Why Do I Feel Valued and Why Do I Contribute A Relational Approach To Employee S

Uploaded by

Copyright:

Available Formats

J Management Studies - 2013 - Liu - Why Do I Feel Valued and Why Do I Contribute A Relational Approach To Employee S

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

J Management Studies - 2013 - Liu - Why Do I Feel Valued and Why Do I Contribute A Relational Approach To Employee S

Uploaded by

Copyright:

Available Formats

bs_bs_banner

Journal of Management Studies 50:6 September 2013

Why Do I Feel Valued and Why Do I Contribute?

Figure 1. Conceptual model of the study

THEORETICAL BACKGROUND AND HYPOTHESES

LMX and OBSE

Hypothesis 1: LMX positively relates to employee OBSE.

Guanxi and OBSE

Hypothesis 2: Employee–supervisor guanxi positively relates to employee OBSE.

Joint Effect of LMX and Guanxi

Hypothesis 3: Employee–supervisor guanxi moderates the effect of LMX on employee

Mediating Role of OBSE

The Moderating Role of Career Mentoring in the OBSE–Job

Hypothesis 5: Career mentoring by the supervisor moderates the effects of OBSE on

c2 d.f. TLI a GFI b RMSEAc

Distinctiveness test models

Notes: a TLI = Tucker–Lewis Index.

Common Method Variance Issues Testing

1. Age 31.45 7.87 –

Notes: a Gender coding: 0 = female; 1 = male.

Model 1 Model 2 Model 3 Model 4

Step 1: Control variables

Notes: a Standardized regression coefficients are reported.

(OBSE), and finally the OBSE–career mentoring interaction. Multi-collinearity prob-

Table IV. Hierarchical regression results of the mediated moderation tests

Task performance Interpersonal facilitation Job dedication

Step 1: Control variables

Step 4: Moderator and interaction

Notes: a Standardized regression coefficients are reported.

High-level of career mentoring

1SD below mean 1SD above mean

1SD below mean 1SD above mean

High-level of career mentoring

Average-level of career mentoring

1SD below mean 1SD above mean

You might also like